Reporting of Securities Loans

AGENCY:

Securities and Exchange Commission.

ACTION:

Final rule.

SUMMARY:

The Securities and Exchange Commission (“SEC” or “Commission”) is adopting a new rule under the Securities Exchange Act of 1934 (“Exchange Act”) to increase the transparency and efficiency of the securities lending market by requiring certain persons to report information about securities loans to a registered national securities association (“RNSA”). The new rule also requires certain confidential information to be reported to an RNSA to enhance an RNSA's oversight and enforcement functions. Further, the new rule requires that an RNSA make certain information it receives, along with daily information pertaining to the aggregate transaction activity and distribution of loan rates for each reportable security, available to the public.

DATES:

Effective date: January 2, 2024.

Compliance date: The applicable compliance dates are discussed in Part VIII of this release.

FOR FURTHER INFORMATION CONTACT:

Elizabeth Sandoe, Senior Special Counsel, James Curley, Special Counsel, Elisabeth Van Derslice, Attorney-Advisor, Theresa Hajost, Special Counsel, Brendan McLeod, Attorney-Advisor, Roland Lindmayer, Attorney-Advisor, Josephine J. Tao, Assistant Director, Office of Trading Practices, or Carol McGee, Associate Director, Office of Derivatives Policy and Trading Practices, Division of Trading and Markets, Securities and Exchange Commission, 100 F Street NE, Washington, DC 20549–7010, at (202) 551–5777.

SUPPLEMENTARY INFORMATION:

The Commission is adopting 17 CFR 240.10c–1a (“final Rule 10c–1a” or “final rule”) under the Exchange Act, which requires covered persons to provide to an RNSA information concerning certain securities loans, in the format and manner required by an RNSA, and within specified time periods. The final rule requires an RNSA to make publicly available certain information it receives, within specified time periods, and to keep confidential certain information it receives. The final rule contains requirements regarding an RNSA's data retention and availability and permits an RNSA to establish and collect reasonable fees.

Table of Contents

I. Introduction

II. Background

III. Statutory Mandate

IV. Proposed Rule 10c–1

V. Overview of Final Rule

VI. Overview of Changes From Proposed Rule

VII. Discussion of the Final Rule

A. Scope of Persons With Reporting Obligations—10c–1a(j)(1)

B. Reporting Agent Overview

1. Use of a Reporting Agent—Rule 10c–1a(a)(2)

2. Reporting Agent Definition—Rule 10c–1a(j)(4)

C. Reporting Agent Requirements—Rule 10c–1a(b)

1. Reporting Agent Reporting Requirements—Rule 10c–1a(b)

2. Recordkeeping Requirements of a Reporting Agent—Rule 10c–1a(b)(5)

D. Scope of Securities Required To Be Reported—Rule 10c–1a(j)(3)

E. Scope of Transactions Required To Be Reported—Rule 10c–1a(j)(2)

F. Information To Be Provided to an RNSA

1. Loan Data Elements—Rule 10c–1a(c)

2. Loan Modification Data Elements—Rule 10c–1a(d)

3. Confidential Data Elements—Rule 10c–1a(e)

4. Removal of Securities Available to Loan Data Element

5. Removal of Securities on Loan Data Element

G. Timing of Required Reporting to an RNSA

1. Timing of Reporting of Loans

2. Timing of Reporting of Loan Modification

3. Timing of Reporting of Confidential Data Elements

H. Definition of Registered National Securities Association—10c–1a(j)(5)

I. RNSA Rules To Administer the Collection of Information—Rule 10c–1a(f)

J. RNSA Publication of Data—10c–1a(g)

K. RNSA Data Retention, Availability, Fees, and Security

1. Data Retention—Rule 10c–1a(h)(1)

2. Data Availability to the Public—Rule 10c–1a(h)(3)

3. RNSA Fees—Rule 10c–1a(i)

4. Data Security—Rule 10c–1a(h)(4)

L. Data Availability to the Commission and Other Persons—Rule 10c–1a(h)(2)

M. Cross-Border Application of Rule 10c–1a

N. Additional Comments

VIII. Compliance Date

IX. Economic Analysis

A. Introduction and Market Failure

1. Introduction

2. Market Failures

B. Economic Baseline

1. Securities Lending

2. Current State of Transparency in Securities Lending

3. Characteristics of the Securities Lending Market

4. Structure of the Securities Lending Market

5. Structure of the Market for Securities Lending Data and Analytics

6. Short Selling Transparency

C. Economic Effects of the Final Rule

1. Benefits of Increased Transparency in the Securities Lending Market

2. Regulatory Benefits

3. Direct Compliance Costs

4. Other Costs

5. Reduced Benefits From Alternative Arrangements

D. Impact on Efficiency, Competition, and Capital Formation

1. Efficiency

2. Competition

3. Capital Formation

E. Alternatives

1. Report Loan Sizes Without a Delay

2. Alternative Timeframes for Reporting or Dissemination

3. Only Require Dissemination of Aggregate or Wholesale Statistics

4. Require Reporting of Shares Available To Lend and Shares on Loan

5. Restrict Covered Persons to Broker-Dealers

6. Expand Reporting Agents Beyond Broker-Dealers and Clearing Agencies

7. Publicly Releasing the Information in 10c–1a(e)

8. Additional Information in the Reported or Disseminated Information

9. Only Allow an RNSA To Charge Fees to Data Reporters

10. Longer Holding Period Requirement

11. Longer Implementation Period

12. Report to the Commission Rather Than to an RNSA

13. Report Through an NMS Plan

X. Paperwork Reduction Act

XI. Regulatory Flexibility Act Certification

XII. Other Matters

I. Introduction

The securities lending market is opaque. There is a general lack of comprehensive information on current market conditions in the securities lending market. Although various market participants, such as certain registered investment companies (“investment companies”), are required to periodically make certain disclosures regarding their securities lending activities, parties to securities lending transactions are not currently required to report the material terms of those transactions. The lack of public information and data gaps creates inefficiencies in the securities lending market. The gaps in securities lending data render it difficult for end borrowers and lenders alike to ascertain market conditions and to know whether the terms that they receive are consistent with market conditions. These gaps also impact the ability of the Commission, RNSAs and other self-regulatory organizations (“SROs”), and other Federal financial regulators to oversee transactions that are vital to fair, orderly, and efficient markets. Indeed, the size of the U.S. securities lending market can only be estimated as the data currently available with respect to securities lending transactions are “spotty and incomplete.” Further, the FSOC 2020 Annual Report noted data gaps in certain important financial markets including transaction data for securities lending arrangements.

Private vendors have attempted to address the opacity in the securities lending market by offering systems that provide data to borrowers and lenders of securities, such as systems that are only available to those who voluntarily provide their transaction data to the data vendor. However, data gaps remain despite the private vendor attempts to address opacity in the securities lending market. The private systems are limited to voluntary submissions of data. Further, the data captured by these private vendors is not available to the general public without a subscription, and is not available in one centralized location. Only subscribers to the private vendor have access to such data, which only provides a limited view into securities lending activity. No single vendor has access to pricing information that reflects all securities lending transactions that take place. In addition, data from certain private vendors is limited to loans from a lending program to a broker or dealer and do not capture loans from a broker or dealer to an end borrower. There have also been calls from industry observers and market participants for the Commission to consider measures to provide additional transparency in the securities lending market.

II. Background

Securities lending is the market practice by which securities are transferred temporarily from one party, a securities lender, to another, a securities borrower, for a fee. A securities loan is typically a fully collateralized transaction. Securities lenders are generally large institutional investors including investment companies, central banks, sovereign wealth funds, pension funds, endowments, and insurance companies. Owners of large, static, unleveraged portfolios, mainly pension funds, increasingly cite securities lending as an important income-enhancing strategy with minimal, or at least controlled, risk. This incremental income not only helps defined-benefit pension funds to generate income, but also provides investment company investors with additional returns.

Traditionally, securities lending and borrowing transactions have been conducted on a bilateral basis. Generally, when an end investor wishes to borrow securities, it may obtain a loan from its broker or dealer from the broker's or dealer's inventory or through customer margin accounts, or the broker or dealer will borrow the securities from a lending agent with whom it has a relationship and will then re-lend the securities to its customer. Loans from lending programs to brokers or dealers occur in what is referred to by market participants as the “Wholesale market,” while loans from a broker or dealer to the end borrower occur in what is referred to by market participants as the “Customer market” (sometimes also known as the “retail market”). Obtaining a securities loan often involves an extensive search for counterparties by brokers or dealers.

Brokers and dealers are the primary borrowers of securities; they borrow for their market making activities or on behalf of their customers. Brokers and dealers who borrow securities typically re-lend those securities or use the securities to cover fails to deliver or short sales arising from proprietary or customer transactions. While the identities of the ultimate securities borrowers are usually unknown, anecdotally, hedge funds rank among the largest securities borrowers and access the lending market mainly through their prime brokers. Brokers and dealers may also lend securities that are owned by the broker or dealer, customer securities that have not been fully paid for ( i.e., have been purchased with a margin loan from the broker or dealer), and the securities of customers who have agreed to participate in a fully paid securities lending program offered by their broker or dealer.

Other securities lending transactions are often facilitated by a third party. Custodian banks have traditionally been the primary lending agent or intermediary and lend securities on behalf of their customers for a fee. Advances in technology and operational efficiency have made it easier to separate securities lending services from custody services. Such developments have given rise to specialist third party agent lenders, who have established themselves as an alternative to custodian banks. Agent lenders provide potential borrowers with the inventory of securities available for lending on a daily basis.

III. Statutory Mandate

Section 984(a) of the Dodd-Frank Wall Street Reform and Consumer Protection Act (“Dodd-Frank Act”) added section 10(c) to the Exchange Act to provide the Commission with authority over securities lending. Section 984(b) of the Dodd-Frank Act mandates that the Commission increase transparency of information available to brokers, dealers, and investors.

On November 18, 2021, to supplement the publicly available information involving securities lending, close the data gaps in this market, and minimize information asymmetries between market participants, the Commission proposed Rule 10c–1 under the Exchange Act (“proposed Rule 10c–1” or “proposed rule”). Proposed Rule 10c–1 was designed to provide investors and other market participants with access to pricing and other material information regarding securities lending transactions in a timely manner. The Commission stated that the data collected and made available by the proposed rule would improve price discovery in the securities lending market and lead to a reduction of the information asymmetry faced by end borrowers and beneficial owners in the securities lending market. In addition, the Commission stated its preliminary belief that the proposed rule would close securities lending data gaps, increase market efficiency, and lead to increased competition among providers of securities lending analytics services and reduced administrative costs for broker-dealers and lending programs. On balance, the final rule requirements are designed to achieve the objectives of the proposed rule, as discussed below, in Parts VII and IX, and are designed to increase the transparency of information available to brokers, dealers, and investors with respect to loans or borrowing securities.

IV. Proposed Rule 10c–1

The Commission proposed that any person that loans a security on behalf of itself or another person (a “Lender”) provide certain securities lending information (“Rule 10c–1 information”) to an RNSA in the time periods specified by the proposed rule ( e.g., transaction data elements and confidential data elements would be reported within 15 minutes after each loan is effected). As proposed, all securities would be within the scope of the rule.

The proposed rule stated that a bank, clearing agency, broker, or dealer that acts as an intermediary to a loan of securities (lending agent) on behalf of a person that owns the loaned securities (beneficial owner) shall provide the Rule 10c–1 information to an RNSA on behalf of the beneficial owner within the time periods specified by the proposed rule or enter into a written agreement with a broker or dealer that agrees to provide the Rule 10c–1 information to an RNSA (reporting agent) in accordance with the proposed rule's requirements. The proposed rule also provided that a beneficial owner would not be required to provide Rule 10c–1 information to an RNSA if a lending agent acts as an intermediary to the loan of securities on behalf of the beneficial owner.

The proposed rule would have permitted persons required to report (including intermediaries) to enter into a written agreement with a reporting agent that is a broker or dealer to provide the Rule 10c–1 information to an RNSA. Such a reporting agent would be required to establish, maintain, and enforce policies and procedures as well as preserve records. The reporting agent would also be required to provide an RNSA with an updated list of persons on whose behalf the reporting agent is providing information under the proposed rule.

The proposed rule would have required that an RNSA make certain reported information publicly available as soon as practicable but no later than the next business day. To track the securities lending transaction, the proposed rule would require an RNSA to assign each securities lending transaction with a unique transaction identifier. Loan modifications would be provided to an RNSA if the modifications to the loan involved any of the terms required to be reported. The terms of the loan modification, but not the parties to the loan, would be made public. In addition, proposed Rule 10c–1 required that by the end of each business day information concerning securities “on loan” and “available to loan” would be provided to an RNSA. Such information would be made publicly available by an RNSA on an aggregated basis per security.

Not all information reported to an RNSA would have been made publicly available under the proposed rule. Certain information reported to an RNSA would be necessary for regulatory functions, but would not have been made publicly available due to the likelihood that it would identify market participants or reveal investment decisions. Proposed Rule 10c–1 would have required an RNSA to keep certain information confidential, subject to the provisions of applicable law.

The proposed rule would have allowed an RNSA to charge fees to lenders, but prohibited an RNSA from charging for or limiting the use of the publicly reported data. Specifically, the proposed rule would have required an RNSA to make the published information available without use restrictions and without charge, for at least five years.

In response to the Proposing Release, the Commission received numerous comments expressing a diversity of perspectives, which are discussed in detail below. Many commenters supported enhanced transparency of information about securities loans. Other commenters did not support the rule. Certain commenters addressed the scope of the proposed rule and the timing for reporting information to an RNSA as discussed below, in Parts VII.A through VII.G.

V. Overview of Final Rule

The final rule requires covered persons to provide securities loan information concerning reportable securities to an RNSA, in the format and manner required by an RNSA, and within specified time periods (“Rule 10c–1a information”). The term “covered person” is defined to mean: (1) any person that agrees to a covered securities loan on behalf of a lender (“intermediary”); (2) any person that agrees to a covered securities loan as a lender when an intermediary is not used unless 17 CFR 240.10c–1a(j)(1)(iii) (“final Rule 10c–1a(j)(1)(iii)”) applies to a broker or dealer borrowing fully paid or excess margin securities; or (3) a broker or dealer when borrowing fully paid or excess margin securities. In addition, 17 CFR 240.10c–1a(a) (“final Rule 10c–1a(a)”) specifies that any covered person that agrees to a covered securities loan must comply with the rule.

If any person agrees to a covered securities loan on behalf of the lender (an “intermediary”), the intermediary has the obligation to provide Rule 10c–1a information to an RNSA. If an intermediary is not used, the lender is required to provide Rule 10c–1a information to an RNSA. If a covered securities loan consists of a broker or dealer borrowing fully paid or excess margin securities, only the broker or dealer is required to provide the Rule 10c–1a information to an RNSA, not the lender.

However, in a change from the proposed rule, as discussed below, in Part VII.A, the final rule excludes from the definition of the term “covered person” a clearing agency when providing only the functions of a central counterparty as defined pursuant to 17 CFR 240.17Ad–22(a)(3) (“Rule 17Ad–22(a)(2)”) of the Exchange Act or a central securities depository as defined pursuant to 17 CFR 240.17Ad–22(a)(3) (“Rule 17Ad–22(a)(3)”) of the Exchange Act. Thus, a clearing agency is not required to report Rule 10c–1a information to an RNSA for a covered securities loan when acting in the capacity or engaged in activities as a central counterparty or a central securities depository in connection with a covered securities loan.

The final rule permits a covered person to rely on a reporting agent that is a broker, dealer, or registered clearing agency to provide Rule 10c–1a information to an RNSA to fulfill such covered person's reporting obligation. To do so, the covered person must enter into a written agreement with a reporting agent, that agrees to provide Rule 10c–1a information to an RNSA, and provide such reporting agent with timely access to such information. If the reporting agent receives the Rule 10c–1a information from the covered person on a timely basis, the reporting agent assumes responsibility for compliance with the reporting requirements under the final rule.

The final rule defines a “covered securities loan” as a transaction in which any person on behalf of itself or one or more other persons, lends a “reportable security” to another person (except for a position at a clearing agency that results from certain central counterparty or central securities depository services). In addition, the use of margin securities, as defined in 17 CFR 240.15c3–3(a)(4) (“Rule 15c3–3(a)(4)”), by a broker or dealer is not a covered securities loan for purposes of the final rule unless the broker or dealer lends such margin securities to another person.

The term “reportable security” is defined as any security or class of an issuer's securities for which information is reported or required to be reported to the consolidated audit trail as required by Rule 613 of the Exchange Act and the CAT NMS Plan (“CAT”), the Financial Industry Regulatory Authority's Trade Reporting and Compliance Engine (“TRACE”), or the Municipal Securities Rulemaking Board's (“MSRB”) Real-Time Transaction Reporting System (“RTRS”), or any reporting system that replaces one of these systems.

The final rule requires a covered person, directly or indirectly through a reporting agent, to report three types of data, which together comprise the Rule 10c–1a information. The first type of data concerns the material terms of the covered securities loan and must be provided to an RNSA by the end of the day on which the covered securities loan is effected (“data elements”). The second type of data concerns modifications to a covered securities loan and must be provided to an RNSA by the end of the day on which a covered securities loan is modified, if the modification occurs after other information about the covered securities loan has already been provided to an RNSA, and results in a change to such information. The third type of data concerns confidential information in connection with a covered securities loan and must be provided to an RNSA by the end of the day on which a covered securities loan is effected. The final rule requires that an RNSA keep the third type of information confidential subject to applicable law. The final rule also requires an RNSA to make the Rule 10c–1a information available to the Commission; or other persons as the Commission may designate by order upon a demonstrated regulatory need.

An RNSA is required to make publicly available the following information not later than the morning of the business day after the covered securities loan is effected: (1) the unique identifier assigned to a covered securities loan by an RNSA and the security identifier; (2) the data elements, except for loan amount; and (3) information pertaining to the aggregate transaction activity and the distribution of rates among loans and lenders (“distribution of loan rates”) for each reportable security and related unique identifier. An RNSA is also required to make publicly available the loan amount on the twentieth business day after the covered securities loan is effected along with loan and security identifying information.

In addition, an RNSA is required to make publicly available any modification to the data elements, except for modification to the loan amount, not later than the morning of the business day after the covered securities loan is modified. An RNSA is also required to make publicly available modifications to the loan amount on the twentieth business day after the loan amount is modified along with loan and security identifying information.

The final rule requires an RNSA to implement rules regarding the format and manner of its collection of information and make publicly available such information in accordance with rules promulgated pursuant to 15 U.S.C. 78s(b) (“section 19(b)”) and 17 CFR 240.19b–4 (“Rule 19b–4”) of the Exchange Act. The final rule also contains requirements regarding an RNSA's data retention and availability. Specifically, an RNSA must maintain the information on its website or a similar means of electronic distribution, without use restrictions, for a period of at least five years. In addition, the final rule permits an RNSA to establish and collect reasonable fees pursuant to rules promulgated pursuant to section 19(b) and Rule 19b–4.

Final Rule 10c–1a is designed to provide access to timely, comprehensive securities loan information to market participants, the public, and regulators, which will help provide borrowers and lenders with better tools to assess the terms of their securities loans and enhance the ability of regulators to oversee the securities lending market. In addition, the final rule will result in the public availability of new information for investors and other market participants to consider in the mix of information about the securities lending market and the securities markets generally to better inform their decisions. The final rule will also provide regulators with information that may be used in conjunction with other information that is currently available to regulators to help assess market events.

VI. Overview of Changes From Proposed Rule

The Commission is adopting final Rule 10c–1a with certain modifications from the proposed rule made in response to comments. The final rule:

• Modifies the scope of persons required to report by:

○ Specifying the persons who have a reporting obligation with a new definition of “covered person” to distinguish persons who have a reporting obligation from those persons who do not; specifically, requiring reporting by any person that agrees to a covered securities loan on behalf of the lender, any person that agrees to a covered securities loan as a lender if an intermediary is not used, or a broker or dealer when borrowing fully paid or excess margin securities, and providing that an intermediary need not be a certain type of entity;

○ Excluding clearing agencies from the new definition of “covered person” when engaged only in certain central counterparty or central securities depository activities;

○ Separating the requirements for covered persons and reporting agents into distinct paragraphs; and

○ Adding a new definition for “reporting agent” to specify that a covered person may rely on a reporting agent that is a broker, dealer, or registered clearing agency provided there is a written agreement between the covered person and the reporting agent under which the reporting agent agrees to establish, maintain, and enforce policies and procedures to ensure compliance with the final rule and if the covered person provides timely information to the reporting agent.

• Modifies the scope of securities for which loans must be reported by adding a new definition for “reportable security” to mean any security or class of an issuer's securities for which information is reported or required to be reported to the CAT, TRACE, or RTRS, or any reporting system that replaces one of these systems; and

• Specifies the type of loan transaction to which the final rule applies by defining a new term “covered securities loan,” which excludes the use of margin securities by a broker or dealer ( e.g., rehypothecation) other than the lending of such margin securities by a broker or dealer, as well as a position at a clearing agency that results from certain central counterparty or central securities depository services.

• Streamlines information required to be reported by:

○ Removing the requirements in paragraph (e) of the proposed rule to provide securities “available to loan” and securities “on loan” information to an RNSA and the requirement for an RNSA to make such information public; and

○ Replacing the requirement to report a description of the loan modification with a requirement to report the specific modification and the specific data element being modified.

• Specifies that all data elements for covered securities loans that were not required to be reported on the date agreed to or on the date last modified, but which subsequently become covered securities loans, must be reported when the covered securities loan is modified.

• Modifies the timing of reporting by:

○ Replacing the requirements to report data elements and confidential data elements to an RNSA within 15 minutes after each loan is effected with requirements to report such elements by the end of the day on which a covered securities loan is effected; and

○ Replacing the requirement to report loan modification data elements to an RNSA within 15 minutes after each loan is modified with a requirement to report such elements by the end of the day on which a covered securities loan is modified.

• Specifies the treatment of open-ended loans in existence prior to the final rule.

• Modifies the responsibilities of an RNSA by:

○ Defining the term “RNSA” to specify that such term refers to an association of brokers and dealers that is registered as a national securities association pursuant to 15 U.S.C. 78o–3 (“section 15A”) of the Exchange Act;

○ Separating an RNSA's data publication requirements into new paragraph (g) of the final rule to more clearly delineate an RNSA's requirements from the reporting requirements applicable to covered persons and reporting agents;

○ Replacing the requirement for an RNSA to publish the loan amount (as well as any modifications to loan amount) as soon as practicable and instead: (1) make the loan amount public on the twentieth business day after the covered security loan is effected (or modified), and (2) make specified loan and security identifying information public on the twentieth business day after the covered security loan is effected (or modified);

○ Adding a new requirement to paragraph (g) of the final rule for an RNSA to publish, on a daily basis, information pertaining to the aggregate transaction activity and distribution of loan rates for each reportable security in order to provide market participants with timely access to loan rate information that incorporates information about loan sizes; and

○ Removing the requirement in paragraph (g) of the proposed rule that collected information be made available to the public “without charge,” and removing the requirement in paragraph (h) of the proposed rule that fees only be paid from persons who provide Rule 10c–1a information directly to an RNSA.

In addition, the final rule includes some technical modifications, such as modifying the rule designation to conform with Federal Register requirements as well as additional technical or conforming modifications that are minor and not substantive. Responses to comments requesting clarification about the cross-border scope of the rule are also provided below, in Part VII.M. Further, the Commission is providing compliance dates requiring that: (1) RNSAs propose rules pursuant to final Rule 10c–1a(f) within four months of the effective date of final Rule 10c–1a; (2) the proposed RNSA rules are effective no later than 12 months after the effective date of final Rule 10c–1a; (3) covered persons report Rule 10c–1a information to an RNSA starting on the first business day 24 months after the effective date of final Rule 10c–1a (the “reporting date”); and (4) RNSAs publicly report Rule 10c–1a information pursuant to final Rules 10c–1a(g) and (h)(3) within 90 calendar days of the reporting date.

VII. Discussion of the Final Rule

A. Scope of Persons With Reporting Obligations—10c–1a(j)(1)

Proposed Rule

The proposed rule would have required that any person that loans a security on behalf of itself or another person shall provide Rule 10c–1 information to an RNSA. The Proposing Release stated that the term “person,” for purposes of the Exchange Act, means a natural person, company, government, or political subdivision, agency, or instrumentality of a government.

The proposed rule also stated that “a bank, clearing agency, broker, or dealer, that acts as an intermediary to a loan of securities (lending agent) on behalf of a person that owns the loaned securities (beneficial owner) shall . . . provide the Rule 10c–1 information to an RNSA . . . .” However, if a person loaned a security on behalf of itself, and an intermediary did not act on its behalf, such person would be the one required to provide the Rule 10c–1 information to an RNSA.

The Commission proposed that when “a bank, clearing agency, broker, or dealer” acts as an intermediary (or “lending agent”) on behalf of a person that owns loaned securities the lending agent would have the reporting obligation. When discussing the rationale for this requirement, the Commission stated that lending agents are in the best position to know when securities have been loaned from the portfolios that the lending agent represents, and that the owner may not know that the lending agent has lent securities from their portfolio until after the time prescribed by proposed Rule 10c–1 to provide Rule 10c–1 information to an RNSA. The Proposing Release also stated that custodian banks have traditionally been the primary lending agent or intermediary and lend securities on behalf of their customers.

The Commission proposed a single-sided approach of only applying the rule's reporting requirements to securities lenders and intermediaries acting on behalf of lenders, and not to borrowers. The Proposing Release explained that such an approach could avoid the potential double counting of transactions. It also explained that lenders are more likely to have access to the Rule 10c–1 information, but a borrower may not be privy to all of the information required to be provided to an RNSA under the proposed rule. The Commission also stated that to the extent smaller entities engage in securities lending, they generally employ lending agent intermediaries, which would relieve them from having to provide proposed Rule 10c–1 information to an RNSA. Accordingly, the Commission stated its preliminary belief that requiring only securities lenders or intermediaries to a loan of securities to provide the proposed Rule 10c–1 information will alleviate the potential for the double counting of transactions and limit the burdens of proposed Rule 10c–1 to larger institutions.

Final Rule

Many commenters requested additional clarity regarding the market participants that would be required to provide the proposed Rule 10c–1 information to an RNSA. As discussed below in this part, to address commenter concerns, and to specify which persons are required to provide Rule 10c–1a information to an RNSA, the final rule defines the term “covered person” to mean: (1) any person that agrees to a covered securities loan on behalf of a lender (“intermediary”) other than a clearing agency when providing only the functions of a central counterparty or central securities depository; (2) any person that agrees to a covered securities loan as a lender when an intermediary is not used, unless the borrower is a broker or dealer borrowing fully paid or excess margin securities; or (3) a broker or dealer when borrowing fully paid or excess margin securities. Accordingly, if a person uses an intermediary, such as a lending agent that is a custodian bank, to run its lending program to lend securities to other persons on its behalf, the custodian bank is the covered person for purposes of the final rule. If, however, a person does not use an intermediary to lend securities on their behalf, such person is the covered person for purposes of the final rule. For example, a fund is a covered person for purposes of the final rule if that fund runs its own lending program and lends securities to other persons without the use of an intermediary, such as custodian bank. Further, a broker or dealer who borrows fully paid or excess margin securities from its customer is the covered person for purposes of the final rule, and the customer of the broker or dealer is not a covered person.

The definition of “covered person” uses the term “agrees to” instead of “loans a security” as proposed to make clear that a customer of a broker or dealer that loans a reportable security to the broker or dealer in a fully paid lending arrangement is not required to comply with the final rule. Instead, the broker or dealer that borrows the reportable security is required to comply with the final rule and must report Rule 10c–1a information on the covered securities loan. Further, as discussed below, in Part VII.E, the term “agrees to a covered securities loan” also provides that Rule 10c–1a information is not limited to covered securities loans that have been settled.

One commenter recommended that the Commission not “limit permitted `lending agents' for purposes of the rule to banks, clearing agencies, brokers, or dealers, as proposed,” but instead “permit any lending agent that acts as an intermediary to a loan of securities to report on behalf of a beneficial owner.” The commenter stated that as long as a lending agent could meet the requirements of the proposed rule, such an expansion would “reflect the variety of lending agents that funds use” and “would facilitate reporting by beneficial owners that prefer to use non-broker-dealer lending agents to report their securities loans.” The Commission agrees, and has modified the final rule to define “covered person” to encompass “[a]ny person that agrees to a covered securities loan on behalf of the lender.”

One commenter sought clarification of whether a “branch or the entity” would be responsible for proposed Rule 10c–1 information reporting if the branch enters into a covered securities loan, although the commenter did not define the terms “branch” or “entity.” Determining the “person” required to report in complex organizational structures will depend on the facts and circumstances. A branch or office of a covered person, as opposed to a subsidiary or affiliate, may not constitute a distinct covered person, but rather different parts of the same covered person. The designation of a business unit as a branch, office, or otherwise, is not dispositive of whether such business unit has the obligation to report Rule 10c–1a information. One commenter sought confirmation that a clearing agency's provision of depository or central clearing services, including novation, processing, settlement, netting, and incidental services, do not make the clearing agency an intermediary to a loan of securities. The Commission agrees that a clearing agency's provision of depository or central clearing services to securities lending transactions is not the type of activity that final Rule 10c–1a is designed to cover. By acting as a central counterparty, including by novating a loan, a clearing agency steps into the shoes of the parties to the loan and technically may become a person that agrees to a loan on its own behalf or on behalf of another person. However, requiring reporting by a registered clearing agency providing such central counterparty or securities depository services would not provide additional transparency and would be duplicative of the reporting requirements that apply to the loan's lender or intermediary prior to it being cleared. Accordingly, the final rule specifically excludes from the “covered person” definition a clearing agency when providing only the functions of a central counterparty pursuant to 17 CFR 240.17Ad–22(a)(2) (“Rule 17Ad–22(a)(2)”) or a central securities depository pursuant to 17 CFR 240.17Ad–22(a)(3) (“Rule 17Ad–22(a)(3)”). The exclusion would not apply when the clearing agency is acting in a different capacity ( e.g., as an intermediary providing the services of a lending agent).

Certain commenters supported the proposed single-sided reporting structure ( i.e., requiring reporting by any person that loans a security on behalf of itself or another person, but not requiring reporting by the borrower) to alleviate concerns about double counting loans of securities. One commenter stated that “we strongly support the Proposal's approach of requiring single-sided reporting . . . the approach would reduce the potential for double counting of securities lending transactions and limit the burden on lenders. A single-sided reporting regime avoids problems with reconciling reports by each party to a transaction.” Other commenters supported the application of the final rule's reporting requirements to all lenders to facilitate a comprehensive view of the securities lending market. In consideration of these comments, the final rule generally employs a single-sided approach to reporting, with the Rule 10c–1a information reporting obligations placed on either an intermediary that agrees to a covered securities loan on behalf of the lender, or any person that agrees to a covered securities loan as a lender when an intermediary is not used. Consistent with the proposed rule, this approach is designed to avoid the potential for the double counting of transactions that could arise if both sides, borrowers and lenders, were required to report final Rule 10c–1a information to an RNSA.

However, one commenter opposed the proposed single-sided reporting approach, and recommended that the final rule require reporting by SEC-registered brokers or dealers only and for transactions in which they act as borrowers, lenders, or lending agents. The commenter stated that “whether acting as borrower or lender acting in a principal or agency capacity, [requiring SEC-registered broker-dealers to report] would substantially reduce overall implementation costs, would not impose costs on a single side of the market and would still provide for sufficient market data.” The commenter reasoned that reporting costs would be lower, due to brokers or dealers having existing reporting infrastructure and connectivity. The commenter cited to the Office of the Financial Research Pilot Survey (“OFR Pilot Survey”) to support the view that although such a structure would be “less comprehensive than lender reporting, the data loss entailed should not be substantial” and “would be substantially captured” and “sufficient.” The commenter also stated that its proposed structure “would also provide better data integrity as broker-dealers are better positioned than beneficial owners and Lending Agents to evaluate whether a securities loan is made for the purpose of facilitating short sales or settlement of fails rather than some other purpose.”

The Commission is not adopting the commenter's recommended reporting structure. Requiring reporting by brokers or dealers acting as borrowers, lenders, or lending agents introduces the same double counting problem that the final rule's single-sided reporting approach is designed to address. Additionally, under the final rule brokers or dealers are permitted to act as reporting agents provided specified requirements are met. The Commission also does not agree with the commenter that the data lost if its recommended reporting structure were adopted would not be substantial. The OFR Pilot Survey cited by the commenter states that banks, credit unions, pension funds, and hedge funds, all of whom would not be required to report covered securities loans under the commenter's proposed structure, are the borrowers of nearly 15 percent of the value of all outstanding securities loans. Based on the survey, the Commission estimated that brokers or dealers facilitate between 60 percent and 90 percent of transactions in the equity lending market. Accordingly, while the Commission acknowledges that the precise percentage of broker or dealer facilitated transactions is unknown, the survey indicates that a significant percentage of transactions in the equity lending market, between 10 percent and 40 percent, are facilitated by non-brokers or dealers. Therefore, the commenter's recommended approach would exclude a significant percentage of securities lending transactions from being reported. Any improved “data integrity” achieved by the commenter's proposed structure would be undermined by its exclusion of securities loans that do not involve an SEC-registered broker or dealer acting as borrower, lender, or lending agent. Additionally, excluding certain types of entities from reporting would create incentives for market participants to agree to covered securities loans through such excluded entities, which would further diminish the comprehensiveness of the reported information. Instead, the Commission agrees with one commenter's statement that, “all lenders (or their lending or reporting agents) should be required to report, whether or not they are registered with the Commission. Otherwise, the data will be incomplete.” As such, the final rule's definition of “covered person” is not limited to brokers or dealers.

One possible alternative to address double counting under the commenter's suggestion of requiring only brokers or dealers to report as lenders and borrowers would be to require identification (such as using a flag) of loans reported by borrowers and loans reported by lenders. However, this would not address the lack of securities loan data from persons that are not brokers or dealers, or the incentives to avoid reporting by migrating to such persons. Further, expanding the reporting obligation generally to all lenders and borrowers, and requiring each to identify whether they are reporting as a lender or borrower, would increase burdens unnecessarily, resulting in the collection of duplicative (albeit identified) data by more persons.

One commenter suggested that “it would be appropriate for the borrower broker-dealer (not the lender customer) to provide the Rule 10c–1 information to an RNSA where such broker-dealer borrows a customer's fully paid securities.” The Commission agrees with this recommendation. Such an approach would help avoid placing a reporting obligation on customers who, upon lending fully paid or excess margin securities, may not have timely access to the required information and less familiarity with reporting information to RNSAs than their broker or dealer. Therefore, the definition of the term “covered person” under the final rule generally does not apply to a borrower, except in the instance of a broker or dealer borrowing fully paid or excess margin securities from a customer. The definition specifically includes a broker or dealer when borrowing fully paid or excess margin securities pursuant to 17 CFR 240.15c3–3(b)(3) (“Rule 15c3–3(b)(3)”). This definition will result in a broker or dealer (as opposed to the customer) being responsible for reporting Rule 10c–1a information to an RNSA if it borrows fully paid or excess margin shares from one of its customers. Pursuant to final Rule 10c–1a, the broker or dealer would also be responsible for reporting Rule 10c–1a information to an RNSA should the broker or dealer act as a lender of those shares to a third-party.

B. Reporting Agent Overview

1. Use of a Reporting Agent—Rule 10c–1a(a)(2)

Proposed Rule

The Commission proposed that “[a] person required to provide Rule 10c–1 information . . . including a lending agent, may enter into a written agreement with a broker or dealer that agrees to provide the Rule 10c–1 information to an RNSA (reporting agent).” The proposed rule stated that a “reporting agent is required to provide the Rule 10c–1 information to an RNSA if it has entered into [the required] written agreement . . . and is provided timely access to the Rule 10c–1 information.” The proposed rule also stated that “[a]ny person that enters into [the required] written agreement . . . with a reporting agent is not required to provide the Rule 10c–1 information to an RNSA if the reporting agent is provided timely access to the Rule 10c–1 information.” Therefore, under the proposed rule, only when a person fulfilled the requirements of: (i) entering into a written agreement with a broker or dealer that agrees to provide the Rule 10c–1 information; and (ii) providing the reporting agent with timely access to the Rule 10c–1 information, would the person shift its Rule 10c–1 information reporting obligation for a loan of securities from itself to a reporting agent.

The Commission in the Proposing Release stated its preliminary belief that it is appropriate that lenders, including a lending agent, be able to enter into a written agreement with a broker or dealer acting as a reporting agent to provide Rule 10c–1 information to an RNSA on behalf of the lender because such an arrangement would ease burdens on lenders, including lending agents, that do not have and do not want to establish connectivity to an RNSA. The Commission also stated that the use of reporting agents could reduce the costs for non-RNSA-members, because rather than incurring the costs associated with directly reporting Rule 10c–1 information, such persons would have the option to use a third-party to provide the Rule 10c–1 information to an RNSA.

Final Rule

The Commission received broad support from commenters for the proposed rule's provisions permitting covered persons to use a reporting agent if certain conditions are met. One commenter stated that it “agree[s] with the Commission that permitting the use of reporting agents to report 10c–1 information will `ease burdens on Lenders, including lending agents, that do not have or do not want to establish connectivity to an RNSA.' ” The Commission continues to believe it is appropriate to permit a covered person to use a reporting agent to help fulfill Rule 10c–1a information reporting obligations. Furthermore, many entities that may act as reporting agents may have existing RNSA connectivity, experience with reporting information to an RNSA, and economies of scale that could benefit covered persons with final Rule 10c–1a reporting obligations.

Thus, consistent with the proposed rule, the final rule requires that, in order to use a reporting agent to fulfill its Rule 10c–1a information reporting obligations, a covered person must: (1) enter into a written agreement with the reporting agent, and (2) provide the reporting agent with timely access to the required Rule 10c–1a information. Requiring a written agreement between the covered person and the reporting agent will memorialize the contractual obligations for the reporting agent to provide the Rule 10c–1a information to an RNSA. Consistent with the Proposing Release, for purposes of final Rule 10c–1a, “timely access” means that the reporting agent has access to the Rule 10c–1a information with sufficient time to provide such information to an RNSA ( i.e., in the format and manner required by the rules of an RNSA and within the time periods specified in paragraphs (c) through (e) of the final rule). However, this definition of “timely access” differs from the Proposing Release in that the timing of the required reporting of certain securities loan information to an RNSA has changed ( i.e., to no longer require reporting “within the fifteen minutes after the securities loan is effected or the terms of the loan are modified”). As discussed below, in Part VII.G, the required timing for the reporting of data elements pursuant to final Rules 10c–1a(c), (d), and (e) is now by the end of the day on which a covered securities loan is effected or modified.

If the reporting agent is unable to provide Rule 10c–1a information to an RNSA because it lacks timely access to it, the covered person who enters into the written agreement with the reporting agent is responsible for providing such information to an RNSA. Ultimately, responsibility for non-compliance will be a facts and circumstances determination. For instance, if the covered person fails to provide the reporting agent with access to accurate Rule 10c–1a information, then the covered person remains responsible for compliance with Rule 10c–1a. However, if the reporting agent that receives timely and accurate information from the covered person provides late or inaccurate information to an RNSA, as discussed below in Part VII.C.1, the reporting agent is responsible for compliance with final Rule 10c–1a.

2. Reporting Agent Definition—Rule 10c–1a(j)(4)

Proposed Rule

The Commission proposed that, “[a] person required to provide Rule 10c–1 information . . . may enter into a written agreement with a broker or dealer that agrees to provide the Rule 10c–1 information to an RNSA (reporting agent) within the time periods specified in Rule 10c–1.” The Proposing Release stated that limiting who can act as a reporting agent to a broker or dealer that is regulated directly by the Commission, would aid the Commission in overseeing compliance with proposed Rule 10c–1 and provide RNSAs with the ability to oversee the activity of its members that perform a reporting agent function. The Proposing Release also stated that if reporting agents were to include other, non-broker or dealer entities, the Commission might lack an efficient way to oversee how the entity is complying with its responsibility to provide Rule 10c–1 information to an RNSA.

Final Rule

The Commission received numerous comments discussing what types of entities should be permitted to act as a reporting agent, including differing views on the proposed rule's restriction to only brokers or dealers. The Commission also received comment stating that there was a lack of clarity in the Proposing Release around the term “reporting agent.”

The Commission received a comment supporting the requirement that reporting agents be registered brokers or dealers and stating that it “would allow an RNSA to oversee the activity of reporting agents and enable it to fulfill an essential regulatory function that promotes just and equitable principles of trade.” Another commenter expressed support for permitting brokers or dealers to act as reporting agents. However, other commenters recommended that entities other than brokers or dealers should be eligible to act as reporting agents.

One commenter recommended that registered clearing agencies should be eligible to serve as a reporting agent because such agencies are subject to regulation and examination by the Commission and have experience providing the infrastructure that would be used by, and function as, a reporting agent. Another commenter suggested that reporting agents should include clearing agencies that operate stock lending platforms. Having considered the commenters' recommendations, the Commission agrees that registered clearing agencies should be permitted to act as reporting agents on behalf of covered persons, and therefore be included under the final rule's definition of the term “reporting agent.” Allowing registered clearing agencies to act as reporting agents will facilitate cost-effective and efficient reporting. Additionally, like brokers and dealers, registered clearing agencies are subject to the Commission's direct oversight and examination functions.

Some commenters stated that allowing non-brokers or dealers to act as reporting agents could facilitate low-cost service providers. Another commenter stated that “by opening up the reporting agent role to non-broker entities, the Proposal could promote more competition amongst intermediaries—including lending agents—in the securities lending marketplace.” One commenter stated that limiting the reporting agents to only brokers or dealers would unfairly restrict the market for such services. Other commenters specifically stated that certain non-brokers or dealers have existing technological capability to be able to act as a reporting agent ( e.g., data vendors and entities that report information for other regulatory purposes). One such commenter supported leveraging the existing technology of non-broker-dealers to reduce implementation time and staff training costs relating to the implementation of proposed Rule 10c–1.

Allowing clearing agencies, as well as brokers or dealers, to act as reporting agents should help facilitate low-cost service providers, introduce more competition, and not unduly restrict the market for reporting agent services to only brokers or dealers. Expanding the definition of reporting agent to include registered clearing agencies strikes a balance between increasing participation and competition in the marketplace for such services, while only applying the definition to entities over which the Commission has direct oversight. Limiting who can act as a reporting agent to brokers, dealers, and registered clearing agencies, all of which are regulated directly by the Commission, will assist the Commission in overseeing compliance with final Rule 10c–1a. Including other entities would leave the Commission without an efficient way to oversee how the entity is complying with its responsibility to provide Rule 10c–1a information to an RNSA on behalf of a covered person. Two commenters stated that some non-brokers or dealers should be eligible to serve as reporting agents because they are already permitted to report information in connection with other Commission reporting regimes and have experience doing so. The commenters identified the CAT, TRACE, and FINRA's Large Options Positions Report (“LOPR”) as examples of reporting regimes that allow brokers or dealers to use non-brokers or dealers as reporting agents. In each of the reporting regimes identified by the commenters, the reporting requirements apply to entities that are registered with the Commission or are a member of an RNSA or an exchange, and that retain legal liability for compliance notwithstanding the existence of an agreement for a third party to provide connectivity services on behalf of the registrant.

The Commission is also clarifying that the ability to use a reporting agent does not prevent covered persons from contracting privately with third-party vendors to assist in reporting. The proposed rule required that “[a]ny person that loans a security on behalf of itself . . . shall provide to a [RNSA] the [Rule 10c–1 information].” However, the proposed rule did not address the use of a third-party vendor by a lender to help facilitate its reporting obligations. The rule as adopted does not prohibit the use of third-party vendors by covered persons. The use of third-party vendors by covered persons to help facilitate the reporting of Rule 10c–1a information should allow covered persons flexibility and decrease costs, and help address a commenter's concerns with the ability of certain covered persons to report. The difference between a covered person relying on a reporting agent to fulfill its Rule 10c–1a reporting requirements and using a third-party vendor to help facilitate its Rule 10c–1a reporting is solely with respect to liability and responsibility under final Rule 10c–1a. When a covered person uses a reporting agent and meets the conditions of the rule, the covered person may rely on a reporting agent to fulfill its reporting obligations. However, the use of other third-party vendors that are not reporting agents would not relieve a covered person of its obligation to report Rule 10c–1a information to an RNSA, as reliance on a reporting agent would.

Some commenters stated that providing the proposed Rule 10c–1 information to a reporting agent that is a broker or dealer could reveal confidential information to such broker or dealer relating to covered securities loans. One such commenter stated that “[w]ithout an alternative, beneficial owners would either be forced to build their own direct reporting to RNSAs to mitigate these concerns, thereby significantly increasing their costs, or to exit the market entirely, thereby reducing overall liquidity.” Another commenter mentioned operational and confidentiality considerations involved with appointing a reporting agent.

Brokers, dealers, and clearing agencies, like all persons, are subject to prohibitions on misuse of material non-public information. Brokers, dealers, and clearing agencies are subject to Commission examination for compliance with this, and other, requirements. To the extent that a covered person may, nonetheless, be concerned about providing sensitive information to another person, it may elect to hire third party vendors for specific tasks to help with compliance obligations. As noted above, the covered person would retain legal responsibility for reporting.

One commenter appeared to believe that all reporting required by the proposed rule must be done through members of FINRA (the only currently existing RNSA). The final rule does not require that all covered persons that are required to report Rule 10c–1a information to an RNSA be members of that RNSA. The final rule requires only that reporting agents, if relied upon by a covered person to fulfill its reporting obligations, must be a broker, dealer, or registered clearing agency. Covered persons, including persons that are not RNSA members, that elect not to use a reporting agent, are responsible for providing the Rule 10c–1a information to an RNSA directly and may do so without becoming RNSA members.

Having considered the commenters' recommendations discussed above, in this part, and for the foregoing reasons, the final rule defines the term “reporting agent” to mean “a broker, dealer, or registered clearing agency that enters into a written agreement with a covered person under paragraph (a)(2) of [the final rule].” Therefore, the final rule adds “registered clearing agencies” to the proposed rule's scope of entities that are permitted to act as reporting agents, which was limited to brokers or dealers.

C. Reporting Agent Requirements—Rule 10c–1a(b)

1. Reporting Agent Reporting Requirements—Rule 10c–1a(b)

Proposed Rule

As discussed above, in Part VII.B, paragraph (a)(1)(ii)(B) of the proposed rule required the reporting agent to provide the Rule 10c–1 information to an RNSA if the reporting agent had entered into a written agreement (with the covered person) to provide the Rule 10c–1 information to an RNSA pursuant to paragraph (a)(1)(ii)(A) of the proposed rule and the reporting agent had been provided timely access to such Rule 10c–1 information by the covered person. The proposed rule, in paragraphs (a)(2)(i) through (a)(2)(iv), also included requirements for the reporting agent to assist both an RNSA and the Commission with surveillance and compliance with RNSA requirements and proposed Rule 10c–1. In proposing the specific requirements for reporting agents, the Commission preliminarily believed it was appropriate for a reporting agent to be responsible for providing Rule 10c–1 information to an RNSA, provided the reporting agent has contractually agreed to provide such information to an RNSA and the reporting agent has also been provided with timely access to such information.

In addition, the proposed rule's requirement that the reporting agent enter into a written agreement with an RNSA was intended to evidence the explicit permission the reporting agent has to provide Rule 10c–1 information on behalf of the covered person. Similarly, the reporting agent would also have been required to provide an RNSA with a list of each covered person on whose behalf the reporting agent would be providing the Rule 10c–1 information (and also to update the list by the end of the day when the list changes). As the Commission explained in the Proposing Release, requiring the reporting agent to provide the identities of each covered person on whose behalf the reporting agent is providing Rule 10c–1 information to an RNSA is intended to provide the Commission with the ability to obtain the identities of the covered persons from an RNSA in order to aid the Commission with its oversight of the covered persons that have entered into agreements with reporting agents, including with their compliance with the proposed rule. In addition, the proposed requirement for a reporting agent to have written policies and procedures was intended to provide regulators with a means to examine and enforce a reporting agent's compliance with proposed Rule 10c–1. Moreover, the proposed recordkeeping requirements were intended to help facilitate the Commission's oversight of reporting agents and to review the reporting agents' compliance with the requirements to provide the Rule 10c–1 information to an RNSA.

Final Rule

The Commission sought specific comment regarding the role and requirements of a reporting agent. The Commission received a number of reporting agent-related comments in response; however, most of these comments focused primarily on a covered person's use of a reporting agent or the eligibility requirements for a reporting agent, as discussed above, in Part VII.B.2, rather than the requirements of an already eligible reporting agent, as relevant here. The Commission is adopting the specific reporting agent requirements, substantially as proposed, but with some non-substantive modifications.

First, the requirements are relocated into a separate paragraph (b) under the final rule. This non-substantive, technical modification is intended to streamline and simplify the proposed rule text, by combining into one paragraph, the requirements specific to a reporting agent (as distinguished from the requirements applicable to the covered person) in order to help clarify which party is ultimately responsible for providing the Rule 10c–1a information to an RNSA.

Second, to reduce redundancy and complexity in the proposed rule text, paragraph (b) of the final rule requires that “any reporting agent” that assumes the reporting obligation on behalf of the covered person, pursuant to paragraph (a)(2) of the final rule, comply with the specific requirements in paragraph (b) of the final rule. A covered person that enters into a written agreement with a reporting agent to provide the Rule 10c–1a information to an RNSA on behalf of the covered person as required by the final rule, and that provides the reporting agent with timely access to the Rule 10c–1a information, may rely on the reporting agent to fulfill the covered person's reporting obligations under paragraph (a)(1) of the final rule.

Lastly, the final rule combines the reporting-related requirements that were previously located in paragraphs (a)(1)(ii)(B) and (a)(2) of the proposed rule with the proposed reporting agent recordkeeping requirements in paragraphs (a)(1)(ii)(A) and (a)(1)(ii)(B), into paragraph (b) in the final rule.

Paragraph (b) of the final rule's specific inclusion of the term “reporting agent,” which is defined in final Rule 10c–1a(j)(4) to mean a broker, dealer, or registered clearing agency that enters into a written agreement with a covered person under paragraph (a)(2) of the final rule, as well as inclusion of the “assumes the reporting obligation” language, sets forth the reporting agent's ultimate responsibility for reporting Rule 10c–1a information to an RNSA. The final rule provides that a reporting agent will be required to provide the Rule 10c–1a information to an RNSA under the new reporting regime only when the covered person has both: (1) entered into the required written agreement with the reporting agent, and (2) provided the reporting agent with timely access to the Rule 10c–1a information. Thus, if a reporting agent is unable to provide the Rule 10c–1a information to an RNSA because it lacks timely access to it, the covered person who enters into the written agreement with the reporting agent under paragraph (a)(2)(i) of the final rule remains responsible for providing the required Rule 10c–1a information to an RNSA, consistent with the proposed rule.

Additionally, the Commission is adopting: 17 CFR 240.10c–1a(b)(1) (“final Rule 10c–1a(b)(1)”); 17 CFR 240.10c–1a(b)(2) (“final Rule 10c–1a(b)(2)”); and 17 CFR 240.10c–1a(b)(3) (“final Rule 10c–1a(b)(3)”) as new paragraphs (b)(1) through (b)(3) of the final rule, the specific reporting-related requirements of a reporting agent, substantially as they were proposed in paragraphs (a)(2)(i) through (a)(2)(iii) of the proposed rule. Paragraph (b)(2) of final Rule 10c–1a will require a reporting agent to establish, maintain, and enforce written policies and procedures that are reasonably designed to provide Rule 10c–1a information to an RNSA. However, in the final rule, as modified with non-substantive, technical, or clarifying changes to the rule text consistent with the other changes made to the overall final rule text, such Rule 10c–1a information is provided (by the reporting agent) to an RNSA on behalf of a “covered person” (which is a newly defined term in the final rule, rather than the term “another person,” as was used in the proposed rule); and, also consistent with the proposed rule, such information is to be provided “in the format and manner required by the applicable rule(s) of an RNSA, and within the time periods specified in paragraphs (c) through (e)” (which is designed to clarify the proposed rule text, “in the manner, format, and time consistent with Rule 10c–1” ). Apart from the aforementioned changes, the policies and procedures requirement of the reporting agent remains unchanged from the proposed rule.

Additionally, consistent with the proposed rule, paragraph (b)(3) of the final rule requires the reporting agent to enter into a written agreement with an RNSA that permits the reporting agent to provide the required Rule 10c–1a information to an RNSA ( e.g., establish the required connectivity with such RNSA) on behalf of a covered person. More specifically, the final rule provides that a reporting agent shall “[e]nter into a written agreement with an RNSA that permits the reporting agent to provide Rule 10c–1a information to an RNSA on behalf of a covered person.” Other than replacing the proposed term “another person” with the newly defined term “covered person” in the final rule, the written agreement requirement of the reporting agent remains unchanged from the proposed rule.

Moreover, the Commission is adopting 17 CFR 240.10c–1a(b)(4) (“final Rule 10c–1a(b)(4)”), which requires the reporting agent to “provide an RNSA with a list naming each covered person on whose behalf the reporting agent is providing Rule 10c–1a information to an RNSA,” consistent with the proposed rule, although paragraph (b)(4) of final Rule 10c–1a adds the term “naming” to replace the less-descriptive term “of,” as was used in the proposed rule. For consistency, the final rule also includes the newly defined term “covered person” in place of “person and lending agent,” as was used in paragraph (a)(2)(iii) of the proposed rule. Moreover, paragraph (b)(4) of the final rule clarifies the proposed rule by streamlining the text, “provide an RNSA with any updates to the list of such persons by the end of the day such list changes,” by removing the extra “on the day” in the sentence. Also, the final rule slightly modifies the rule text “an updated list” to read instead “any updates to the list” to clarify that any updates (or changes) to the list (and the updated list itself) must be provided to an RNSA “by the end of the day such list changes.” A reporting agent's written policies and procedures will need to address the above requirements.

2. Recordkeeping Requirements of a Reporting Agent—Rule 10c–1a(b)(5)

Proposed Rule

Paragraph (a)(2)(iv) of the proposed rule would have required that the reporting agent maintain certain information for a period of three years, the first two in an easily accessible place. The information required to be maintained included the Rule 10c–1 information provided by the covered person to the reporting agent, including the time of receipt, as well as the Rule 10c–1 information that the reporting agent provided to an RNSA, and time of transmission. Additionally, under the proposed rule, the reporting agent would have to retain the written agreements between the reporting agents and beneficial owners, lending agents, and an RNSA. As the Commission explained, the proposed recordkeeping requirements were designed to help facilitate the Commission's oversight of reporting agents and review the reporting agents' compliance with the requirement to provide the Rule 10c–1 information to an RNSA.

Final Rule

One commenter stated that it was redundant and unnecessary for the reporting agent to retain data, as FINRA already has the data. However, pursuant to 17 CFR 240.10c–1a(b)(5) (“final Rule 10c–1a(b)(5)”), the reporting agent would not only be required to retain data that it submitted to an RNSA (including time of transmission), but would also be required to retain the data it received from the covered person (including time of receipt), which would demonstrate compliance with the timeliness requirements of the rule or whether the reporting agent itself received the data in an untimely manner from the covered person, as well as the written agreements required in order to be eligible as a reporting agent. Such records would be helpful to regulators in identifying and reconstructing the reasons for missing, incomplete, or untimely data ( e.g., whether it was the result of untimely receipt by the reporting agent or issues with the reporting agent's operations). The data would also help reporting agents respond to Commission inquiries as to whether the reporting agent or the covered person appropriately complied with the final rule. The records of data submitted to an RNSA would provide context for such inquiries, as well as allow for a comparison of data received (by the reporting agent) and submitted to an RNSA, to the data published by an RNSA, allowing regulators to determine whether there are issues with an RNSA's processing of data received.

Accordingly, the Commission is adopting, in paragraph (b)(5) of final Rule 10c–1a without substantive changes, the proposed recordkeeping requirements contained in paragraphs (a)(2)(iv)(A) and (B) of the proposed rule that provide for the reporting agent to preserve for a period of not less than three years, the first two years in an easily accessible place, both the Rule 10c–1a information obtained by the reporting agent from the covered person pursuant to paragraph (a)(2) of the final rule, including the time of receipt, and the corresponding Rule 10c–1a information provided by the reporting agent to an RNSA, including the time of transmission to an RNSA ; and, the written agreements as required under paragraphs (a)(2) and (b)(3) of the final rule.

D. Scope of Securities Required To Be Reported—Rule 10c–1a(j)(3)

Proposed Rule

In the Proposing Release, the Commission stated its preliminary belief that proposed Rule 10c–1 should apply to all securities “to ensure that a complete picture of transactions involving the loan of securities is provided to the RNSA.” Accordingly, the proposed rule would have required the reporting of loans of any security ( e.g., both equity and debt).

Specifically, the Commission stated that, “if the Commission were to limit the scope of the proposed Rule ( e.g., to only equity securities) then a significant number of securities lending transactions would be excluded and the market efficiencies and reduction of information asymmetry that the Commission anticipates will result from proposed Rule 10c–1 would not accrue to non-equity securities.”

Final Rule

In response to proposed Rule 10c–1, some commenters expressed general support for increased transparency and price discovery in the securities lending market by increasing the amount and availability of data in the market. Other commenters offered support for the inclusion of all loans of securities. Another commenter expressed the view that “[t]here is an urgent need to require securities lenders to provide greater details of their loans to a RNSA as outlined by the proposed changes to Rule 10c–1 . . . [m]arket participants and regulators alike will greatly benefit from the greater transparency that comes from reporting every securities lending transaction as a result of the proposed changes to Rule 10c–1, just as they have from the trade-by-trade reporting that was instituted in the corporate bond market since TRACE was introduced.”

One commenter recommended that the Commission make explicit “the types of securities loans to which Rule 10c–1 applies so that market participants have clarity as to which types of securities loans must be reported.” The final rule makes explicit the types of transactions to which the final rule applies and the securities that are within the scope of the final rule through two newly defined terms: “covered securities loan” and “reportable security.” The definition of the term “covered securities loan” in paragraph (j)(2) of the final rule makes explicit the types of loans to which the final rule applies, as discussed below, in Part VII.E. The definition of the term “covered securities loan” refers to a transaction in which any person on behalf of itself or one or more other persons, lends a “reportable security” to another person. The final rule defines the term “reportable security” as “any security or class of an issuer's securities for which information is reported or required to be reported to the consolidated audit trail as required by § 242.613 (“Rule 613”) of the Exchange Act and the CAT NMS Plan (“CAT”), the Financial Industry Regulatory Authority's Trade Reporting and Compliance Engine (“TRACE”), or the Municipal Securities Rulemaking Board's Real Time Reporting System (“RTRS”), or any reporting system that replaces one of these systems.” The definition of the term “reportable security” aligns the securities for which loans must be reported with securities for which transactions are currently being reported to existing reporting regimes, specifically, the CAT, TRACE, and RTRS. At this time, data on securities loans should be consistent with existing reporting regimes for other transactions in the same securities.

Aligning the scope of securities for covered securities loans with existing transaction reporting systems will provide a number of benefits. For instance, such alignment will allow regulators to obtain a more complete view across the different types of transactions for the same securities. In addition, many market participants are already familiar with the existing systems, either as reporting entities or, in the case of TRACE and RTRS, as users of the data.

Presently, securities that fall outside of the existing reporting regimes include: (1) for TRACE: fixed-income transactions in securities with a maturity of one calendar year or less, such as money market instruments, and non-U.S. dollar-denominated debt are excluded from reporting requirements in TRACE; (2) for the CAT: equity transactions in “restricted securities,” as that term is defined in 17 CFR 230.144(a)(3) (“Rule 144”) of the Securities Act, are generally not reportable to the CAT because they are not subject to prompt last sale reporting rules; and (3) for RTRS: municipal securities transactions excepted under MSRB Rule G–14, including a small number of transactions for securities without assigned Committee on Uniform Securities Identification Procedure (“CUSIP”) numbers, municipal fund securities ( i.e., 529 Plans, ABLE programs, local government investment pools), and inter-dealer transactions ineligible for comparison of trade settlement date at a clearing agency. To provide consistency with existing reporting regimes, final Rule 10c–1a also incorporates these exclusions.

The Commission also received comments with suggestions to limit the scope of securities for which loans must be reported under the proposed rule. One commenter favored narrowing the proposed scope of securities to only equity securities that transact in significant volume in the U.S., namely equity securities listed or traded on a national securities exchange “to avoid uncertainty for non-U.S. borrowers and lenders transacting in securities that are primarily relevant in non-U.S. markets and to provide for a measured approach in introducing data reporting and dissemination into the marketplace.” Investors will also benefit from the increased transparency into loans of equity securities that are not listed or traded on a national securities exchange ( i.e., loans of equities that are traded over-the-counter (“OTC”)) provided by the final rule. The Commission understands that existing lending transparency systems currently include information about loans of exchange-listed and OTC securities. Thus, lenders should generally already be accustomed to providing and receiving such data. Further, OTC securities tend to be less liquid and publicly available information about an issuer of an OTC security may be limited or nonexistent, and thus such securities may be harder to borrow. Accordingly, data concerning loans of less liquid securities, such as OTC securities, will be beneficial to market participants by reducing information asymmetries and improving pricing efficiency by facilitating benchmarking for the loans of less liquid securities.

Commenters also requested clarification regarding which types of securities would be reportable, and some suggested limiting the scope of reportable securities. For example, one commenter suggested limiting the scope to equities for a period of time while the Commission evaluates whether to expand the scope of reporting to additional asset classes. Another commenter stated that the final rule should “capture the most relevant and appropriate securities for the reporting regime [referring to equity securities] while not precluding the inclusion of other securities over time as the SEC and market participants gain more experience with the reporting regime.” The definition of the term “reportable security” in the final rule is not limited to equities for the reasons discussed below, in this part. The definition of the term “reportable security” in paragraph (j)(3) of the final rule will help provide certainty to non-U.S. borrowers and lenders regarding the securities that are within the scope of the final rule ( i.e., securities or any class of an issuer's securities for which information is reported or required to be reported to the CAT, TRACE, or RTRS, or any reporting system that replaces one of these systems). In addition, Part VII.M below discusses the domestic application of the final rule, which should also help provide certainty to non-U.S. borrowers and lenders transacting in securities that are primarily relevant in non-U.S. markets about the application of the final rule. With respect to taking a measured approach in introducing data reporting and dissemination into the marketplace, the Commission is establishing the specific compliance dates discussed below in Part VIII. Such compliance dates should provide adequate time for covered persons to implement systems for data reporting and for RNSAs to implement data dissemination systems. Another commenter recommended that the Commission begin with reporting requirements for loans of U.S. equity securities because such loans “mostly occur on electronic trading platforms, making the generation of trade data for these loans more straightforward than loans of other asset classes.” As discussed above, the Commission has provided a compliance date that will provide adequate time for covered persons to comply with the final rule, even for loans of securities that may not occur on an electronic trading platform, and will not further delay transparency for OTC investors, which are primarily retail.

Some commenters requested a phased approach to the overall implementation of the final rule, which is discussed below, in Part VIII, including a recommendation that implementation of the final rule begin with loans of equity securities. However, current lending systems include data on loans of fixed income securities. Thus, lenders should generally be accustomed to providing and analyzing data on loans of fixed income securities. Further, as discussed below, in Part VIII, such a phased approach is not provided for in the final rule because it would unduly delay the public availability of Rule 10c–1a information and limit the initial benefits of the final rule.

One commenter suggested excluding debt from the proposed rule, particularly corporate debt and asset-backed securities, which the commenter described as having small issuance sizes, diverse characteristics, and a small number of lending transactions. However, information about loans on such securities will be beneficial to lenders, investors, and regulators precisely because of such characteristics. Data on such loans may be useful for lenders to assess rates and for regulators and investors to understand the lending market for such securities. For example, market participants will be able to use the information made publicly available by the final rule and pool such information across debt securities to help market participants compare the terms of a loan of one debt security with loans of other debt securities that have similar characteristics or time horizons. Thus, including debt securities in the final rule will help improve the ability of market participants to benchmark their loans of debt securities.

The Commission also received comment regarding the size of the debt market. One commenter stated that “[w]hile the number of publicly traded equity securities of U.S. issuers is around 3,600, there are over two million unique issuances of corporate and government bonds and asset-backed securities in circulation.” The same commenter stated that utilization rates decrease as products become more individualized and there is only a small market for shorting bonds. Less liquid markets should benefit from the increased transparency provided by the final rule because improving securities lending transparency, even in the debt market, can lead to reduced information asymmetry and increased pricing efficiency.

One commenter suggested excluding government securities from the proposed rule because “there is sufficient liquidity and demand for these securities on platforms and venues that have a high degree of transparency,” but did not provide details regarding the transparency of the platforms and venues. As discussed below, in Part IX.B.2, there is some transparency in the government securities lending market, but some information may not be as timely or as robust ( i.e., the data may contain aggregated volume information) as under the final rule. The final rule includes loans of U.S. Government securities between market participants to help ensure that comprehensive and timely information about these loans is made publicly available so that market participants in the government securities lending market benefit from increased transparency.

Some commenters requested clarification on whether “crypto assets” or as some commenters referred to them, “cryptocurrencies” would be covered by proposed Rule 10c–1. A crypto asset, which is also sometimes referred to as a “digital asset,” may meet the definition of a “security” under the federal securities laws. As defined in the final rule, paragraph 10c–1a(j)(3) applies to reportable securities, which are securities for which transactions are reported to the CAT, TRACE, and RTRS. Accordingly, if a crypto asset is a security that meets one of these same criteria, the crypto asset is a reportable security.

One commenter believed that infrequent lending of debt securities could result in a “far greater risk of loss of anonymity and the use of such data to anticipate or reverse engineer the trading of competitors.” The final rule's requirements for end-of-day reporting of data elements and loan modification data elements in paragraphs (c) and (d) and delaying the publication of loan amount and modifications to loan amount in paragraph (g) should help reduce the potential to anticipate and reverse engineer the trading of competitors. In addition, paragraph (g)(4) of the final rule requires that following the receipt of confidential information pursuant to paragraph (e) “an RNSA shall keep such information confidential, in accordance with the provisions of paragraph (h) of this section and applicable law.” Further, 17 CFR 240.10c–1a(h)(4) (“final Rule 10c–1a(h)(4)”) requires that an RNSA “[e]stablish, maintain, and enforce reasonably designed written policies and procedures to maintain the security and confidentiality of confidential information required by paragraph (e) of this section.” Thus, the only existing RNSA's experience with managing nonpublic information, coupled with the explicit requirements for protecting confidential information in paragraphs (g)(4) and (h)(4) of the final rule, will help ensure that an RNSA implements data security measures that will protect anonymity.

E. Scope of Transactions Required To Be Reported—Rule 10c–1a(j)(2)

Proposed Rule

The Commission proposed that any person that “loans a security” on behalf of itself or another person shall provide Rule 10c–1 information to an RNSA. The Commission did not define the term “loans a security,” but requested comment on whether a definition was necessary. In the Proposing Release, the Commission stated its preliminary belief that any person that loans a security on behalf of itself or another person should be required to provide the material terms to an RNSA to ensure that proposed Rule 10c–1 is appropriately “designed to increase the transparency of information available to brokers, dealers, and investors, with respect to the loan or borrowing of securities.”

Final Rule

Many commenters requested clarification of the scope of securities loan transactions to which the rule's reporting requirements would apply. One commenter stated that the proposed rule sought to regulate “a host of transactions that do not involve the `loan or borrowing of securities.' ” Other commenters recommended that a definition of the term “loans a security” refer only to lending for a particular purpose. Some of those commenters expressed concerns about including data for loans that were not considered traditional loans in the industry, particularly loans for which the parties did not enter into a written contract or transfer collateral, and that such loans might “skew the data and negatively impact the utility of the information provided.” Other commenters recommended that the final rule include a defined term for the types of transactions that are covered by the rule.

To address commenters seeking greater clarity and commenters recommending a definition as to what types of transactions are required to be reported, the final rule includes a definition of the term “covered securities loan” to mean “a transaction in which any person on behalf of itself or one or more other persons, lends a reportable security to another person,” with certain exceptions. As discussed below, in this part, the exceptions remove from the scope of the final rule's reporting requirements certain clearing agency positions and certain uses of margin securities by a broker or dealer that are not consistent with, or traditionally recognized as, securities lending transactions.

The Commission received a comment recommending that the term “loans a security” be limited to lending to unaffiliated borrowers. Other commenters suggested that information on securities lending transactions between affiliates not be disseminated under the rule, with one stating that, “loans between a broker-dealer and its affiliates, may not represent the actual rates available in the securities lending market and therefore could cause confusion and lead to misinterpretation if published.” The Commission does not agree that it would be appropriate to exclude loans to affiliated borrowers from the scope of the final rule. Inter-affiliate loans may be arms-length transactions and excepting such loans would result in the loss of data relevant to the lending market.

In addition, an exclusion for inter-affiliate loans could result in market practices to circumvent reporting obligations. For instance, in arranged financing, a broker or dealer “may offer arranged financing programs (sometimes called `enhanced lending' or `short arranging products') through which a customer can borrow shares from the firm's domestic or foreign affiliate and use those shares to close out a short position in the customer's account.” In describing arranged financing, a market participant has stated that “[w]ith respect to . . . arranged financing/enhanced lending models at member firms, certain ones may involve the loan of shares to customers from domestic affiliates and others may involve the loan of shares to customers from foreign affiliates.” Thus, if the broker or dealer agrees to provide a customer the covered securities loan through an affiliate, such as a foreign affiliate, an exclusion for inter-affiliate loans could exclude the securities loan entirely. Moreover, one researcher has observed that “owners of large portfolios . . . often conduct their own lending programs with an affiliated agent lender . . . .” An exclusion for inter-affiliate loans would not capture such loans. Therefore, under the final rule a covered person is required to report covered securities loans with affiliates.

Certain commenters recommended that a definition of the term “loans a security” refer only to transactions made pursuant to a written lending agreement or documented as a securities loan on the lender's books and records. One commenter stated that such an approach would “ensure consistent reporting and avoid confusion and misinterpretation of data by the public.” Another commenter stated that it could avoid “the public dissemination of incomplete, inaccurate, and misleading information that could have an adverse impact on the securities lending market.” The Commission disagrees that the definition of the term “covered securities loan” should require that the transaction be agreed to pursuant to a written securities lending agreement such as the Master Securities Loan Agreement, or a transaction documented as a securities loan on a lender's books and records. Such a requirement could provide an easy way to avoid the requirement, particularly for unregulated entities not otherwise subject to documentation and books and records requirements. Moreover, such a requirement could cause competitive harm to entities that are subject to such requirements or who follow such practices as risk management tools. Requiring the reporting of only securities loans made pursuant to a written agreement or subject to a covered person's books and records requirements would create confusion by introducing such distinctions, which may be interpreted or applied differently across the different types of persons that meet the definition of “covered person.” Additionally, such requirements would result in data that excludes reporting by covered persons that do not use written agreements or have books and records requirements. Such gaps in the data could contribute to the misinterpretation of data by the public, instead of ameliorating it as the commenter suggests. Thus, the final rule covers a wide variety of loans without limitation as to a particular loan's design or structure by including all loans that occur when a covered person “agrees to a covered securities loan.”

Some commenters recommended that the term “loans a security” specifically refer to the temporary lending of securities against a transfer of collateral. One of the commenters stated that such a provision would help “ensure consistent reporting and avoid confusion and misinterpretation of data by the public.” Another stated that defining the term “loan of securities” to include a transfer of collateral would “better achieve the Commission's goals” by avoiding “the public dissemination of incomplete, inaccurate, and misleading information.” Such a definition would have been consistent with the Proposing Release's statement that a securities loan is typically a fully collateralized transaction. However, the proposed rule did not limit the application of its reporting requirements to securities loans that were fully collateralized, as it was designed to capture all loans of securities and provide a more complete and timely picture of trading for securities loans. While the provision of collateral is currently a common industry practice for securities loans, it is not necessarily a characteristic of all securities loans. For example, some lenders may structure securities loans to include a bank guaranty, insurance policy, or other credit enhancement in lieu of collateral. Therefore, limiting the scope of loans covered by the final rule to only include lending against a transfer of collateral could leave certain loans out of the scope of the final rule's reporting requirements. The exclusion of uncollateralized loans could decrease the availability of pricing and activity information in the securities lending market, resulting in the reporting and dissemination of incomplete, inaccurate, and misleading information for which the commenters voiced concern. Furthermore, the reporting of uncollateralized loans would provide additional data to market participants, and would not result in misleading duplicative information like other types of transactions that are discussed below in this part ( e.g., loans arising from certain clearing agency services). Additionally, such a restriction could incentivize market participants to use credit enhancement mechanisms that differ from collateral ( e.g., bank guaranties or insurance policies) in order to avoid the final rule's reporting requirements. Therefore, the scope of final Rule 10c–1a's reporting requirements is not limited to only include loans against a transfer of collateral.

One commenter recommended that the Commission limit the scope of the rule's reporting requirements to securities loans where a lender seeks to earn compensation or a return from the transaction. The commenter stated that, “[l]imiting the scope in such a way would help focus reporting on primary transactions of interest, where the borrower is seeking to `gain access to the security itself,' and distinguish it from repurchase and other agreements, which are `typically used for short-term financing.' ” Another commenter proposed limiting the scope of securities loans to transactions in which a borrower obtains use of the securities for a fee. However, a limitation of the final rule's reporting requirements to only securities loans made by a lender seeking to earn compensation or a return from the transaction, or for a fee could potentially result in evasion, as securities loans could be structured ( e.g., via over-collateralization, haircuts, or use of non-cash collateral with varying maturities, credit ratings, or income characteristics) to avoid the identification of a fee. At the very least, such a requirement could result in confusion or inconsistent reporting as market participants may interpret differently the loan structures that incorporate a fee or are made by lenders seeking to earn compensation or a return from the transaction. If the final rule only applied to securities loans agreed to for compensation, a fee, or for the purpose of earning a return, securities loans made to affiliates could be structured to fall outside the final rule's reporting requirement ( i.e., made without an expectation of return or for no fee), while a loan still would have occurred, and the affiliate would still benefit from possession of the loaned securities. Further, the final rule does not exclude securities loans based on the purpose or duration of the loan, such as for “short-term financing,” in order to capture a wide variety of loans regardless of purpose or duration. Finally, the final rule provides that a covered securities loan is a transaction in which a person lends a reportable security, which distinguishes covered securities loans from other agreements.

Multiple commenters were concerned that the proposed rule would require that short sales or short positions be reported as securities loans. Such commenters requested clarification of whether the proposed rule would treat short sales as loans of securities. To provide such clarification, under the final rule covered persons will not be required to report short sales as defined by 17 CFR 242.200(a) (“Rule 200(a)”), but will be required to report loans that are used for short sales.

The Commission stated in the Proposing Release that it was not proposing to include repurchase and sale agreements (commonly known as “repos”) within the scope of the proposed rule because section 984 of the Dodd-Frank Act focuses on the loan or borrowing of securities. The Proposing Release detailed distinctions between the typical securities lending and repo markets, including that repos typically are primarily used for short-term financing while other securities loans typically are used to gain access to the security itself, and that loans generally allow the lender to recall the security on demand while repos do not. Some commenters agreed with the Commission, recommending that repos should not fall within the meaning of the term “loans a security” in the proposed rule. One commenter suggested that loans of a security should be distinguished from repos. However, one commenter stated that there is “significant overlap in the functionality between repos and securities lending transactions.” The commenter also suggested “a broad anti-evasion provision that prohibits any person from engaging in any practice intended to evade the rule's reporting requirements.” The same commenter recommended defining the term “securities loan” to include repos. The Office of Financial Research (“OFR”) recently proposed a rule to collect data on repos. Accordingly, at this time, it is not necessary to include repos within the scope of the final rule's information reporting requirements, as a potential OFR rule, if adopted, could address any potential reporting gaps, thereby reducing the incentives to evade final Rule 10c–1a.

The Commission also received comment requesting that the final rule “make clear that the novation and processing of a securities lending transaction by a registered clearing agency does not give rise to a new loan or the modification of an existing loan subject to reporting under Rule 10c–1.” The same commenter also stated that “[n]ovation is . . . the legal mechanism through which NSCC guarantees to each counterparty the performance of the obligations under a transaction that the counterparties already negotiated and executed away from the NSCC and that was already subject to a reporting obligation.” Another commenter directly responded to the Proposing Release's request for comment asking whether the Commission should define what it means to “loan a security.” The commenter stated that the definition should “expressly exclude loan positions that result from the central counterparty novation function that a clearing agency like OCC provides to securities loans.” The commenter, which is an SEC-registered clearing agency, also stated that it “does not have access to the majority of information sought under proposed Rule 10c–1 . . . and, therefore could not report such information to an RNSA.” The Commission agrees with the commenters' recommendations that positions that result from central counterparty services by a clearing agency, including the novation and processing of a securities lending transaction, should not give rise to any reporting obligation under the final rule, including the requirement to report loan modification data elements under paragraph (d) of the final rule. As discussed above, in Part VII.A, clearing agencies providing the functions of a central counterparty or central securities depository do not lend or borrow shares on their own behalf, but instead novate or process the loans of lenders and borrowers ( i.e., central counterparty or central securities depository services) for the purpose of efficient clearance and settlement. Although novation of the loan by the clearing agency technically creates a new loan in which the clearing agency would step into the shoes of the counterparties, the clearing agency would not be modifying the key terms of the loan (other than changing the identities of the counterparties to that of the clearing agency), such as the rate or size, and requiring clearing agencies to report such loans would result in the misleading appearance of additional loan volume containing otherwise duplicative loan information. Therefore, the final rule excludes from the “covered securities loan” definition “a position at a clearing agency that results from central counterparty services pursuant to Rule 17Ad–22(a)(2) of the Exchange Act or central securities depository services pursuant to Rule 17Ad–22(a)(3) of the Exchange Act.”

The Commission also received a comment stating that “it is not uncommon for a lending agent to pool together available supply of a given security across multiple lenders in their lending program” to satisfy a large loan of securities and reallocate such loan among the lenders in the program, as the inventory of individual lenders changes, to avoid recalling a loan. The commenter also stated that, in such instances, “[g]iven no change in the economics of the trade or any physical movement of securities, these intraday record entries are not market trades and should not be reported as such.” However, the proposed rule did not limit its information reporting requirements to securities loans arising from “market trades” or the “physical movement of securities,” neither of which the commenter has defined or described in the context of the securities lending market. Changing the party or parties to a covered securities loan, which is required to be confidentially reported under 17 CFR 240.10c–1(e)(1) (“final Rule 10c–1a(e)(1)”), creates a new covered securities loan that would require reporting as a new covered securities loan to an RNSA under final Rule 10c–1a(a)(1), and not as a modification under final Rule 10c–1a(d). The identity of the lender or lenders is a material term of a covered securities loan. The identity of the lender is not made public, but is important information for regulatory purposes, such as surveillance of activity pertaining to the individual loan. Whether the parties to a covered securities loan change for purposes of the reporting requirements under final Rule 10c–1a(e)(1) depends on how a pool or lending program is structured ( e.g., whether the pool or lending program itself or the individual underlying participants are the party or parties identified as the lender for the loan). For example, if a lending program as an individual entity is the party that is the lender of the covered securities loan, changes to the underlying participants, inventory providers, or customers of that lending program will not constitute a change to the parties of the covered securities loan. In that instance, unlike the lending program, the individual participants will not be lenders directly to the borrower of the covered securities loan. In that case, where a covered securities loan remains open and the only change that occurs is a reallocation among the pool's underlying constituents, that is not a change that will require reporting as a new covered securities loan or as a modification under paragraph (d) because there will be no change to a data element in paragraph (c) of the final rule.

However, if the multiple, individual participants are all parties identified as lenders to the loan, with no lending program or other entity interposed between them and the borrower, a change in their composition (the removal or addition of a lender) would constitute an assignment of the loan and therefore would require reporting as a new loan pursuant to final Rule 10c–1a(a)(1). Whether a reallocation of a loan among participants in a lending program requires the reporting of a new covered securities loan depends upon the facts and circumstances, including the structure of such lending program.

One commenter recommended a definition of the term “loans a security” that would apply at a minimum “any time the lender loses the right to vote the securities during the time of the loan.” Voting rights do not provide an adequate criterion to define what a securities loan is because not all reportable securities necessarily carry voting rights ( e.g., some preferred stock, options, fixed-income securities, and treasuries). However, the facts and circumstances, including the loss of voting rights, may be an indicator as to whether a transaction is a loan.

The Commission also received comment letters expressing support for limiting the reporting required under the proposed rule to the Wholesale market of the securities lending market. Some commenters recommended narrowing the scope of loans to the Wholesale market due to what they characterized as potential negative effects on short selling from data that could be revealed through loans in the Customer market. One of the commenters stated that “immediately publicly disclosing short selling activity would signal to all other market participants that a short position is being established.” Such concerns are largely addressed by requiring that an RNSA delay public disclosure of the loan amount by 20 business days. This modification should significantly reduce the novelty of the information disseminated under the final rule regarding short sellers' positions, such that it is less timely than pre-existing sources of short selling transparency, such as FINRA's bimonthly short interest data.

Another commenter suggested that retail loans may not be structured as securities lending, but rather structured as brokerage agreements, and thus may not be suitable for standardized data collection. This commenter may be referring to the use of margin securities by a broker or dealer in which the broker or dealer does not lend such margin securities to another person ( e.g., a rehypothecation). The final rule's exclusion of such uses from the definition of “covered securities loan,” as discussed below, in this part, should address the commenter's concern.

A commenter stated that the Commission had not fully explained why greater transparency, or transaction-by-transaction data, in the Customer market would be valuable. More broadly, limiting data to Wholesale market loans would significantly reduce the benefits of the rule stemming from increased transparency into the securities lending market. In the Proposing Release, the Commission stated that the rule “is designed to address . . . inefficiencies in the securities lending market by making more comprehensive information regarding securities lending transactions publicly available, which could better protect investors by eliminating certain information asymmetries.” Data regarding Customer market loans will, for example, provide end-borrowers with valuable information as to the competitiveness of the rates they are being charged for their loans. Such data will also provide information to customers that have agreed to loan their fully paid securities as to the competitiveness of the rates they are receiving from brokers or dealers borrowing such securities. Excluding such a fundamental part of the securities lending market from the scope of transactions to which the rule's reporting requirements apply is inconsistent with the Commission's stated goal of making comprehensive information regarding securities loans publicly available. Therefore, the definition of “covered securities loan” in the final rule does not distinguish between the Wholesale market and Customer market and is not limited to Wholesale market transactions.

One commenter recommended that the term “loans a security” specifically refer to securities loaned “for a permitted purpose pursuant to Regulation T of the Board of Governors of the Federal Reserve System.” However, limiting the final rule's reporting requirements to only securities loans made for permitted purposes pursuant to Regulation T of the Board of Governors of the Federal Reserve System, would be overly narrow as the principal purpose of Regulation T is to regulate extensions of credit by brokers and dealers. As the commenter states, the “permitted purposes” contemplated under section 220.10(a) of Regulation T are limited to “allowing the borrower to make delivery of the borrowed securities in the case of short sales, failure to receive securities required to be delivered, or other similar situations.” Regulation T states that “[i]ts principal purpose is to regulate extensions of credit by brokers and dealers; it also covers related transactions within the Board's authority under the Act. It imposes, among other obligations, initial margin requirements and payment rules on certain securities transactions.” However, the securities loan market includes lenders other than brokers and dealers and may involve loans of securities for purposes other than short sales, failures to receive securities required to be delivered, and similar situations. Furthermore, the purpose of final Rule 10c–1a is to increase the transparency of information available to broker, dealers, and investors, with respect to the loan and borrowing of securities without limitation to the purpose of the loan or borrow, whereas the stated purpose of Regulation T is regulating extensions of credit. Thus, Regulation T does not provide an appropriate framework to limit final Rule 10c–1a's information reporting requirements.

The Commission also received comment expressing support for a de minimis threshold for reporting loans of securities. In some contexts, reporting thresholds may be useful to reduce burdens on persons required to report, or to reduce the possibility that strategies or identities of reporters will be revealed. However, a de minimis reporting threshold is not necessary in light of modifications from the proposed rule to require end-of-day (versus 15 minute) reporting, as well as the delay in publication of size information, which are expected to reduce the possibility that strategies or identities of reporting entities will be revealed. Including a de minimis threshold, which would primarily affect data for smaller securities loans, could provide the public with a distorted view of securities loan activity. For example, a large volume of relatively small securities loans could constitute a significant amount of the overall lending activity for an individual security. Alternatively, a single, small securities loan could constitute the only outstanding loan for an issuer, and its exclusion would leave market participants without information about the lending market for the security. The Proposing Release stated that “granular information about certain material terms of securities lending transactions would allow investors, including borrowers and lenders, to evaluate not only the rates for such transactions, but also any signals that such rates provide.” The public disclosure of information for securities lending transactions of all sizes will help improve price discovery in the securities lending market. Additionally, providing a de minimis threshold could result in avoidance of the final rule's reporting requirements if industry participants structure otherwise reportable securities loans into smaller tranches in order to fit below a reporting threshold.

The Commission received a comment recommending that the final rule not capture a “broker-dealer hypothecation of customer margin securities.” Another commenter stated that it understood the proposed rule to exclude rehypothecation of a customer's margin securities. When a broker or dealer hypothecates (or uses) customer margin securities, the customer is not loaning them to the broker or dealer as the customer has already pledged the securities in a margin account to the broker or dealer, as collateral for a margin loan. Accordingly, the final rule excludes hypothecation of securities, but provides that the loan of such customer margin securities by a broker or dealer to another person is a covered securities loan. Specifically, the definition of the term “covered securities loan” under the final rule includes the provision that, “the use of margin securities, as defined in Rule 15c3–3(a)(4) of the Exchange Act, by a broker or dealer will not be a covered securities loan for purposes of this rule.” It also provides that “if a broker or dealer lends such margin securities to another person, the loan to the other person is a covered securities loan” for purposes of the final rule. Therefore, should a broker or dealer use a customer's margin securities for purposes other than to lend them to another person, such a transaction would not fall within the definition of the term “covered securities loan” under the final rule. Additionally, if a broker or dealer borrows fully paid or excess margin securities from a customer, that loan is a covered securities loan, and the broker or dealer borrowing the fully paid securities is responsible for reporting the loan to an RNSA.

The Commission also received comment recommending that the final rule not apply to “funding trades.” The commenter further stated its belief that such transactions, which are defined as loans of cash against securities being pledged as collateral, are not the type of activity that should be captured through the proposed rule. To the extent that such funding transactions constitute a use of margin securities by a broker or dealer pursuant to paragraph (j)(2)(iii), they would not fall within the final rule's definition of “covered securities loan.” Otherwise, commenters were not specific as to whether there would be variations of transactions used to “fund trades.” The final rule does not specifically exclude transactions to “fund trades.” Whether a transaction to “fund trades” is a covered securities loan will depend on the facts and circumstances of each transaction.

After considering commenters' concerns and perspectives regarding the proposed scope of transactions for which proposed Rule 10c–1 reporting would have been required, the final rule defines the term “covered securities loan” to mean “[a] transaction in which any person on behalf of itself or one or more other persons, lends a reportable security to another person.” Further, in paragraph (a) of the final rule, the term “loans a security” is replaced with the phrase “agrees to a covered securities loan” in its reporting requirement for covered persons. In response to commenters' request for clarification as to whether a loan should be reported before or after it is settled, the term “agrees to” clarifies that covered securities loans are required to be reported after the parties agree to the loan, which is before settlement.

For the reasons described above, in this part, the definition of the term “covered securities loan” in the final rule further provides that “a position at a clearing agency that results from central counterparty services pursuant to Rule 17Ad–22(a)(2) of the Exchange Act or central securities depository services pursuant to Rule 17Ad–22(a)(3) of the Exchange Act will not be a covered securities loan for purposes of this rule.” The final rule also provides that “the use of margin securities, as defined in Rule 15c3–3(a)(4) of the Exchange Act, by a broker or dealer, will not be considered a covered securities loan for purposes of this rule,” provided, however, that “if a broker or dealer loans such margin securities, such loan will be considered a covered securities loan for purposes of this rule.”

The definition of the term “covered securities loan,” in conjunction with the definition of the term “reportable security,” discussed above, in Part VII.D, sets forth the scope of the final rule and distinguishes “covered securities loans” from other types of transactions that are not required to be disclosed under final Rule 10c–1a. The final rule's definition of the term “covered securities loan” will facilitate the public availability of information regarding securities lending transactions. The definition's scope, encompassing transactions in which any person, on behalf of itself or one or more other persons, lends a reportable security to another person, is designed to provide market participants, the public, and regulators with information that broadly captures activity in the securities lending market. The defined scope of transactions required to be reported to an RNSA under final Rule 10c–1a will contribute to better decision-making by investors, reduced costs of business for brokers or dealers, improved performance and reduced costs for lending programs, new business opportunities for data vendors, improvements to shareholder monitoring, and improved market stability and price discovery in the securities lending market. Additionally, the exceptions to the definition of “covered securities loan” under the final rule should help prevent the double counting of securities loans and support the integrity of publicly available data by excluding redundant and potentially misleading information.

F. Information To Be Provided to an RNSA

To facilitate transparency of the securities lending market, the Commission proposed Rules 10c–1(b) through (e) to require certain loan-level data elements ( i.e., specified material terms of securities loans) to be provided to an RNSA. Because an RNSA would be required to implement rules regarding the format and manner to administer the collection of information, proposed Rule 10c–1 lists the data elements that persons would be required to provide to an RNSA, but does not specify granular instructions for data elements or the formatting required for submission of the information to an RNSA.

1. Loan Data Elements—Rule 10c–1a(c)

Proposed Rule

To facilitate transparency in the securities lending market, the proposed rule required certain loan transaction data elements to be provided to an RNSA, on a transaction-by-transaction basis, within 15 minutes of the loan being effected, followed by an RNSA assigning each loan a unique transaction identifier, and then making such information publicly available as soon as practicable. The proposed rule included twelve transaction data elements required to be provided to an RNSA.

In proposing the list of loan data elements required to be provided to an RNSA, the Commission explained that the data elements in paragraphs (b)(1) through (b)(5) of the proposed rule, which include the legal name and legal entity identifier (“LEI”) of the issuer; ticker symbol, International Securities Identification Number (“ISIN”), CUSIP, or Financial Instrument Global Identifier (“FIGI”) or other security identifier; date and time loan was effected; and platform or venue loan was effected, identify each loan of securities. In particular, the Commission explained, they contain material terms that are not negotiated between the parties but are elements that would provide important information to allow market participants and regulators to track, understand, and perform analyses on the negotiated terms contained in paragraphs (b)(6) through (b)(12) of the proposed rule, which include the amount of security loaned; securities lending fee or rate (or any other fee or charges); type of collateral used; rebate rate (or any other fee or charges); percentage of collateral to value of loaned securities; termination date of loan; and, whether the borrower is a broker/dealer/customer/clearing agency/bank/custodian/other person. The Commission also explained that the proposed data elements in paragraphs (b)(1) through (b)(5) of the proposed rule would provide an RNSA with enough information to create a unique transaction identifier as required by the proposed rule.

In proposing the negotiated data elements in paragraphs (b)(6) through (b)(12) of the proposed rule, the Commission explained that, because the data elements reflect material terms that are negotiated when securities loans are arranged, increasing the transparency of information will provide market participants with meaningful data that could be used when structuring, pricing, or evaluating loans of securities. The Commission also explained that increasing transparency would allow market participants to analyze signals obtained from the securities lending market when considering investment or trading decisions for a security; and, would also permit an RNSA to perform in-depth monitoring and surveillance of securities lending transactions to identify trends and any anomalous market patterns.

Final Rule

The Commission sought comment with respect to each of the proposed transaction data elements. In response, commenters were generally supportive of the Commission's stated goals of increasing transparency and price discovery in the securities lending market by increasing the amount and availability of data in the market. For example, one commenter stated in support of the proposed rule that, “certain data elements [should be made] publicly available, thereby increasing transparency of the securities lending market and reducing competitive advantages that may exist in the marketplace.” Another commenter also stated in support that the proposed public dissemination of securities lending information will, among other things, improve price discovery in the securities lending market, reduce information asymmetries, close data gaps, and increase market efficiency. Another commenter stated that, “Proposed Rule 10c–1 represents a properly-tailored way to bring more transparency to this dark area of the market.” According to this commenter, “[t]he Proposal would establish a system for the reporting and dissemination of the material terms of securities lending transactions without attribution, providing issuers, investors, and regulators the necessary data . . . while protecting the identity and intellectual property of any individual market participant.” Another commenter expressed support for the proposed rule's inclusion of standards as part of the proposed data elements to be reported, as well as the Commission's efforts to include the LEI in the data elements for the identification of issuers and parties to a lending transaction. Another commenter supported the proposal allowing a choice in use of identifiers, such as the [FIGI], as a possible required data element to be collected and disseminated by an RNSA.” Another commenter, in response to the Commission's specific request for comment, stated that retail investors/borrowers without LEIs should not be required to obtain a LEI as “[r]egulators can easily identify who we are by going to our brokers.” Another commenter suggested that the Commission should modify the proposal to require “lending agent” be a reported field (instead of requiring the lender's LEI to be disclosed) “to avoid the potentially confusing appearance of tens or even hundreds of individual loans that are, in reality, part of the same overall loan transaction negotiated between a borrower and a lending agent or third-party intermediary on behalf of the underlying lender(s) for an aggregate notional amount.”

With regard to the proposed “rebate rate or any other fee or charges” data element for cash collateralized loans, several commenters pointed out that the relevant data point is actually the lending spread to the reference rate (most commonly the Overnight Bank Funding Rate or “OBFR”) as the rebate rate will change daily based on the current level of the reference rate, even as the negotiated lending spread remains fixed. Thus, commenters requested that the Commission clarify (where applicable) that pricing data may be reported as a spread to a benchmark rate and, as discussed below in Part F.2, that such pricing data element does not need to be updated for changes in the value of the benchmark rate.

Another commenter expressed concern that fee and rebate data could be misleading in the absence of additional contextual information, stating that “[f]ee/rebate data, without consideration of firm(s) collateral requirements, counterparty/asset exposure(s), applied lending restrictions or jurisdictional obligations, may create an unrealistic and misleading portrayal of prevailing rates.” The final rule requires pricing information in 17 CFR 240.10c–1a(c)(8) (“final Rule 10c–1a(c)(8)”) and 17 CFR 240.10c–1a(c)(9) (“final Rule 10c–1a(c)(9)”) because pricing is a material term of a covered securities loan. Further, paragraph (g)(5) of the final rule requires that an RNSA will make a distribution of loan rates for each reportable security publicly available to help market participants compare the pricing of their covered securities loan against the pricing of other covered securities loans. Information about the distribution of loan rates recognizes that the cost-to-borrow securities can be influenced by a number of factors and can give market participants information to help compare the pricing of their loan against other loans. Another commenter requested clarification that only one identifier for a security is required to be reported under the rule ( i.e., ticker or ISIN or CUSIP or FIGI) rather than all of them. The final rule permits a covered person to determine which identifier to report, or if it prefers, it may also report multiple identifiers, but one identifier is required to be reported. Further, the final rule also permits the covered person to report a different identifier than the ticker symbol, ISIN, CUSIP, or FIGI. In addition, this commenter requested further clarification on what type of system would represent a “platform or venue” for purposes of proposed paragraph (b)(5). The final rule does not specify what type of systems are a “platform or venue” for purposes of the data elements required to be reported under 17 CFR 240.10c–1a(c)(5) (“final Rule 10c–1a(c)(5)”). The terms are intended to capture a wide variety of potential vehicles that covered persons might use to transact securities loans. Further, an RNSA has discretion to determine, pursuant to rules that would be subject to the section 19(b) and Rule 19b–4 process, whether to categorize such vehicles ( e.g., on-line venue, exchange, OTC).

Many commenters suggested that the proposed transaction-by-transaction reporting requirement be replaced with an aggregated-only requirement ( i.e., with some of the commenters wanting both the reporting and publishing of the data elements to be on an aggregated basis only, while other commenters were supportive of reporting to an RNSA on a transaction-by-transaction basis, but then any dissemination to the public must be made by an RNSA on an aggregated-only basis). For instance, one commenter recommended “transaction-by-transaction data be reported to an RNSA and made available to regulators, but that the RNSA analyze and normalize the reported transaction data to provide aggregated and, where appropriate, averaged transaction terms that do not expose firm and customer investment strategies, and that better reflect the transaction terms available to lenders and borrowers in the securities lending market.” Another commenter stated that the Commission should narrow the scope of transaction data information to be reported to an RNSA (and thereafter made public by an RNSA) to only aggregated transaction data. Some of the comments appeared to be focused on an RNSA's publication or dissemination obligations when they stated that the final rule should require an RNSA to publish aggregated pricing and volume data (rather than transaction-by-transaction data). However, this was not always clear as many of the comments received were less specific in that their comments on aggregated disclosure were phrased in reference to the proposed rule's reporting requirement or “reveal short selling trading strategies” or disclosing all individual customer short selling positions on a transaction-by-transaction basis.

The requirements in paragraph (b) of final Rule 10c–1a are limited in scope to the reporting agent's obligation to provide certain specified data elements to an RNSA on behalf of a covered person. As such, the discussion in this part is limited to the reporting agent's provision of the Rule 10c–1a information directly and solely to an RNSA, which does not involve or require any public disclosure or public dissemination of such information. By contrast, the dissemination or disclosure of the required information is the obligation of an RNSA, and paragraph (g) of the final rule, and Part VII.J below, discuss an RNSA's responsibilities with respect to disseminating certain Rule 10c–1a information to the public.

After considering the comments received recommending that the final rule require only aggregated pricing and volume data to be publicly disclosed (discussed below in Part VII.J), the Commission has determined not to modify the proposed transaction-by-transaction reporting requirement with respect to the provision of Rule 10c–1a information to an RNSA. Such a modification to paragraph (c) of the final rule is unnecessary because the requirement involves reporting of the information directly and only to an RNSA (and not to the public). As such, the reporting requirement does not present the concerns raised by commenters that would warrant requiring the reporting of the data elements on an aggregated-only basis, as was suggested by some commenters. Likewise, another commenter suggested that, to decrease the likelihood that published data will be potentially misleading or confusing or that it could reveal short trading strategies that could prompt short sellers to exit the market, the Commission should adjust the information that is made publicly available by an RNSA to be only aggregated securities lending data, including, among other things, a volume-weighted average borrowing fee aggregated across all firms, for each security loaned. Similarly, for the same aforementioned reasons, it is unlikely that providing data information directly and solely to an RNSA will reveal such short trading strategies or mislead or confuse the public to warrant such a modification to the final rule. Instead, because paragraph (c) of the final rule only requires the reporting of information to an RNSA (rather than to disclose the information to the public), the risk of revealing short trading strategies or confusing the public is not applicable. Moreover, an RNSA's responsibilities with respect to the subsequent publication of certain Rule 10c–1a information are subject to the requirements and limitations set forth in paragraph (g) of the final rule, which are discussed below, in Part VII.J. Thus, the Commission is adopting the proposed requirement to report Rule 10c–1a information, on a transaction-by-transaction basis, directly and solely to an RNSA.

In addition, the Commission is adopting, essentially unchanged, the twelve data elements originally set forth in paragraphs (b)(1) through (b)(12) in the proposed rule, except the Commission is making the following non-substantive changes to the language of some of the proposed data elements. Specifically, the Commission is modifying the language of the “the amount of the security loaned” data element in paragraph (b)(6) of the proposed rule ( i.e., by adding the language “such as the size, volume, or both)” so that it instead reads, as adopted, “the amount, such as size, volume, or both, of the reportable securities loaned.” This change will help ensure the compliance and accuracy of the Rule 10c–1a information submissions to an RNSA, by specifying in the final rule that the term “amount” means “size, volume, or both.”

The Commission also is modifying proposed paragraph (b)(6) by changing “security” to “securities” (thus, making it plural, consistent with the term “securities” in paragraph (b)(10)); and, also modifying both proposed paragraphs (b)(6) and (b)(10) by adding the term “reportable” to each, so that they are both consistent with the final rule's newly defined term, “reportable securities.” The Commission also is modifying language in proposed paragraph (b)(1) by changing the term “active LEI” to “non-lapsed LEI” for accuracy ( i.e., by focusing more precisely on the status of the application for the LEI-indicator, rather than on the entity itself). The Commission has modified proposed paragraph (b)(2) by providing the full names of specified security identifiers and retaining the corresponding abbreviations within a parenthetical with quotes to 17 CFR 240.10c–1a(c)(2) (“final Rule 10c–1a(c)(2)”).

The required pricing data, “[f]or a loan collateralized by cash, the rebate rate or any other fee or charges,” in paragraph (c)(8) of the final rule generally should include when the pricing data element is negotiated or agreed-to as a spread to an identified benchmark or reference rate. The Commission recognizes that the pricing data element is a material term of the loan that is often negotiated or agreed to between the parties. Consistent with paragraph (b)(9) of the proposed rule, as well as the final rule, this benchmark type of pricing, as described by the commenter, will be captured by the final rule text language “the rebate rate or any other fee or charges.” For the aforementioned reasons, negotiated or agreed-to pricing data, including as a spread to an identified benchmark or reference rate, is required to be reported to an RNSA pursuant to the rule as adopted. This requirement should generally help to address commenters' concerns that may stem from an overly narrow application of the pricing data element in paragraph (b)(9) of the proposed rule. However, because there are various pricing terms or arrangements that can be negotiated or agreed to by parties to a securities loan, the final rule does not specify or endorse one particular pricing method over another and, thus, is adopting the sufficiently broad pricing data element rule text for purposes of the final rule without modification from the proposed rule text. An RNSA, nevertheless, could address in its rules details regarding the manner in which pricing data must be reported, such as further clarifying how lending cost information is to be reported. As such, an RNSA has the necessary flexibility and discretion to structure its systems and processes as it determines, consistent with its obligations under the final rule as well as other applicable provisions of the Exchange Act.

2. Loan Modification Data Elements—Rule 10c–1a(d)

Proposed Rule

Subject to terms agreed to by the parties, covered securities loans may be modified after they are made. To ensure that the loan data elements reported pursuant to proposed Rule 10c–1(b) would accurately reflect currently outstanding covered securities loans and to prevent evasion, proposed Rule 10c–1(c) required that certain loan modification data elements be provided to an RNSA within 15 minutes after each loan is modified if the modification resulted in a change to information required to be provided to an RNSA under paragraph (b) of the proposed rule (and then for an RNSA to make such information available to the public as soon as practicable). More specifically, the proposed loan modification data elements in paragraphs (c)(1) through (c)(3) of the proposed rule required the following data elements to be provided to an RNSA: (1) the date and time of the modification; (2) a description of the modification; and (3) the unique transaction identifier of the original loan.

In the Proposing Release, the Commission explained that it preliminarily believed the proposed loan modification data elements in proposed paragraph (c) are necessary to allow an RNSA to identify which loan was being modified, categorize the type of modification, and make information about the modification publicly available. In addition, as proposed, the requirement to provide the loan modification information to an RNSA was conditioned on the modification resulting in a change to the Rule 10c–1 information required to be provided to an RNSA under paragraph (b) of the proposed rule.

Final Rule

The Commission is adopting the loan modification data elements substantially as proposed. However, the final rule includes a change to the loan modification data elements requirement in response to commenters. Other commenters sought clarification as to when loan modifications should be reported, which is discussed below. In addition, the final rule has been modified to address comments regarding loans that pre-exist the reporting date.

Paragraphs (d)(1)(i) through (iii) of the final rule set forth the data elements that a covered person must report by the end of the day on which a covered securities loan is modified. Paragraph (d)(1)(i) requires covered persons to report “[t]he date and time of the modification;” paragraph (d)(1)(ii) requires covered persons to report “[t]he specific modification and the specific data element in paragraph (c) of this section being modified; and paragraph (d)(1)(iii) requires the covered person to report “[t]he unique identifier assigned to the original covered securities loan under paragraphs (g)(1) or (g)(3) of this section.”

With respect to paragraph (d)(1)(ii), to help ensure the provision of relevant and useful loan modification reports to an RNSA, the Commission has revised this loan modification data element from the proposed requirement by removing the phrase “[a] description of” and, instead, adding both “the specific modification” as well as “and specific data element in paragraph (c) being modified” ( i.e., with respect to the loan modification data element being reported to an RNSA). The revised language “the specific modification” makes explicit that the actual modification must be reported ( e.g., the amount of the security loaned increased by 200 shares), not a vague description of the modification ( e.g., the amount of the security loaned was increased). The Commission is making this change because it will result in the collection of more accurate and useful information concerning the modification being reported. The Commission is also modifying paragraph (d)(1)(iii) to require the reporting of a previously assigned unique identifier whether assigned at the time of execution of the loan, or, with respect to a pre-existing securities loan as discussed below in this section, at the time of the modification. As a result, the Commission is adopting paragraph (d)(1)(iii) of the final rule, with a slight modification from the proposal such that the rule text of (d)(1)(iii) will require the unique identifier assigned to the original covered securities loan under paragraphs (g)(1) or (g)(3) of this section.”

The Commission also received comment seeking clarity regarding how to report various modifications to a loan. Other commenters who discussed the loan modification data elements discussed the different types of modifications that can occur frequently throughout the day, or sought further information on what types of modifications would likely trigger a loan modification reporting obligation under the final rule. One commenter stated that, in the context of cash collateralized loans, the rebate will fluctuate daily as the reference rate value changes, but suggested that loans where the selected reference rate and spread to that reference rate do not change should be deemed out of scope of the required loan modification reporting. As an example, this commenter stated that, where a loan is collateralized by cash and the negotiated pricing data element is a spread to a benchmark rate, such pricing does not need to be updated for changes in the value of the benchmark rate ( i.e., not a reportable loan modification for purposes of the final rule). The same commenter stated more generally that certain modifications to loan information that has already been provided to an RNSA pursuant to proposed Rule 10c–1 should not be required to be provided to an RNSA, claiming that could make data that was already made public potentially misleading. As discussed above, in Part VII.F.1, the pricing data elements required to be reported by paragraph (c)(8) include benchmark pricing, such as a reference to the OBFR in a manner determined by an RNSA. Paragraph (d) of the final rule, requiring reporting of modifications to a paragraph (c) data element, including pricing in paragraph (c)(8), will also capture benchmark pricing. The extent to which the commenter's concern is realized will be determined by how an RNSA chooses to structure the reporting of this variable. For instance, if an RNSA chooses to allow market participants to report a spread and a benchmark, then no modifications will be required to be reported from day to day unless there were a change in the negotiated spread or benchmark. However, if an RNSA chooses to require market participants to report the total fee, then market participants will be required to report changes to the fee if the benchmark changes, which can require daily revisions. However, the Commission does not expect that this will cause confusion. The Commission understands that market participants know that rebates can change regularly and, thus, revisions would not be unexpected. Further, gathering loan modification data is important to facilitate the accurate computation of statistics regarding the cost to borrow by an RNSA and other market participants.

One commenter suggested that the termination of a loan should be reported based on the return of securities by the borrower. Under the final rule, if a borrower returns a portion of loaned shares, the reduction in the amount of shares remaining on loan will be reported as a modification. Similarly, under the final rule, a termination of a covered securities loan is a modification for which information, including the termination date and the reduction of the loaned shares (if any), is required to be reported to an RNSA.

Events such as consolidations and splitting of loans are required to be reported. A consolidation would be reported as the termination (and therefore modification) of existing loans and entry into a new loan, including the reporting of all required data elements for those modifications and for the new loan. Similarly, the splitting of an existing loan into separate loans could be reported as a modification and termination of the original loan and entry into multiple separate loans. However, that there may be certain life-cycle events in the course of an open-ended loan that some market participants may view as a modification to an existing loan that other market participants might view as a termination of an existing loan and the entry into a new loan. For example, when all outstanding loaned securities are returned to the lender and additional (or the same) shares are subsequently lent under an open-ended loan, one lender may view that event as a termination (and therefore a modification) of an existing loan and a creation of a new loan, whereas another lender may view a similar transaction as two modifications (the return of, and subsequent loan of, securities under the same open-ended loan). In such cases, the lender (or lending agent or reporting agent) may elect to report the required information as either a modification (and termination), or as two modifications to an open-ended loan.

Reporting of Data Elements for Pre-Existing Securities Loans

The Commission also received comment seeking clarification on what some commenters called day-one loans, or loans that are agreed to prior to the reporting date of the final rule. One commenter described the day-one problem and stated, “we suggest the Commission find a solution to ensure these loans do not go unreported under a new reporting regime.” Other commenters stated that the Commission should consider ways to report and identify such loans.

Certain commenters stated that many loans are open-ended and continuously modified, and sought clarity regarding the treatment of loans in existence prior to the final rule's reporting date for covered persons that are subsequently modified, which commenters characterized as the day-one problem. For example, one commenter stated that the proposal does not consider a key implementation question regarding the day-one problem for reporting existing loans, specifically, that “most loans are open-ended without a set termination date. Accordingly, loans are continually resized and rerated until either the lender or the borrower recalls or returns all shares, respectively. Given the longevity of the average loan, there will be a substantial number of loans that exist prior to the implementation date of the reporting requirements, and such loans will likely continue to be modified as long as they remain outstanding.” This commenter provided three possible paths for handling such day-one loans: “(1) report each loan the first time it is modified after the implementation date as if it were a new loan, (2) provide all existing loans with an identifier on the day of implementation, or (3) exclude all existing loans from all reporting obligations, including reporting of modifications made to those loans after implementation date.”

The commenter's first alternative for reporting each “day-one” loan the first time it is modified provides an appropriate means of on-boarding data for such loans, as discussed below. As discussed below, in Part IX.C.1, the lack of information about these loans would harm data quality because it would render the data less complete and would thus limit market participants' ability to determine conditions in the lending market. Accordingly, paragraph (d)(2) of the final rule provides that if a modification is made to a covered securities loan for which reporting under paragraph (a) of the final rule was not required on the date the loan was agreed to or last modified (a “pre-existing covered securities loan”) and results in a change to any of the data elements in paragraph (c) of the final rule, there is an obligation to report each of the data elements in paragraph (c) of the final rule as of the date of the modification. Paragraph (d)(2) will provide a snap shot of a pre-existing covered securities loan the first time it is modified on or after the reporting date. If, for example, under the final rule, a pre-existing covered securities loan is first modified two months after the reporting date and the modification results in a change to a data element in paragraph (c) of the final rule, the covered person must provide each of the data elements under paragraph (c) to an RNSA, including the date and time the pre-existing covered securities loan was effected pursuant to 17 CFR 240.10c–1a(c)(3) (“final Rule 10c–1a(c)(3)”) and 17 CFR 240.10c–1a(c)(4) (“final Rule 10c–1a(c)(4)”), which will indicate that this is a pre-existing covered securities loan. Reporting of the date and time that the pre-existing covered securities loan was originally effected will provide market participants and regulators with a more complete picture as to the duration of the outstanding loan that is being modified.

In addition to the date and time that the pre-existing covered securities loan was effected ( i.e., information in paragraphs (c)(3) and (c)(4)), the covered person must also provide information in paragraphs (c)(1), (c)(2), and (c)(5) through 17 CFR 240.10c–1a(c)(12) (“final Rule 10c–1a(c)(12)”) as of the date of the modification. The date and time of the modification will provide market participants with specificity and context regarding the terms of the modifications, including the rates available for modification at that date and time (which may be different than rates available for new loans). Including the date and time will be important because it will indicate if a report of a modification is not filed in a timely fashion.

If, for example, a pre-existing covered securities loan is first modified, either on or after the reporting date, and it is a modification to the loan amount, the covered person will be required to report the number of shares as modified under paragraph (c)(6) of the final rule to an RNSA, as well as each of the other data elements in paragraph (c), including the date and time the pre-existing covered securities loan was effected, by the end of the day on which such loan was modified. In that same example, the date and time of the modification will be required to be reported. Recognizing that a unique identifier will not yet exist for the loan and no data about the loan was previously reported or made publicly available under the final rule, the data elements in (d)(1)(ii) and (d)(1)(iii) will not be required to be reported for such pre-existing covered securities loans. However, any subsequent modifications going forward will be subject to the modification reporting requirements under paragraph (d)(1) of the final rule such that only the information in paragraphs (d)(1)(i) through (iii) of paragraph (d) will be required to be reported to an RNSA. Pre-existing covered securities loans that do not have a modification to a data element in paragraph (c) will not be subject to the reporting obligation under paragraph (d)(2) of the final rule.

The Commission is not adopting the commenter's alternative of requiring reporting of all pre-existing covered securities loans on the first reporting date. Such an alternative would result in costs and burdens for an RNSA to build out the capacity to handle a much larger number of reports on the first reporting date. The Commission is also not adopting the commenter's alternative of omitting all pre-existing covered securities loans, as the data regarding such loans will be useful to investors as well as regulators and the on-boarding of data for such loans when modified will help reduce information asymmetries. These modifications to the final rule address the concerns of other commenters seeking certainty regarding such day one loans.

The commenter also recommended “provid[ing] all existing loans with an identifier on the day of implementation.” However, simply providing an identifier to a pre-existing loan without data about the loan would not provide transparency about the terms of the loan. Instead, the final rule requires that an RNSA assign a unique identifier to the pre-existing covered securities loan when it is reported to an RNSA.

Specifically, 17 CFR 240.10c–1a(g)(3) (“final Rule 10c–1a(g)(3)”) requires that an RNSA assign a unique identifier to a pre-existing covered securities loan that is reported to it pursuant to paragraph (d)(2). Consistent with covered securities loans, an RNSA is also required to make publicly available the data elements in paragraphs (c) and (d)(1)(i) reported to it, other than loan amount data element in paragraph (c)(6), not later than the morning of the business day after the covered securities loan is modified. In addition, an RNSA is required to make the loan amount publicly available on the twentieth business day after the pre-existing covered securities loan is modified. These requirements for pre-existing covered securities loans that are modified after the reporting date are necessary so that the data reported regarding these modified loans are also made publicly available in a manner similar to loans that are covered securities loans under paragraph (g) of the final rule.

3. Confidential Data Elements—Rule 10c–1a(e)

Proposed Rule

The Commission also proposed certain material transaction data elements to be provided to, and retained by, an RNSA, but not made public (“confidential data elements”). As proposed, the confidential data elements provision of the rule is designed to have certain data information used solely by regulators to better understand securities lending, including interest in short selling and price discovery for securities lending, yet without identifying market participants or revealing sensitive information about their internal operations to the rest of the market. Paragraph (d) of the proposed rule also required the confidential data elements to be provided to an RNSA within 15 minutes after each securities loan was effected (and then an RNSA would be required to keep such information confidential, subject to the provisions of applicable law).

As proposed, the confidential data elements included:

(i) The legal name of each party to the securities loan; where applicable, CRD or IARD Number, MPID, and the LEI of each party to the transaction, and whether such person is the lender/borrower/intermediary between the lender and the borrower (if known);

(ii) If the person lending securities is a broker or dealer and the borrower is its customer, to report whether the security is loaned from a broker's or dealer's securities inventory to a customer of such broker or dealer; and

(iii) If known, whether the securities loan is being used to close out a fail to deliver either pursuant to 17 CFR 242.204 (“Rule 204”) of Regulation SHO or outside of Regulation SHO.

Final Rule

The Commission sought and received comment regarding the proposed provision that would require lenders or their agents to provide the confidential data items to an RNSA within 15 minutes after each loan is effected. Commenters stated that the Commission should impose explicit confidentiality obligations on FINRA so as to protect the confidential securities lending information that would be reported. Paragraph (h)(4) of the final rule requires that an RNSA establish, maintain, and enforce reasonably designed written policies and procedures to maintain the security and confidentiality of the collected information, which should enhance protection of the confidential data that it collects under the final rule.

One commenter suggested that the final rule include masking or further additional redactions to better protect an individual's personal non-public information. Another commenter stated that a retail client has increased confidentiality concerns compared to an institutional client. These concerns appear to arise primarily from the potential identification of a broker or dealer's individual fully paid customers. Paragraph (e)(1) of the final rule specifies that a broker or dealer is not required to provide the legal name of (or otherwise make identifiable) the customer when reporting loans of fully paid excess margin securities pursuant to Rule 15c3–3(b)(3). Thus, a broker or dealer may redact or mask such information prior to submitting the information to an RNSA.

Another commenter supported the Commission's proposed requirement that lenders provide to an RNSA the identities of borrowers of securities loans (although they would not be publicly disclosed), explaining that this information is essential for effective market oversight, and failure to collect this information would materially weaken the efficacy of the rule. Under the final rule, the names and identities of lenders (other than customers in a fully paid lending arrangement) will be required to be reported to an RNSA. This data will be fully captured by the final rule requiring a broker or dealer (as the “borrower” under the “fully paid” exception) to report such loans. With regard to fully paid arrangements, the final rule requires the broker or dealer to report such loans as the borrower. Although brokers or dealers will not be required to report the customer names to an RNSA, the brokers or dealers will be required to retain such information and make it available to the Commission and its representatives and will be required to retain per RNSA rules.

Another commenter suggested that, to make the rule more useful to investors, in addition to loan characteristics, there should also be a requirement to publicly disclose the legal names of the parties to the loan (in contrast to keeping that information confidential). In fact, multiple retail investor commenters stated that real reform for securities lending must include notifying the public about who is borrowing and lending shares (not just which company's shares are being borrowed or lent). One commenter stated that there is no reason for any of the information that is proposed to be provided to an RNSA to be restricted from public view and, thus, it should be made public. However, making this information available to the public could be detrimental because it could identify specific market participants or reveal confidential information about the internal operations or investment decisions of specific market participants.

Others commented more generally on the importance of the securities lending disclosure requirements not adversely impacting the market ( e.g., by revealing investors' short selling strategies). To avoid such adverse impacts, these commenters stated that there should be more prescriptive measures (than requiring policies and procedures, for instance) to prevent an RNSA, or the reporting agent, from releasing any of the confidential transaction data that are provided, under the final rule. Paragraph (h)(4) of the final rule sets forth a multi-part requirement that mandates: (1) written policies and procedures; (2) that such policies and procedures must be reasonably designed; and (3) that an RNSA establishes, maintains, and enforces such policies and procedures. This approach is intended to provide flexibility to an RNSA to tailor policies and procedures that will appropriately maintain the security and confidentiality of the confidential information required by paragraph (e) of the final rule, based on its experience with the multiple types of sensitive and confidential trading data that it routinely handles as an RNSA. Policies and procedures generally are regularly monitored, tested, reviewed, and, if needed, updated to ensure that they remain continuously effective and, thus, should be particularly useful in helping to ensure that an RNSA adjusts to ever-changing technologies intended to circumvent data safeguards. Further, an RNSA's policies and procedures are subject to Commission oversight.

Some commenters expressed concerns about reporting agents maintaining confidential data received from covered persons. Consistent with the proposed rule, a covered person has the option of whether or not to entrust its confidential information with a reporting agent or to report such information directly to an RNSA. Further, paragraph (a)(2)(i) of the final rule requires that a covered person who relies on a reporting agent to fulfill its reporting obligations under the final rule must enter into a written agreement with such reporting agent. Such agreement could be used by the covered person and reporting agent to memorialize any measures that they agree to regarding the protection of the covered person's Rule 10c–1a information. Further, registered brokers and dealers and registered clearing agencies are subject to provisions such as section 15(g) of the Exchange Act and thus, have experience with implementing, maintaining, and enforcing policies and procedures and other means of protecting and preventing the misuse of information.

One commenter recommended that the proposed rule be modified to permit an RNSA to confidentially provide regulatory-related information to SROs that are not RNSAs, such as NYSE Regulation, without first obtaining a Commission order. Under the final rule, an RNSA is required to establish, maintain, and enforce policies and procedures specifically tailored to protect the confidential information enumerated in the final rule. To address commenters' concerns regarding the potential market impact of loan information required to be provided to an RNSA by the final rule, the Commission may designate by order upon a demonstrated regulatory need, dissemination to persons other than an RNSA. Thus, as discussed further below, in Part VII.K, it is appropriate to provide such information to persons (other than the Commission) by order, on a case-by-case basis, upon a demonstrated regulatory need. Such case-by-case determination strikes the proper balance between protecting confidential information and facilitating the regulatory function, such as those of an SRO. Another commenter requested that the Commission clarify whether the “if known” parenthetical, included at the end of paragraph (d)(1) of the proposed rule, was meant to apply to each of the confidential data elements ( i.e., CRD, IARD, MPID) in that paragraph. In reviewing this provision, the Commission recognizes that the proposed rule text, by locating the “if known” language at the end of paragraph (d)(1) in the proposed rule (rather than at the beginning as was done with other “if known” language that was used for the confidential data elements in paragraph (d)(3) of the proposed rule), the scope of this “if known” language with respect to the individual confidential data elements in paragraph (d)(1) was not entirely clear. Thus, the placement of the “if known” language similarly at the beginning of the paragraph (e)(1) of the final rule clarifies that each piece of information enumerated in that paragraph ( i.e., following the “if known” language) is required to be reported only to the extent that piece of information is known by the covered person. To clarify this intent, it is appropriate to modify proposed paragraph (d)(1) by moving the placement of the “if known” language that is currently proposed at the end of paragraph (d)(1) (in the proposed rule), and relocating it to the beginning of paragraph (e)(1) in the final rule, consistent with the same “if known” language at the beginning of 17 CFR 240.10c–1a(e)(3) (“final Rule 10c–1a(e)(3)”).

To the extent the data elements of final Rule 10c–1a(e)(1) are known, providing these data elements to an RNSA will allow regulators to understand buildups in risk at market participants and may otherwise assist an RNSA in monitoring and surveilling the security markets. The data elements will also provide an RNSA with the necessary information to administer the collection of all the other data elements provided to it under final Rules 10c–1a(c) and (d), such as ensuring the completeness of submissions, contacting persons that have errors in their provided data, and troubleshooting person-specific technical issues.

The Commission also proposed requiring that information be reported (but not publicly disclosed) on whether the securities loan is being used to close out a fail to deliver pursuant to Rule 204 of Regulation SHO or to close out a fail to deliver outside of Regulation SHO. Rule 204 of Regulation SHO generally requires closing a firm's fail to deliver position in equity securities at a registered clearing agency by purchasing or borrowing securities of like kind and quantity. Accordingly, under the final rule, this data element will provide regulators with information both about the loans used by brokers or dealers to close out fails to deliver as required by Regulation SHO, as well as additional insight into the use of loans, particularly the extent that loans are used to cover account-level fail to deliver positions in reportable securities.

Several commenters were supportive of this confidential data element, as proposed, with one commenter noting the benefits of such data for regulatory oversight, and yet other commenters requested that this data element be required to be publicly reported. Final Rule 10c–1a(e)(3), adopts this data element as proposed. The Commission has elected to adopt without modification ( i.e., not requiring public dissemination) this specific data element in paragraph (d)(3) of the proposed rule because, as a commenter stated, such information could be abused and is designed to be primarily useful for regulatory purposes. As the Commission has stated, fails to deliver may result from long sales as well as short sales, and may be closed out by either purchasing or borrowing shares. Fails to deliver can occur for various reasons, such as “human or mechanical errors or processing delays can result from transferring securities in custodial or other form rather than book-entry form, thereby causing a fail to deliver on a long sale.” Such data will be useful to regulators in determining both the extent of the use of securities loans for fails to deliver, as well as how and when fails to deliver are being addressed by securities loans. However, the Commission agrees with the commenter that such information, if made public, could be abused. For example, market participants that become aware of fails to deliver may seek to short squeeze the security knowing that the person failing to deliver has a time-sensitive need to cover its short position.

Another commenter stated that, in addition to the data elements in paragraph (d)(1) of the proposed rule, it will not have knowledge of whether the loan was used to cover a fail to deliver at the time of providing the confidential report to an RNSA, as would be required by paragraph (d)(3) of the proposed rule. The final rule adopts as proposed the requirement that such information must be reported if known.

As non-substantive changes to the rule, the Commission is replacing the word “person” with the defined term “covered person” as discussed above, in Part VII.A. Thus, as adopted, paragraph (e) of final Rule 10c–1a requires, consistent with the Proposing Release, that “[i]f required by paragraph (a) of this section, a covered person directly, or indirectly using a reporting agent shall provide the following information to an RNSA, if applicable, by the end of the day on which a covered securities loan is effected.” As non-substantive changes to the rule, the Commission is replacing the word “person” with the defined term “covered person” as discussed above, in Part VII.A. Thus, as adopted, paragraph (e) of final Rule 10c–1a requires, consistent with the Proposing Release, that “[a] covered person shall provide the following information to an RNSA, if applicable, by the end of the day on which a covered securities loan is effected.”

4. Removal of Securities Available to Loan Data Element

Proposed Rule

Among the securities lending data elements required to be reported to an RNSA, the Commission included in the proposed rule a requirement to report the total amount of each security that is not subject to legal or other restrictions that prevent it from being lent (“available to lend”). The Commission also required that certain identifying information, including the legal name of the security, and the ticker symbol or other identifier, be reported with the available to lend data. The Commission stated in the Proposing Release that it designed the securities available to loan data elements to allow for the calculation of a “utilization rate” for each security. The utilization rate, which could be calculated by dividing the total number of shares on loan by the total number of shares available for loan, could then be used by market participants to evaluate whether the security will be difficult or costly to borrow.

Final Rule

The Commission received numerous comments regarding the “available to lend” data element of the proposed rule. Certain commenters expressed concern that such data could be inaccurate, with some highlighting a variety of limitations on securities lending that could make the data unreliable. Commenters expressed concern that the available to lend data could be underreported if it omitted data from persons that were not required to provide information to an RNSA or that did not have an open securities loan. One commenter stated that “data would be particularly misleading for large international banking organizations with substantial operations outside of the United States.” Other commenters expressed general concern that the proposed “available to lend” data element could discourage securities lending. Two commenters recommended replacing the “available to lend” metric for the calculation of a utilization rate with publicly available information on the issuer's total share float. However, other commenters supported the reporting of “available to lend” data.

One of the Commission's goals in proposing the “available to lend” data requirements was “to allow for the calculation of a `utilization rate' for each particular security,” which in turn “could be used by market participants to evaluate whether the security will be difficult or costly to borrow.” The Commission stated in the Proposing Release that “the information provided under paragraph (e) [of the proposed rule] should allow market participants to calculate a utilization rate that is likely to be reliable.” However, many commenters expressed concerns that there may be significant challenges to the accurate reporting of “available to lend” data. One commenter stated, before highlighting barriers to obtaining accurate “available to lend” data, that the “[c]alculation of the utilization rate requires an accurate measure of the `securities available to lend' for all market participants.”

Upon consideration, the Commission agrees with commenters' concerns with the potential for underreported or inaccurate “available to lend” data being reported. Therefore, final Rule 10c–1a has been amended from the proposed rule to remove the reporting of “available to lend” data, as described under proposed Rules 10c–1(e)(1)(iii) and (e)(2)(iii), and to remove the corresponding requirement that an RNSA make such information publicly available, as described under proposed Rule 10c–1(e)(3). Additionally, final Rule 10c–1a removes the reporting requirements for the identifying information ( e.g., legal name of the security, LEI of the issuer, ticker symbol, ISIN, CUSIP, FIGI, or other identifier) of such “available to lend” securities, as listed under proposed Rules 10c–1(e)(1)(i) and (ii) and 10c–1(e)(2)(i) and (ii). Therefore, final Rule 10c–1a does not include a requirement that “available to lend” data be reported to an RNSA, by a lending agent, a person that does not employ a lending agent, or otherwise. The elimination of this requirement from the proposed rule responds to commenter concerns, will limit the burden of the final rule's reporting requirements on covered persons, and will preempt the potential for inaccurate or unreliable data to be made publicly available to market participants.

5. Removal of Securities On Loan Data Element

Proposed Rule

The Commission proposed a requirement that, by the end of each business day that a lending agent was required to provide Rule 10c–1 information to an RNSA or had an open securities loan about which it was required to provide information to an RNSA, the lending agent shall provide to an RNSA the total amount of each security on loan that has been contractually booked and settled (“security on loan”). The proposed rule also required that by the end of each business day that a person that does not use a lending agent provides Rule 10c–1 information to an RNSA, or had an open securities loan about which it was required to provide information to an RNSA, the person shall also provide to an RNSA the total amount of each security on loan that has been contractually booked and settled (“security on loan”). The Commission also proposed that “[f]or each security about which the RNSA receives [such] information . . . . The RNSA shall make available to the public only aggregated information for that security.”

The Commission stated its preliminary belief that securities on loan data could help market participants plan their borrowing activity. The Commission also stated that it designed the proposed securities on loan data element to allow for the calculation of a “utilization rate” for each particular security. The utilization rate could then be used by market participants to evaluate whether the security will be difficult or costly to borrow.

Final Rule

Commenters provided differing views on the proposed reporting requirement for the securities on loan data element. Certain commenters supported or did not object to the requirement to report securities on loan data in the proposed rule. One commenter recommended that more granular securities on loan data be made available, including that the “total number of shares of each stock on loan [should be] disaggregated by lender.” Multiple commenters supported the proposed requirement for an RNSA to make securities on loan data publicly available on an aggregate basis, with one recommending that “[f]or each security, FINRA could disseminate the total volume of securities on loan by shares or principal value (as applicable) and as a percentage of the shares or principal value (as applicable) of all securities that are outstanding.” Another commenter disapproved of the proposed securities on loan data reporting requirement, expressing concerns that the data “may be misleading,” and recommended removing it. The commenter stated that some brokers use approaches to calculating securities on loan data that “routinely misrepresent the true number of borrowed shares.”

Similar to the concerns described above, in Part VII.F.4, about the proposed securities available to loan data element, the Commission agrees with the commenters that there may be challenges to the collection of accurate securities on loan data. Additionally, aggregated information similar to the proposed securities on loan data will be made available by RNSAs, without the additional, potentially duplicative, reporting requirements of proposed Rule 10c–1(e)(1)(iv) and (e)(2)(iv). Under the final rule, RNSAs will be required to design and provide not only information pertaining to the aggregate transaction activity for each reportable security, but also distribution of loan rates for each reportable security. These requirements should provide market participants with information to help plan their borrowing activity, while avoiding potentially duplicative reporting obligations. Such aggregated data would be consistent with some commenters' support for requiring RNSAs to make “securities on loan” data available on an aggregated basis. Additionally, RNSAs will be required to make the disaggregated “amount of the security loan” data publicly available, comparable to the disaggregated “securities on loan” data recommended by one commenter. However, as discussed below, in Part VII.J, such disaggregated data will only be publicly available on a delayed basis, on the twentieth business day after the covered securities loan is effected, in order to protect the sensitive nature of individual securities lending “amount” information ( e.g., such as size or volume). Furthermore, for the reasons described above, in Part VII.F.4, final Rule 10c–1a will not require the reporting of an available to loan data element, which therefore would not be available to pair with the proposed securities on loan data element in order to calculate a “utilization rate” metric. Therefore, while certain benefits derived by market participants from the reporting and disclosing of securities available to lend data and securities on loan data elements together may not be realized, market participants may benefit from the public availability of aggregate transaction activity and distribution of loan rates.

For the foregoing reasons, final Rule 10c–1a has been changed from the proposed rule to remove the reporting of securities on loan data, as described under proposed Rule 10c–1(e)(1)(iv) and (e)(2)(iv), and to remove the corresponding requirement that an RNSA make such information publicly available, as described under proposed Rule 10c–1(e)(3). Therefore, final Rule 10c–1a does not include a requirement that a lending agent or other covered person report securities on loan data to an RNSA.

G. Timing of Required Reporting to an RNSA

1. Timing of Reporting of Loans

Proposed Rule

As discussed above, in Part VII.F.1, paragraph (b) of the proposed rule would have required certain loan transaction data elements to be reported to an RNSA, on a transaction-by-transaction basis, within 15 minutes of the loan being effected, followed by an RNSA assigning each loan a unique transaction identifier and then making such information publicly available as soon as practicable. As part of the proposed rule, the Commission sought specific comment as to the proposed 15-minute requirement for reporting specified data elements to an RNSA.

Final Rule

While the proposed 15-minute reporting requirement received some support, most of the comments received by the Commission from larger institutional market participants strongly opposed the proposed rule's requirement that the specified data elements be reported to an RNSA within 15 minutes after a loan is effected ( i.e., in addition to the data being publicly disseminated on a transaction-by-transaction basis). For example, most of these larger market participants explained that the terms of securities loans change during the day and are generally not finalized until the end of the day. These changes include reallocations of securities loans among lenders, re-pricings, and changes in collateral. As a result, the commenters stated that requiring transaction-by-transaction reporting, particularly on a 15-minute/intraday basis, would result in significant unintended negative consequences, including the public dissemination of incomplete or misleading information, which could adversely impact the securities/lending markets. Another commenter also stated that the proposed 15-minute reporting requirement is impractical and logistically challenging and would “create noise and misleading information in the market.” In addition commenters stated that there would be costs to participants to create and maintain an entirely new infrastructure for loan data reporting and dissemination.

Other commenters stated that the proposed 15-minute reporting time period would exponentially increase the number of execution (and modification) reports that would be required to be filed. Another commenter stated that requiring reporting every 15 minutes gives away too much proprietary information to the market regarding closely guarded trading strategies, risking exposures to short squeezes, front running, reverse-engineering—particularly with hard-to-borrow securities. Another commenter raised the concern that a 15-minute reporting requirement may be unnecessarily frequent and that there did not appear to be any stated rationale in the Proposing Release for how a 15-minute reporting interval would be helpful to market participants or why this frequency is appropriate.

To address these concerns, commenters offered possible modifications for the final rule. Many of the commenters who suggested an alternative approach favored an “end-of-day” reporting requirement. They explained that end-of-day reporting would make more sense and result in more worthwhile information being reported as it takes into account that most securities loan trades are not finalized until end of day. One of the commenters explained how most securities loan market participants already use/rely on “end-of-day” data because they view it as the most relevant and reliable measurement of market activity. This commenter also stated that, even though intraday data may already be available from service providers, it is still widely seen as indicative in nature and subject to frequent correction.

Other commenters suggested reporting by end of the next day ( i.e., on a T+1 basis similar to what the EU/UK's Securities Financing Transactions Regulation (“SFTR”) reporting regime requires). For example, one commenter stated it did not believe there is any additional value to investors in providing intraday data, given the nature of pricing for securities lending transactions, the potential for intraday data to confuse investors, and the unnecessary costs and burdens it would pose to market participants. Another commenter who favored next day reporting maintained that it was a way to avoid the operational challenges, disproportionate costs, and compliance complexities associated with the proposed 15-minute reporting requirement. Another commenter asked the Commission to amend the 15-minute reporting requirement in favor of end-of-day reporting on a next day (T+1) basis and stated that the 15-minute reporting requirement would be impractical, and that a T+1 standard would address transparency, reduce implementation costs, and align with the existing SFTR securities loan reporting regime. Another commenter suggested that the proposed rule be modified to allow the lending agent to have until the following business day to make the required report to avoid reporting errors.

Commenters also stated that next day or T+1 reporting would provide more appropriate flexibility for developing systems and processes for reporting at substantially lower cost. Some commenters expressed support for modifying the reporting requirement to require next business day ( i.e., on T+1 basis) but also for an earlier reporting requirement ( i.e., end of the same day) provided that the Commission provides clarity regarding the precise time period for such reporting ( e.g., the exact point in time that will constitute “end of day” for purposes of the final rule) so that market participants will have the clarity and certainty they will need to comply and report accurately.

After reviewing and considering the public comments and recommendations regarding the proposed 15-minute reporting requirement, the Commission is adopting final Rule 10c–1a(c), substantially as proposed, but with targeted modifications to the proposed rule's reporting period to address commenters' concerns about the proposed timing requirement being unworkable and overly burdensome by replacing the proposed 15-minute reporting period with end-of-day reporting, as well as bringing the different regulatory regimes in closer alignment with the final rule's public dissemination requirements. Adopting the final rule with an end-of day reporting requirement will help reduce the concerns raised by the commenters, while still furthering the underlying objectives of the final rule.

As modified, the final rule's end-of-day reporting requirement will help prevent an excessive number of incomplete or slightly modified reports that otherwise would occur throughout the day yet without providing any incremental value. Thus, in modifying the final rule to include an end-of-day reporting requirement, rather than requiring frequent intraday reporting, the Commission understands the frequency with which parties to a securities loan may agree to some of the basic terms initially, but that some or many of the securities loan terms may not be agreed to (or may be updated throughout the day and, thus, not finalized) until the end of the day. Nonetheless, whether or not a loan has been effected is a legal/factual question and a delay in settlement (or if one of the agreed to loan terms is modified the next day) does not impact the initial requirement to report all loans (and modifications) within the required timeframes under the final rule.

As adopted, paragraph (c) of the final rule requires a covered person (directly, or indirectly using a reporting agent) to provide the Rule 10c–1a information, if applicable, to an RNSA “by the end of the day on which a covered securities loan is effected.” By replacing the proposed 15-minute reporting requirement with a more flexible end-of-day (or “EOD”) reporting requirement in final Rule 10c–1a(c), the Commission recognizes that many covered securities loans are not likely to be finalized within that 15-minute time period or until the end of the day. Unlike SFTR in Europe, final Rule 10c–1a's reporting requirement ( i.e., to an RNSA) will not extend until the next day, on a T+1 basis. While final Rule 10c–1a and the SFTR both deal with securities lending, they are two unique and different regulatory regimes, with differences in their underlying objectives and scope of regulations. Final Rule 10c–1a, however, does allow an RNSA until the morning of the next business day (thus on T+1, the same as SFTR) to disseminate the Rule 10c–1a information it receives the night before. This allows an RNSA sufficient time it needs between receipt of the Rule 10c–1a information and its dissemination to the public, which necessitates the Rule 10c–1a information being collected no later than the night before.

2. Timing of Reporting of Loan Modification

Proposed Rule

As discussed above, in Part VII.F.2, paragraph (c) of the proposed rule would have required that certain loan modification data elements be provided to an RNSA within 15 minutes after each loan is modified if the modification results in a change to the Rule 10c–1 information that is already required to be provided to an RNSA under paragraph (b) of the proposed rule (and for an RNSA to make such information available to the public as soon as practicable). In issuing the proposed rule, the Commission sought comment specifically regarding the proposed timing requirement in paragraph (c), which would have required specified loan modification data elements to be reported to an RNSA within 15 minutes after a loan is modified.

Final Rule

Most of the comments received by the Commission on the proposed loan modification data elements in paragraph (c) of the proposed rule were focused on the proposed 15-minute loan modification reporting requirement, with many of the commenters suggesting that it be replaced with end-of-day reporting ( i.e., similar to the end-of-day timing modification discussed above with respect to the securities loan data elements information originally reported to an RNSA).

In addition to the comments discussed above, in Part VII.F.2, commenters opposing the proposed 15-minute reporting requirement also raised concerns that new trades can be executed for market delivery at any time during the day before the close of settlement at DTC and, as a result, that some market participants will book trades in “batches” or at their discretion, as opposed to individually when agreed. Some commenters stated that other reporting regimes do not require intraday reporting, as certain activity reported intraday might ultimately not result in an executed trade.

Reporting the loan modification data elements for each covered securities loan that is modified is important to ensure that data elements previously reported to an RNSA accurately reflect currently outstanding covered securities loans and to prevent evasion of the rule. However, in response to commenters' concerns, particularly with respect to operational difficulties associated with the proposed 15-minute loan modification reporting requirement and uncertainties as to the scope of its intended application, the Commission is adopting the loan modification data elements provision in paragraph (d) of the final rule, substantially as proposed, but with a modification as to the timing of the reporting period and some clarifying changes to the proposed rule text.

More specifically, to promote the integrity and consistency of the reporting under the final rule, the Commission has determined to remove the proposed requirement to report loan modification data elements to an RNSA within 15 minutes after each loan is modified and to replace it with the same end-of-day reporting the Commission is requiring for the original loan data elements under paragraph (c) of the final rule. For the same reasons discussed above, in Part VII.F.1, allowing covered persons until the end of the day to report any required loan modification data information to an RNSA is appropriate because it will help to address commenters' concerns and operational difficulties with frequent intraday reporting while still furthering the transparency objectives of the final rule.

As modified, the final rule requires the specified loan modification data elements to be provided by the covered person (directly or indirectly using a reporting agent) to an RNSA, by the end of the day on which a covered securities loan is modified. By replacing the proposed 15-minute reporting requirement with a more flexible end-of-day reporting requirement in final Rule 10c–1a(d), the timing modification should help reduce the concerns raised by commenters, that is, by allowing covered persons and reporting agents additional time (until the end of each day, rather than just within 15 minutes), to report any required loan modification information to an RNSA and also to help reduce the overall implementation cost for market participants, and allow such participants to better manage the flow of data in line with the existing internal processes for reporting.

3. Timing of Reporting of Confidential Data Elements

Proposed Rule

As discussed above, in Part VII.F.3, the Commission also proposed that certain confidential data elements would be provided to, and retained by an RNSA, but not made publicly available by an RNSA. Such confidential data was intended to provide regulators with specific information about the loan (such as the identity of the parties to the loan) that would be kept confidential due to concerns about potential misuse of such information. The Commission proposed that such information be reported to an RNSA within 15 minutes after the loan is effected.

Final Rule

The Commission is modifying the proposed timing requirement for reporting the confidential data elements, as adopted in paragraph (e) of the final rule, similar to the timing revisions the Commission made with respect to the proposed loan and loan modification data elements adopted in paragraphs (c) and (d) of the final rule ( i.e., by replacing the proposed 15-minute reporting requirement with the similar end-of-day reporting period). As modified, and otherwise consistent with the rule as proposed, paragraph (e) of the final rule will continue to require that any confidential data elements be reported to an RNSA; however, an RNSA will be required to keep such information confidential, “in accordance with the provisions of paragraph (h) of this section and applicable law.”

H. Definition of Registered National Securities Association—10c–1a(j)(5)

Proposed Rule

The Commission proposed that any person that loans a security on behalf of itself or another person shall provide to an RNSA certain specified information. However, the Commission did not include a definition of the term “RNSA.” The Proposing Release stated that the only existing RNSA has experience establishing and maintaining systems that are designed to capture transaction reporting, similar to the requirements of final Rule 10c–1a.

Final Rule

The Commission received comments stating that FINRA, as the only existing RNSA, is best positioned for such a role. To provide additional clarity regarding the regulatory body that Rule 10c–1a information must be reported to, the final rule defines the term “RNSA” to mean “an association of brokers and dealers that is registered as a national securities association pursuant to 15 U.S.C. 78 o –3 (`section 15A') of the Exchange Act.” This definition applies to any association of brokers and dealers that is registered as a national securities association pursuant to section 15A of the Exchange Act now or in the future.

I. RNSA Rules To Administer the Collection of Information—Rule 10c–1a(f)

Proposed Rule

The Commission proposed that “[t]he RNSA shall implement rules regarding the format and manner to administer the collection of information in paragraphs (b) through (d) of [the proposed rule] and distribute such information in accordance with the rules approved by the Commission pursuant of section 19(b) of the Exchange Act and Rule 19b–4 thereunder.” In the Proposing Release the Commission stated its preliminary belief that permitting an RNSA to implement rules regarding the administration of the collection of securities lending transactions would enable an RNSA to maintain and adapt potential technological specifications and any changes that might occur in the future. The Commission also affirmed that it would retain oversight of an RNSA's adoption of rules to administer the collection of information under proposed Rule 10c–1.

Final Rule

The Commission asked in the Proposing Release whether Rule 10c–1 should require that lenders provide material information to an entity other than an RNSA. One commenter stated that FINRA, the only existing RNSA, is best positioned to serve in such a role due to its expertise and experience in administering trade reporting rules. Other commenters also supported the role of RNSAs in collecting and distributing information under the proposed rule, with one specifically stating that the format and manner through which information will be provided to an RNSA should be defined by an RNSA and should not be specified in proposed Rule 10c–1. The Commission agrees with the commenters that an RNSA is well positioned to define, consistent with section 19(b) and Rule 19b–4 of the Exchange Act as well as an RNSA's own internal business requirements and systems designs, the format and manner in which it collects Rule 10c–1a information.

Alternatively, another commenter recommended that the Commission should eventually transition the collection and dissemination of the data collected under proposed Rule 10c–1 to an internal process at the Commission to enhance enforcement of the proposed rule and allow the Commission to share relevant data. However, in addition to providing securities lending information to the public, the final rule is intended to provide additional tools to RNSAs to enhance RNSAs' surveillance over the securities markets. An RNSA is appropriately suited for collecting and analyzing such data for such surveillance purposes.

Some commenters expressed concerns or provided recommendations for an RNSA's administration of information collected under the final rule. One commenter stated the importance of RNSAs' compliance with section 19(b) and Rule 19b–4 of the Exchange Act when implementing RNSA rules required by final Rule 10c–1a. Consistent with the proposed rule, any proposed changes to an RNSA's rules required by final Rule 10c–1a, including its Rule 10c–1a information collection and dissemination practices, will also be subject to notice, public comment, and Commission review pursuant to section 19(b) and Rule 19b–4 prior to implementation. This requirement, including the notice and opportunity for public comment, will help the Commission ensure that the direct reporting to an RNSA by covered persons is logistically feasible, prior to the effectiveness and implementation of such RNSA rule, and that its methodology helps ensure the accuracy and quality of the reported data.

Another commenter expressed concern that the proposed rule “would provide no mechanism for quality assurance.” An RNSA could choose to include mechanisms for quality assurance in its rules to implement the system under the final rule. Further, all RNSA rules implementing the system under the final rule will be subject to notice, public comment, and Commission review under section 19(b) and Rule 19b–4.

Another commenter stated that the Commission “should ensure that direct reporting to the RNSA is logistically feasible for entities that are not broker-dealers.” The final rule permits such entities to enter into agreements with reporting agents that are brokers, dealers, or registered clearing agencies to fulfill their reporting obligations on their behalf, should they determine that direct reporting is not the most appropriate choice for them.

One commenter recommended that the collection, maintenance, and publication of securities lending data, which an RNSA was required to perform under the proposed rule, be subject to a competitive bidding process to select a technology vendor or vendors to provide such services. However, the oversight that the Commission maintains over RNSAs, including oversight of their rules to administer the collection of Rule 10c–1a information pursuant to section 19(b) and Rule 19b–4, would not apply to the more general population of “technology vendors” that the commenter proposes should participate in a competitive bidding process to collect, maintain, and distribute Rule 10c–1a information. The commenter's proposed approach would not be appropriate in the absence of such oversight. The Commission agrees with one commenter's statement that the existing RNSA is well positioned to serve in this role under final Rule 10c–1a given its expertise in administering other trade reporting systems.

Having considered commenters' submissions, for the foregoing reasons, the Commission is adopting the requirement that an RNSA shall implement rules regarding the format and manner to administer the collection and dissemination of certain information, substantially as proposed. The final rule includes minor changes to reflect that the relevant Rule 10c–1a information paragraphs have been renumbered to align with the format of the final rule, to provide citations to the statutory provision and regulation that such rules must be promulgated pursuant to, the term “distribute such information” has been modified to “make publicly available such information” to more accurately reflect an RNSA's responsibilities for the publication of data under final Rule 10c–1a(g), and the term “approved by the Commission” has been modified to “promulgated” to more accurately reflect the process by which RNSA rules are implemented pursuant to section 19(b) and Rule 19b–4 of the Exchange Act.

J. RNSA Publication of Data—10c–1a(g)

Proposed Rule

The proposed rule would have required an RNSA to make available to the public the information required by proposed paragraph (b), including the legal name of the security issuer, and the LEI of the issuer, if the issuer has an active LEI, and the ticker symbol, ISIN, CUSIP, or FIGI of the security, if assigned, or other identifier. In addition, modifications that resulted in a change to the data elements required to be provided to an RNSA under paragraph (b) of the proposed rule would also have been made public by an RNSA. An RNSA would have also provided identifying information for each security for which aggregate information would be made public. Paragraph (e)(3) of the proposed rule further required that an RNSA keep identifying information about lending agents, reporting agents, and other persons using reporting agents confidential, subject to applicable law. In addition, the confidential data elements in proposed Rules 10c–1(d)(1) through (3) would be kept confidential so as to not identify market participants or reveal information about the internal operations of market participants. The proposed rule also required an RNSA to assign a unique transaction identifier to each loan.

Final Rule

The Commission received many comments that expressed support for increasing transparency and price discovery in the securities lending market by increasing the amount and availability of data to the public. Commenters were generally supportive of the requirement that an RNSA be responsible for making reportable data publicly available.

The proposed rule combined reporting requirements with RNSA publication requirements, which resulted in commenters addressing these respective requirements in conjunction with one another. The Commission did not intend to conflate the requirements. The final rule separates an RNSA's publication responsibilities from a covered person's reporting requirements to improve the structure of the final rule. Therefore, final Rule 10c–1a is modified to separate the requirements of an RNSA from the reporting requirements applicable to covered persons and reporting agents. Final Rule 10c–1a now places an RNSA's publication of data elements into new paragraph (g).

In response to the comments received that the disclosure of reported information by an RNSA on a transaction-by-transaction basis could increase the risk of revealing short sale strategies, final Rule 10c–1a bifurcates paragraph (g) requirements such that certain data elements will be made publicly available on a transaction-by-transaction basis and aggregate transaction activity and distribution of loan rates) will also be made publicly available in an aggregated format for each reportable security. With respect to aggregated data, the final rule includes a new requirement for an RNSA to make “information pertaining to the aggregate transaction activity and distribution of loan rates for each reportable security and the security identifier(s) under paragraphs (c)(1) or (2) of the final rule for which an RNSA determines is appropriate to identify” publicly available “as soon as practicable, and not later than the morning of the business day after the covered securities loan is effected or modified.” The term “aggregate transaction activity,” as used in the final rule, refers to information pertaining to the absolute value of transactions such that net position changes should not be discernable in the data, and is intended to help ensure that only aggregate information about net positions changes, rather than individualized information, is provided to the public. The addition of the term “aggregate transaction activity” responds to commenters' concerns about the potential exposure of proprietary information, while still providing volume transparency to market participants. Providing information about the distribution of loan rates for each security recognizes that the cost-to-borrow for loans of securities is influenced by a number of factors ( e.g., counterparty-creditworthiness) and, thus, information about loan rates on a transaction-by-transaction basis may not facilitate a perfect comparison of such rates between loans of the same security. Consequently, knowing the distribution of loan rates for a given security can give market participants information to help market participants compare the pricing of their covered securities loan against the pricing of other covered securities loans. This can facilitate conversations between beneficial owners and their lending agents or end borrowers with their brokers or dealers regarding the terms of their loan.

Additionally, the final rule delays an RNSA's publication of loan amount, on a transaction-by-transaction basis in paragraph (c)(6), to the twentieth business day after the covered securities loan is effected along with the loan and security identifying information specified in paragraphs (g)(1)(i)(A) and (C) of this section. The final rule also delays an RNSA's publication of a modification to loan amount, on a transaction-by-transaction basis, in paragraph (c)(6), to the twentieth business day after the covered securities loan is modified with the loan and security identifying information specified in paragraphs (g)(1)(i)(A) and (C).

For purposes of compliance with the final rule, the countdown to the “twentieth business day” starts the day after the covered securities loan is effected. For example, if a covered securities loan were effected at 4:00 p.m. on a Wednesday, and the applicable Rule 10c–1a information is received by an RNSA by the end of that day, that Thursday after the Wednesday on which the covered securities loan is effected will start the 20-business day period, assuming that Thursday is not a holiday.

The final rule contains requirements for an RNSA to make data elements other than loan amount publicly available as soon as practicable, and not later than the morning of the business day after the covered securities loan is effected. Similarly, the final rule requires an RNSA to make loan modifications, excluding modifications to loan amount, publicly available, as soon as practicable, and not later than the morning of the business day after the covered securities loan is modified. The final rule also requires that an RNSA make the data elements in paragraph (c) of the final rule, other than loan amount, publicly available as soon as practicable, and not later than the morning of the business day after the covered securities loan is effected, if a securities loan is modified when such covered securities loan is a pre-existing covered securities loan ( i.e., a covered securities loan for which reporting under paragraph (a) was not required on the date the loan was agreed to or last modified). This requirement is designed to provide transparency for covered securities loans that have not been previously reported to an RNSA prior to having a data element of paragraph (c) modified. As discussed above, in Part VII.F.2, the final rule also requires that an RNSA make publicly available the data elements in paragraph (c) of the final rule for pre-existing covered securities loans. The final rule also contains a requirement for an RNSA to keep confidential data elements confidential, in accordance with subparagraph (h) and consistent with applicable law. Final Rule 10c–1a also contains updated terminology from the proposed Rule 10c–1, including that the “unique transaction identifier” is updated to “unique identifier;” and references to “loan” are updated to “covered securities loan.” Revisions to final Rule 10c–1a(g) are updated with the relevant paragraphs renumbered to align with the format of the final rule.

Final Rule 10c–1a requires an RNSA to assign each covered securities loan a unique identifier in paragraphs (g)(1)(i)(A) and (g)(3). The final rule makes a revision to terminology by replacing the term “unique transaction identifier” assigned by an RNSA to the original loan, as used in the proposed rule, with “unique identifier” assigned by an RNSA to the original covered securities loan. The removal of “transaction” is to avoid confusion with the defined term “covered securities loan,” which is the accurate description of what the “unique identifier” represents.

While the final rule continues to require an RNSA to “assign each covered securities loan a unique identifier,” a general theme received from commenters was about the operational efficiency of allowing market participants to create and submit their own unique identifiers. In response to the Proposing Release, one commenter stated that “beneficial owners or those reporting on their behalf should have flexibility to generate their own Transaction Identifiers provided they meet minimum standards for integrity and identification” and that “this Transaction Identifier [be used] in connection with any loan modification reporting in connection with the relevant loan.” Another commenter recommended that “allowance is made for the UTI to be provided to the RNSA by the reporting party, with an on-receipt issuance model only available should firms have no existing UTI for a reportable transaction.” Another commenter agreed that firms should be allowed to internally assign UTIs and report them to an RNSA, but alternatively suggested that if the UTI is required to be assigned by an RNSA then an RNSA should return this internal identifier to the reporting firm along with an RNSA-assigned UTI in order to link a modification to the original reported loan. This commenter also recommended that, under either approach, UTIs should be different in the U.S. and other jurisdictions, as applicable. Another commenter stated that under the final rule, if there was a loan that was reported to both an RNSA and the SFTR, the loan would have an identifier for each system and prevent global aggregation.

The Commission understands that for many reporting entities, obtaining a unique identifier from an RNSA on a post-trade basis and then tagging it to the trade for accurate reporting may add complication and expense and introduces operational risk for post-trade errors in aligning trades with transaction identifiers (particularly if a trade has been modified during the course of intraday trade reporting). According to one commenter, “the objective of a Transaction Identifier is to uniquely identify a particular transaction; so long as that objective is satisfied, the SEC should be neutral as to the manner in which the Transaction Identifier is produced.” Doing so, the commenter suggests, “would [provide] the flexibility to an RNSA to allow reporting parties to provide their own Transaction Identifiers [which] would likely materially reduce costs for any such parties.” According to the commenter, this approach has been successfully deployed in Europe, under the SFTR, and the U.S. in connection with security-based swap reporting rules. Accordingly, this commenter urged the Commission to permit the reporting agent to generate the transaction identifiers prior to reporting to an RNSA, with an RNSA available to provide Transaction Identifiers as a back-up for those reporting parties who are unable, for whatever reason, to generate their own transaction identifiers. Another commenter stated that while the data elements required to be reported are sufficient to allow for an RNSA to identify loans and to create a unique transaction identifier, the Commission should specify how such identifier will be shared with the reporter.

Having a unique identifier assigned to each covered securities loan is necessary for an RNSA to easily track the loan and facilitate the identification and reporting of any subsequent modifications to a particular covered securities loan. The Commission agrees with the comment that the objective is to have a unique identifier for tracking and that the Commission should be neutral as to the manner in which the unique identifier is produced, and that it is appropriate to allow administrative details of the process to be left to the discretion of an RNSA, rather than dictated by the Commission. It is appropriate to require an RNSA to assign a unique identifier to the loan, even if such loan already has an identifier that is reported to the SFTR, for consistency with an RNSA's rules and systems. Further, the final rule requires an RNSA to create a system capable of generating unique identifiers, but does not preclude market participants from creating their own unique identifiers when submitting information to an RNSA.

The Commission agrees that FINRA, as the sole existing RNSA, has the experience and controls to implement an appropriate system for market participants, including, if appropriate, how an identifier will be shared with a reporter. Structuring the final rule to provide an RNSA with flexibility to accept unique identifiers and/or use an RNSA assigned unique identifier to publish data is operationally practical. Accordingly, final Rule 10c–1a(g) maintains that an RNSA shall assign a unique identifier to the covered securities loan and make publicly available as soon as practicable, and not later than the morning of the business day after the covered securities loan is effected.

K. RNSA Data Retention, Availability, Fees, and Security

1. Data Retention—Rule 10c–1a(h)(1)

Proposed Rule

Under the proposed rule the Commission required that RNSAs retain the information collected pursuant to paragraphs (b) through (e) of the proposed rule in a convenient and usable standard electronic data format that is machine readable and text searchable without any manual intervention for a period of five years. In the Proposing Release, the Commission stated that requiring an RNSA to retain records for five years was consistent with other retention obligations of records that Exchange Act rules impose on an RNSA. As examples, the Commission identified Rule 17a–1 and 17 CFR 242.613(e)(8) (“Rule 613(e)(8)”) of Regulation NMS, both of which require RNSAs to keep documents for a period of not less than five years.

Final Rule

The Commission received limited comments concerning this element of the proposed rule. One commenter stated that a retention period of six years plus the current year would simplify GDPR and UK tax compliance. The Commission recognizes the importance of compliance with other regulatory regimes; however, requiring RNSAs to retain records for not less than five years is consistent with other record retention obligations that Exchange Act rules impose on RNSAs. Including a retention period that is consistent with other rules applicable to RNSAs could reduce the burden for an RNSA to comply with the retention requirements in proposed Rule 10c–1 because RNSAs will have developed experience and controls around administering similar record retention programs. Therefore, the Commission is adopting RNSA data retention requirements as proposed ( i.e., in a convenient and usable standard electronic data format that is machine readable and text searchable without any manual intervention for a period of five years), with the relevant paragraphs renumbered to align with the format of the final rule.

2. Data Availability to the Public—Rule 10c–1a(h)(3)

Proposed Rule

The Commission proposed that RNSAs make certain data available to the public in a convenient and usable standard electronic data format on its website or similar means of electronic distribution, without charge and without use restrictions, for at least a five-year period. The Commission stated its preliminary belief that requiring an RNSA to provide certain information to the public would further the direction by Congress in section 984(b) of the Dodd-Frank Act for the Commission to promulgate rules that are designed to increase the transparency of information to brokers-dealers and investors, with respect to the loan or borrowing of securities. The Commission acknowledged that establishing and maintaining a system to provide public access to Rule 10c–1 information is not without cost, and expressed the preliminary belief that such costs should be borne by an RNSA in the first instance and be permitted to be recouped by an RNSA from market participants who report securities lending transactions to an RNSA. The Commission also stated its belief that any restrictions on how the publicly available Rule 10c–1 information is used could impede the utility of such information and limit the ability of investors, commercial vendors, and other third parties, such as academics, from developing uses and analyses of the information.

Final Rule

It is appropriate to remove the phrase “without charge” from an RNSA's availability of information requirements in final Rule 10c–1a(h)(3). Some commenters supported the Commission making certain securities lending information available without charge, as proposed. One commenter stated generally that it would be extremely helpful to have freely available data. The Commission agrees that having access to securities lending data without charge will benefit market participants. As discussed below, in Part VII.K.3, the Commission received comment recommending that RNSAs be permitted to charge fees to entities other than lending agents and beneficial owners. After considering commenter input, the Commission is removing the “without charge” requirement from the final rule to provide RNSAs with greater flexibility to structure reasonable fees. As discussed below, in Part VII.K.3, and as proposed, any such fees will have to be filed with the Commission pursuant to section 19(b) of the Exchange Act and Rule 19b–4, and would be published for notice and public comment. Additionally, consistent with the Proposing Release, and the provisions that govern RNSAs under the Exchange Act, the final rule requires that RNSAs may only establish and collect fees that are reasonable. Furthermore, as proposed, the final rule prohibits use restrictions on the publicly available information. Any restrictions on how the publicly available Rule 10c–1a information is used could impede its utility and limit the ability of investors, commercial vendors, and other third parties, such as academics, from developing uses and analyses of the information.

The Proposing Release included a request for comment asking if a period of five years is the appropriate length of time for an RNSA to make information available to the public. One commenter responded, stating that, “transparency is a good thing, and lessons can be learned from history, `in perpetuity' would be the ideal period of time.” The Commission acknowledges the value of historical studies of securities markets, and the final rule does not preclude an RNSA from making certain reported data public in perpetuity. However, requiring that RNSAs make certain information publicly available indefinitely could prove unduly burdensome on RNSAs when compared with the utility of old and potentially less informative data. Furthermore, making information available in a convenient and usable standard electronic data format indefinitely would necessitate that RNSAs retain the data indefinitely, which would be inconsistent with the final rule's data retention requirements and other retention obligations of records that Exchange Act rules impose on RNSAs. Five years is the appropriate length of time for an RNSA to be required to make information available to the public, because such a time period will provide broker-dealers and investors with an opportunity to identify trends occurring in the market and in individual securities based on changes to the material terms of securities lending transactions.

For the foregoing reasons the Commission is adopting the collected information availability requirements for RNSAs, with respect to making certain collected information available to the public, substantially as proposed, with the phrase “without charge” removed and the relevant paragraphs renumbered to align with the format of the final rule.

3. RNSA Fees—Rule 10c–1a(i)

Proposed Rule

The Commission proposed that RNSAs may establish and collect reasonable fees, pursuant to rules that are effective pursuant to section 19(b) of the Exchange Act and Rule 19b–4 thereunder, from each person who provides any data set forth in paragraphs (b) through (e) of this section directly to an RNSA. Under the Exchange Act, RNSAs are allowed to adopt rules that impose the equitable allocation of reasonable fees among members and issuers, and other persons that use an RNSA facility or system. In the Proposing Release the Commission stated its preliminary belief that it would be appropriate to establish and collect reasonable fees from each person who directly provides the information set forth in the rule to an RNSA.

Final Rule

One commenter supported an RNSA acting in accordance with section 19(b) of the Exchange Act and Rule 19b–4 thereunder when implementing rules regarding fees. Another commenter recommended that reporting costs should not only be borne by persons who provide data directly to an RNSA and recommended that “costs incurred by FINRA . . . be shared among all those that benefit from securities lending activity, not solely on lending agents and beneficial owner clients.” Similarly, another commenter recommended “that the costs incurred by the RNSA to establish and operate the reporting system for securities lending data should be equitably shared by both borrowers and lenders, along with a tiered fee structure that eliminates costs for General Collateral (“GC”) transactions.”

The Commission recognizes that additional flexibility may be warranted in how costs incurred by an RNSA are recouped. To provide more flexibility in the establishment and collection of reasonable fees by RNSAs, the final rule has been amended from the proposed rule to remove the requirement that fees only be collected “from each person who provides any data set forth in paragraphs (b) through (e) of this section directly to the RNSA.” The final rule provides that an RNSA may “establish and collect reasonable fees, pursuant to rules that are promulgated pursuant to section 19(b) and Rule 19b–4 of the Exchange Act.” These changes from the proposed rule will provide RNSAs with greater flexibility to structure reasonable fees, including permitting the charging of fees to entities that use certain value-added services, as discussed above in Part VII.K.2, when accessing the information made publicly available by an RNSA.

4. Data Security—Rule 10c–1a(h)(4)

Proposed Rule

The Commission proposed a requirement that RNSAs establish, maintain, and enforce reasonably designed written policies and procedures to maintain the security and confidentiality of certain confidential information required by the proposed rule. In the Proposing Release the Commission stated its preliminary belief that an RNSA needs to protect Rule 10c–1 information from intentional or inadvertent disclosure to protect investors that provide such information by establishing reasonably designed written policies and procedures because the distribution of such information would identify market participants or could reveal information about the internal operations of market participants, which could be adverse to those providing information to an RNSA.

Final Rule

For the reasons discussed above, in Part VII.F.3, the Commission is adopting the final rule as proposed and requiring an RNSA to establish, maintain, and enforce written policies and procedures to maintain the confidentiality of the confidential data elements collected under the final rule.

L. Data Availability to the Commission and Other Persons—Rule 10c–1a(h)(2)

Proposed Rule

The Commission proposed that RNSAs be required to make certain information, including confidential information, collected pursuant to the proposed rule available to the Commission or other persons as the Commission may designate by order upon a demonstrated regulatory need.

Final Rule

Two commenters recommended that the final rule require RNSAs to make collected confidential information available to SROs without obtaining an SEC order. However, another commenter recommended that access to such information by securities exchanges should require a clear regulatory purpose. The Commission recognizes that other regulators and SROs may require access to the confidential information reported to RNSAs pursuant to final Rule 10c–1a for regulatory purposes. However, it is appropriate to only provide such information by order of the Commission upon a demonstrated regulatory need. Imposing this condition on access to reported information should help ensure the security of the reported data elements, while taking into account that other regulators may require access. For the foregoing reasons the Commission is adopting the requirement that RNSAs make collected information available to the Commission, and make it available to other persons as the Commission may designate by order upon a demonstrated regulatory need, with a technical edit, with the relevant paragraphs renumbered to align with the format of the final rule.

M. Cross-Border Application of Rule 10c–1a

The Commission received a number of comments about the rule's intended cross-border application. Many of these comments requested that the Commission provide cross-border guidance to help promote legal certainty and competitive equity, and that in doing so the Commission rely on existing definitions for U.S. and non-US market participants to avoid new documentation requirements. Among other things, these commenters requested clarification or specific guidance as to when non-U.S. persons would be subject to the final rule and clarity as to the circumstances under which a securities loan will be deemed to be within the U.S. market for purposes of applying reporting requirements. Some commenters stated that because the rule would apply broadly to “any person” that loans a security on behalf of itself or another person, that could mean the rule would apply to “all lenders” regardless of whether the lender loans a U.S. security or a non-U.S. security and whether or not the lender was a U.S. person or not. Other commenters, in contrast, warned that, because a significant amount (up to 18 percent) of lender transactions in the U.S. market are provided by funds outside of the U.S., the Commission risks not capturing this considerable volume of activity if it does not clarify cross-border scope of the final rule. Many of these commenters raised questions about the scope of the Commission's cross-border regulatory authority under section 10(c).

Although the Proposing Release did not propose rule text addressing the cross-border aspects of proposed Rule 10c–1 or otherwise discuss the rule's cross-border reach, the Commission is addressing commenters' contentions about the Commission's cross-border authority under section 10(c) and offering general guidance as to the rule's cross-border scope. As an initial matter, the Commission advises that final Rule 10c–1a is intended to reach the full scope of the cross-border authority provided for by section 10(c).

Turning to the cross-border scope of section 10(c), the Commission's understanding of that provision's cross- border reach is based on the territorial approach that the Commission has applied when crafting rules to implement other provisions of the Exchange Act. Consistent with that territorial approach (which is based on Supreme Court precedent, including Morrison v. National Australia Bank, Ltd. and its progeny) the Commission examined the relevant statutory provision to determine the domestic conduct that is covered by the provision. By its terms, section 10(c) requires reporting when, directly or indirectly, a person has “effect[ed], accept[ed], or facilitate[d]” a transaction involving the loan or borrowing of securities. Based on that language, the Commission concludes that the relevant domestic conduct that triggers the Commission's regulatory authority under section 10(c) is conduct within the U.S. that comprises (in whole or in part) effecting, accepting, or facilitating of a borrowing or lending transaction. Because the Commission intends final Rule 10c–1a to be co-extensive with the regulatory scope of section 10(c), the Commission is of the view that the rule's reporting requirements will generally be triggered whenever a covered person effects, accepts, or facilitates (in whole or in part) in the U.S. a lending or borrowing transaction.

One significant exclusion proposed by commenters concerned non-U.S. residents. Having considered those comments, such an exclusion would not be appropriate. Section 10(c) is focused on increasing transparency regarding lending and borrowing transactions that are effected, accepted, or facilitated (in whole or in part) within the U.S. Excluding borrowing and lending transactions by non-U.S. persons when those transactions are effected, accepted, or facilitated (in whole or in part) within the U.S. could impact the completeness of the data, providing less transparency and potentially resulting in U.S. market participants receiving misleading information as a result of those omissions. Moreover, the exclusion of non-U.S. persons suggested by the commenters could cause competitive harm to U.S. market participants as it would allow non-U.S. persons engaging in the same activities as U.S. persons to benefit from the transparency about the U.S. securities lending market provided by final Rule 10c–1a without contributing to that transparency. For these reasons, the Commission declines to adopt such an exclusion.

Commenters suggested several other exclusions from final Rule 10c–1a that the Commission has also decided not to adopt, including that the final rule should apply only to loans of U.S.-exchange traded securities offered by a U.S. lender. One commenter stated “that market participants—particularly those outside of the United States—might determine to limit or cease trading or interacting with U.S. intermediaries, or leave the U.S. markets altogether” in order to prevent public disclosure of their transactions. Another commenter stated that capturing such activities would not further the transparency goals of the rule but, instead, could inadvertently result in market participants segregating their foreign lending activities so as to avoid reporting under final Rule 10c–1a. Other commenters suggested that final Rule 10c–1a should apply only to loans of securities in which: (1) the country of issue and primary trading market of the securities is the U.S., and (2) the beneficial owner/lender or Lending Agent is a U.S. person. Having considered these comments, as a policy matter, it is not appropriate to exclude any of these categories of lending transactions. Each of these types of transactions could have a potential direct impact upon U.S. markets and U.S. market participants and, thus, consistent with the breadth of the cross-border scope of section 10(c) it is appropriate to retain these categories of transactions within the rule's reporting requirements so as to enhance overall transparency in the U.S. securities market.

Finally, some commenters asked the Commission to consider the issue of potential overlap with the SFTR that is already implemented in the EU and UK. These commenters stated that the SFTR uses the domicile of the transacting entities ( i.e., both the lender and the borrower) to determine which entity has the reporting obligation and expressed concern about potential adverse impacts from regulatory overlap with Rule 10c–1a to the extent it covers non-U.S. persons. One commenter suggested that the Commission, at a minimum, include a carve-out from proposed Rule 10c–1 (or allow for substituted compliance with respect to the rule) for loans by EU or UK lenders that are subject to reporting under the EU or UK SFTR. The Commission acknowledges that there will be some overlap with the SFTR, however, the overlap that may occur with respect to entities established in the EU or UK (or to the EU or UK branch of a non-EU/UK entity) is not sufficient to warrant an exclusion. To the extent transactions are subject to both the EU or UK SFTR reporting rules and Rule 10c–1a, this overlap will be due to the occurrence within the U.S. of the relevant domestic activities as specified by section 10(c). For the reasons discussed above, related to transparency and competition within the U.S. market, Rule 10c–1a will apply to those transactions notwithstanding any potential overlap and resulting adverse impacts.

N. Additional Comments

One commenter stated that the Commission has already exhausted its rulemaking authority under section 984(b) of the Dodd-Frank Act, that the authority to promulgate rulemaking has expired, and that the Commission is seeking to regulate transactions outside the scope of the intended Congressional authority. Section 984(a) of the Dodd-Frank Act, codified in section 10(c)(1) of the Exchange Act, has no deadline for Commission action. Although, section 984(b) includes such a date, the “[p]assage of the statutory action deadline does not deprive the agency of the power to act.” Section 984(b) of the Dodd-Frank Act provides that the Commission shall promulgate “rules” and does not limit the number of any such rules. The Commission did not “exhaust” its authority under sections 984(a) and (b) by adopting the Investment Company Reporting Modernization rules.

The commenter also stated that, by “linking” securities loan transactions entered into in order to fulfill delivery obligations arising from customer short sales, the Commission would be exceeding the scope of its rulemaking authority. With regard to authority, section 10(c)(1) of the Exchange Act confers the Commission with plenary authority over both short sales and the loan or borrowing of securities, with a limited exception to authority over securities loans related to the safety and soundness or systemic risk of certain financial institutions such as banks and credit unions. A loan of a security may be used for short sales or other purposes, and the use of a securities loan to facilitate a short sale is separate and distinct from the short sale itself. While it is possible that some persons may seek out short sale information through securities lending information that will be provided under this rule, the modifications made from the proposed rule, in particular delaying the publication of the amount of the loan, should make it more difficult for such persons to succeed at such efforts, thereby addressing the potential negative consequences of such “linking” of securities loan information to short sale information ( e.g., short sellers seeking to use securities lending information to alter their short selling strategies, thereby potentially reducing liquidity and harming investors) raised by the commenter.

Some commenters urged the Commission to take additional or different regulatory and non-regulatory actions than the approaches that were proposed, including actions that the Commission did not propose. These suggestions covered a variety of areas, including: notifications to retail investors; securities lending transaction terms; other SEC regulations; SEC enforcement actions and penalty provisions; studies, audits, and roundtables; reporting technologies; broker-dealer revenue; and “onward lending.” These issues are outside the scope of the proposal, and the final amendments appropriately further the Commission's objectives of promoting investor protection, enhancing market efficiency, and facilitating capital formation by implementing the requirements of section 984(b) of the Dodd-Frank Act and enhancing the transparency of securities lending and borrowing.

VIII. Compliance Date

The Proposing Release did not include a specific proposed compliance date for the final rule, but commenters expressed a range of views on the length and structure of an implementation period.

The Commission received several comments recommending an implementation period that runs from the time that technical specifications are published by an RNSA. One commenter proposed an “initial compliance period of at least 18 months to develop necessary systems and otherwise prepare for compliance.” Another commenter proposed a period of “at least 18 months after publication of technical specifications by the registered national securities association to come into compliance with the new requirements.” Another commenter recommended that the Commission “provide a minimum two-year implementation period” for the final rule. One commenter stated that ample time should be provided for covered persons to develop “procedures and protocols and properly train their compliance and trading personnel before subjecting them to additional regulation.” However, other commenters encouraged the Commission to expedite the adoption of the final rule.

Certain commenters requested a phased approach to implementation, including a recommendation that the implementation of the final rule begin only with loans of U.S. equity securities. Other commenters recommended initially reporting only securities listed or traded on a U.S. exchange and reporting other U.S. equity securities or debt securities after further study. Other commenters recommended deferring public dissemination of data to allow the Commission to gain experience with the data. Similarly, one commenter supported “an approach to implementation that requires confidential reporting for regulatory purposes to give the SEC experience with the data to refine it as necessary to ensure it serves its purpose. Any further stages that would require public dissemination of the data should only be considered following additional economic analysis and public consultation.”

Another commenter recommended that “[a]ny proposed timeline should carefully consider the universe of entities required to report transactions for the first time.” The Commission recognizes that covered persons may have differing degrees of expertise and familiarity with existing RNSA reporting systems, and that lacking such familiarity could increase the time needed to prepare for compliance with final Rule 10c–1a's requirements. However, the Commission also recognizes that the final rule permits the use of reporting agents, as well as other third party vendor service providers, either of which could help facilitate preparation for fulfillment of the final rule's reporting requirements by less experienced covered persons. For example, some commenters stated that existing third party vendors have experience and technological capacity to be able to provide reporting-related services ( e.g., data vendors and entities that report information for other regulatory purposes).

Another commenter proposed means of expediting the implementation of the final rule, including initially: (1) applying it to only larger market participants; (2) requiring only certain data elements; and (3) not enforcing it for several years. However, limiting the reporting of Rule 10c–1a information ( i.e. to larger market participants or only requiring certain data elements) would provide the public with a narrow and incomplete view of the securities lending market activity. Similarly, not enforcing the rule for several years could incentivize non-reporting and skew the reported data to disproportionally represent market participants that voluntarily report.

Taking into consideration commenters' wide-ranging recommendations, the following compliance dates strike an appropriate balance between making securities loan information publicly available and providing adequate time for industry participants to come into compliance. Specifically, the final rule's compliance dates require that: (1) an RNSA propose rules pursuant to final Rule 10c–1a(f) within four months of the effective date of final Rule 10c–1a; (2) the proposed RNSA rules are effective no later than 12 months after the effective date of final Rule 10c–1a; (3) covered persons report Rule 10c–1a information to an RNSA starting on the first business day 24 months after the effective date of final Rule 10c–1a (the “reporting date”); and (4) RNSAs publicly report Rule 10c–1a information pursuant to final Rules 10c–1a(g) and (h)(3) within 90 calendar days of the reporting date for covered persons to report Rule 10c–1a information to an RNSA. Additionally, upon the reporting date requiring covered persons to report Rule 10c–1a information to an RNSA, RNSAs are required to fulfill the data retention and availability requirements—including relevant information security policies and procedures—pursuant to paragraphs (h)(1), (h)(2), and (h)(4), and may establish and collect reasonable fees pursuant to paragraph (i) of final Rule 10c–1a.

This approach to the final rule's compliance dates balances the Commission's goal of increasing transparency in the securities lending market with providing RNSAs and market participants with adequate time to implement systems and processes to comply with the final rule's reporting requirements. The Commission has also taken into consideration that the technology to collect and disseminate Rule 10c–1a information already exists. Specifically, FINRA, which is currently the only RNSA, already operates facilities that collect and disseminate transaction information, including TRACE. FINRA's extensive experience in developing and operating such facilities will enable it to design and propose rules regarding the format and manner of its collection of information within four months following the effective date of the final rule. Additionally, requiring that RNSA rules are effective no later than 12 months after the effective date of final Rule 10c–1a will help to ensure that information improving the transparency of the securities lending market is made available to the public without unnecessary delay.

A reporting date for covered persons of 24 months after the effective date of final Rule 10c–1a is sufficient for covered persons (and eligible reporting agents) to prepare for compliance with the final rule's reporting requirements. Two commenters recommended an implementation period of two years, and another commenter recommended a compliance period of at least 18 months. One commenter's proposed implementation period of eighteen months after publication of technical specifications by an RNSA could provide more or less implementation time to covered persons than the final rule does. Depending on when an RNSA adopts rules pursuant to final Rule 10c–1a(f), covered persons could potentially have more than 12 months to begin reporting Rule 10c–1a information, including more than the 18 months the commenter recommends. Further, the final rule has been modified to reduce certain burdens, such as requiring end-of-day reporting rather than reporting in 15-minute increments, and removal of the available to lend and securities on loan requirements, which should reduce the amount of time required to prepare for compliance. As stated above, in this part, the reporting date for covered persons strikes a balance between making securities loan information publicly available and providing industry participants with sufficient time to come into compliance. Covered persons' permitted reliance on reporting agents and ability to use third party vendors to help facilitate the fulfillment of reporting obligations will allow for the outsourcing of certain functions to prepare for and comply with the final rule's requirements.

A compliance date that requires RNSAs to publicly report Rule 10c–1a information within 90 calendar days of the reporting date for covered persons reporting Rule 10c–1a information to an RNSA will help ensure the utility of such data once it is made publicly available. Providing RNSAs with up to 90 calendar days between covered persons being required to publicly report Rule 10c–1a information and RNSAs being required to make such information available to the public pursuant to final Rules 10c–1a(g) and (h)(3) will help RNSAs resolve any initial issues with collecting, aggregating, and publishing Rule 10c–1a information following the reporting date for covered persons. Additionally, upon the reporting date for covered persons an RNSA is required to fulfill the data retention and availability requirements—including relevant information security policies and procedures—pursuant to paragraphs (h)(1), (h)(2), and (h)(4), and may establish and collect reasonable fees pursuant to paragraph (i) of final Rule 10c–1a.

IX. Economic Analysis

A. Introduction and Market Failure

1. Introduction

The Commission has considered the economic effects of final Rule 10c–1a and, wherever possible, the Commission has quantified the likely economic effects of the final rule. The Commission is providing both a qualitative assessment and quantified estimates of the potential economic effects of the final rule where feasible. The Commission has incorporated data and other information to assist it in the analysis of the economic effects of the final rule. However, as explained in more detail below, because the Commission does not have, and in certain cases does not believe it can reasonably obtain, data that may inform the Commission on certain economic effects, the Commission is unable to quantify certain economic effects. Further, even in cases where the Commission has some data, quantification is not practicable due to the number and type of assumptions necessary to quantify certain economic effects, which render any such quantification unreliable. Our inability to quantify certain costs, benefits, and effects does not imply that such costs, benefits, or effects are less significant.

Commenters raised a number of concerns with the analyses and conclusions in the Proposing Release. The Commission reviewed all of the comments received and, in a few instances, have modified the subsequent economic analysis in response to additional information provided by commenters. We have also conducted analyses of changes to the proposed rule text.

The Commission believes that the final rule will increase transparency in the securities lending market by making available the public portion of Rule 10c–1a information, which is more comprehensive than existing data, and by making such data available to a wider range of market participants and other interested persons than currently are able to access existing data.

The subsequent benefits include a reduction of the information disadvantage faced by end borrowers and beneficial owners in the securities lending market, improved price discovery in the securities lending market, increased competition among providers of securities lending analytics services, reduced costs associated with tracking market conditions for broker-dealers and lending programs, and improved decision-making by investors, beneficial owners and other market participants. The Commission believes that final Rule 10c–1a will also likely reduce the borrowing costs of some securities, which will improve price discovery, liquidity, and capital formation in the underlying security markets. The Commission also believes the final rule will benefit investors by increasing the ability of regulators to surveil, study, and provide oversight of both the securities lending market and individual market participants.

The Commission believes that the final Rule 10c–1a will result in costs. The final rule will lead to direct compliance costs as entities providing Rule 10c–1a information to an RNSA will have to build or adjust systems to meet the requirements of the final rule. The entities that provide Rule 10c–1a information to an RNSA may absorb these costs in the form of lower profits or may pass them on to their customers in the form of increased fees for broker-dealer services or lending program services. The final rule will also impose direct costs on an RNSA responsible for collecting, maintaining, and distributing the data, who may pass on these costs by imposing fees on entities that provide Rule 10c–1a information to an RNSA and/or consumers of Rule 10c–1a data (“Rule 10c–1a data”). Additionally, the Commission believes that the final rule could render existing securities lending data services less valuable, potentially leading to less revenue for the firms currently compiling and distributing these data for a fee. Also, broker-dealers and lending programs will have costs in the form of lost information advantage when dealing with beneficial owners and end borrowers in the securities lending market. For securities lending data that are currently not reported, or to which access is limited, making the data public may affect the profitability of certain trading strategies as investors use the data to learn about market sentiment and adjust their trading strategies accordingly.

2. Market Failures

In the securities lending market, the cost to borrow a given security depends on a number of factors, including the current demand for the security, the length of the loan, and the type and amount of collateral used, among others. Information about loan prices along with information about loan characteristics helps inform market participants about whether the price of a given securities loan is consistent with the current market rate. At the same time, the securities lending market is characterized by information frictions that stem from the fact that access to timely securities lending data is limited for some market participants. This means that, at any point in time, there is incomplete information on market conditions and some market participants have better information than others on borrowing costs and transactions. Such incomplete information and asymmetric information may lead to inefficient prices for securities loans (including loans of equity securities and fixed income securities).

There is a general lack of comprehensive information on current market conditions in the securities lending market. Most providers of commercial securities lending data currently focus on loans from lending programs to broker-dealers (“Wholesale” loans) and largely use a “give-to-get” model, where entities who wish to obtain securities lending data are typically required to: (1) be participants in the Wholesale lending market themselves, with data that they could provide, and (2) provide their data to the commercial vendor in order to access the full dataset provided by the vendor. This means that only those market participants with data to report for themselves are able to access the data, and other market participants have a very limited view into the Wholesale market. Furthermore, participation in give-to-get data products is voluntary, meaning that relevant observations could be missing from the data in a systematic fashion, thus biasing the impression it creates of the lending market.

Some commenters disagreed with the Commission's assessment of the current opacity of the securities lending market. One commenter pointed out that there are some commercial securities lending datasets that are available to all subscribers, and another commenter stated that “the `give-to-get' model is not the only model for commercial data for securities lending.” Based on the Commission's experience, the collection of data for these currently available commercial datasets that are not give-to-get largely relies on surveying asset managers, and potentially others, in the Customer segment of the market about their borrowing experiences. To differentiate this data from the give-to-get data, this type of dataset will be referred to going forward as “Customer market survey” data.

One commenter stated that “the Commission does not explain why [Customer market] survey data is inadequate or unavailable.” The Commission acknowledges that Customer market survey data may be more widely available, as access to these datasets is generally not restricted in terms of which entities can purchase the data. However, the Commission believes that Customer market survey datasets are inadequate for two reasons. First, since they rely on surveys of borrowers in the Customer segment of the market, they lack comprehensiveness and have limited insight into the Wholesale market. Second, similar to the give-to-get datasets, Customer market survey datasets may also contain biases as they rely on voluntary submissions of data from a potentially limited subset of market participants. As discussed above, these two types of datasets differ in terms of the market segments that they cover, and in terms of their accessibility.

One commenter stated that the Commission “has overstated the level of opacity in the securities lending market,” that both the number of data vendors and the detail of their data products have increased over time, and that market participants generally have access to one or more securities lending datasets. However, based on the Commission's experience, the commercial securities lending data products that are currently available lack comprehensiveness and likely suffer from biases. Furthermore, while some market participants could potentially have access to multiple datasets, this would generally be possible only for a limited set of market participants who have access to give-to-get data, and who may find it cumbersome or costly to combine different types of data. The Commission is not aware of any commercially available securities lending dataset that currently provides securities lending data as comprehensive, accessible, and informative about all segments of the securities lending market ( i.e., both the Wholesale market and Customer market), as the data provided by the final rule.

One commenter stated that lending agents frequently provide their clients with benchmark reports based on market data, and that beneficial owners currently benefit from substantial information obtained from lending agents. However, the Commission believes that, even if some lending agents or broker-dealers are incentivized to provide their clients with information about the quality of their securities lending services, it is unlikely that they have incentives to provide information in a standardized fashion. Therefore, it is unlikely that market participants would be able to use this information to compare the quality of securities lending services across lending agents or broker-dealers.

The Commission believes that the problems of incomplete information and information asymmetries in the securities lending market are unlikely to be solved by market forces. Specifically, these information problems are unlikely to be solved by the availability of commercial securities lending data products. Firstly, there is a general a lack of incentives to improve the accessibility of give-to-get datasets. This is because market participants may be discouraged from contributing their data to give-to-get datasets if these datasets are widely available, e.g., to sophisticated investors such as hedge funds who could use this information to learn about the other participants' trading or hedging strategies. Therefore, for give-to-get data vendors, restricting access is likely seen as necessary in order to persuade market participants to contribute to the vendors' data products, and information asymmetry would be expected to persist as a result. Secondly, and more generally, as commercial data vendors lack the authority to mandate the reporting of securities loans, both the commercial data products that currently exist (both give-to-get and Customer market survey models), as well as any new commercial data product that could presumably emerge, can only be based on the voluntary contribution of data. Market participants who choose not to contribute data may so choose because they believe it is in their interest to keep their data out of public view. This makes it unlikely that any securities lending data vendor will be able to produce a securities lending data product that is comprehensive and free from biases. Thirdly, as discussed above, the information that lending agents or broker-dealers are incentivized to provide to their clients is unlikely to be standardized and thus useful for comparing the quality of securities lending services across lending agents or broker-dealers. Thus, the availability of this information would also be unlikely to solve the problems of incomplete and asymmetric information in the securities lending market.

A. Economic Baseline

The baseline against which the costs, benefits, and the effects on efficiency, competition, and capital formation of the final rule are measured consists of the current state of the securities lending market, current practice as it relates to securities lending and availability of data about securities lending, and the current regulatory framework. The economic analysis appropriately considers existing regulatory requirements, including recently adopted rules, as part of its economic baseline against which the costs and benefits of the final rule are measured.

Several commenters requested the Commission to consider interactions between the economic effects of the proposed rule and other recent Commission rules. The Commission recently adopted four of the rules mentioned by commenters as potentially impacting the economic effects of the final Rule 10c–1a, namely the recent Amendments to Form N–PX Adoption, the Settlement Cycle Adoption, the May 2023 SEC Form PF Amending Release, and the Beneficial Ownership Amending Release. These recently adopted rules were not included as part of the baseline in the Proposing Release because they were not adopted at that time. In response to commenters, this economic analysis considers potential economic effects arising from the extent to which there is any overlap between the compliance period for final Rule 10c–1a and the compliance periods for each of these four adopted rules.

Other rules mentioned by commenters remain at the proposal stage. To the extent those proposals are adopted, the baseline in those subsequent rulemakings will reflect the regulatory landscape that is current at that time.

1. Securities Lending

A securities loan is typically a fully collateralized transaction whereby the lender, also known as the beneficial owner, temporarily transfers legal right to a security to the borrower, the counterparty, in exchange for compensation. The form of compensation depends on the type of collateral used to secure the transaction. There are two general types of collateral: cash and non-cash.

In the U.S., the most common form of collateral for equity security loans is cash. The borrower of the security deposits collateral, typically 102 percent or 105 percent of the current value of the asset being loaned. The lender then reinvests this collateral, usually in low-risk interest-bearing securities, then rebates a portion of the interest earned back to the borrower. The difference between the interest earned and what is rebated to the borrower is the lending fee earned by the lender. The portion of the interest earned on the reinvested collateral that is returned to the borrower is called the rebate rate, and is a guaranteed amount set forth in the terms of the loan. It is possible for the lender to lose money on the loan if the interest earned on the reinvestment of the collateral does not exceed the rebate rate. If the security is in high demand in the borrowing market, the rebate rate may be negative, indicating that the borrower does not receive any rebate and must also provide additional compensation to the lender. In contrast, borrowers of loans that are collateralized other than with cash typically must pay a lending fee. The following will refer to “cost to borrow” or “borrowing costs” being higher when rebate rates are lower (or negative) in the case of cash-collateralized loans, and lending fees are higher in the case of non-cash-collateralized loans.

Borrowing costs are influenced by the current demand for the given security, the potential difficulty a particular broker-dealer may face finding an alternative source of loans, the length of the loan, the collateral used, the creditworthiness of the counterparty, and the relative bargaining power of the parties involved, among other factors. Consequently, there is usually a significant range of borrowing costs for loans of the same security on the same day to different entities.

In the Wholesale market, securities loans are most commonly obtained through bilateral negotiations between lending programs and broker-dealers, often with a phone call. Generally, when an end investor wishes to borrow a share, and their broker-dealer does not have the share available in their own inventory or through customer margin accounts to loan, their broker-dealer will borrow a share from a lending agent with whom they have a relationship. Obtaining a securities loan often involves extensive search for counterparties by broker-dealers.

Investors borrow securities in the Customer market for a variety of reasons. In the equity market, a primary reason for borrowing shares is to facilitate a short sale. Investors use short sales to take a directional position in a security, or to hedge an existing position. When investors execute a short sale, they do not borrow the shares on the day of the short sale. Rather, because the stock market settles up to two business days after a transaction occurs (“T+2”) and the lending market has same-day settlement, the loan actually occurs on the settlement day.

Options market activity can also be a source of demand for security loans as short selling is a critical component of delta hedging. Delta hedging occurs when options market participants, particularly options market makers, holding directional positions hedge their inventory exposure by taking offsetting positions in the underlying stock. Equity options markets are often significantly less liquid than the markets for their underlying securities. Delta hedging a long call or short put position requires short selling, which in turn requires borrowing the underlying asset.

Equity security loans can also occur to close out a fail to deliver. Fail to delivers occur when one party of a transaction is unable to deliver at settlement the security that they previously sold. Fail to delivers can occur for multiple reasons. Rule 204 of Regulation SHO generally requires a participant of a registered clearing agency to close out any fail to deliver at the registered clearing agency by purchasing or borrowing securities of like kind and quantity. Thus, a broker or dealer that is required to close out a fail to deliver can borrow shares in the lending market to comply with Rule 204 of Regulation SHO. Doing so allows more time for the individual to source the shares or purchase them in the open market.

The financial management activity of banks also drives securities loans, particularly in the Wholesale fixed-income securities lending market. It is the Commission's understanding that a significant fraction of debt security loans occurs as banks manage liquidity on their balance sheets. Securities loans help banks manage liquidity on their balance sheets because when a security is on loan, legal claim to the security transfers to the borrower. Thus, banks lacking sufficient high-quality liquid assets on their balance sheet may borrow such high-quality assets to bolster their liquidity ratios. U.S. Treasury/Agency bonds are often lent for this purpose, and consequently are the most commonly lent securities.

Also, the Commission understands that some financial entities may use securities loans to obtain the type of collateral required for other agreements they are trying to enter into. For example, if a contract requires a certain kind of fixed income security as collateral, a firm may borrow that security to collateralize the contract.

While a security is on loan, the borrower is the legal owner of the security and receives any dividends, interest payments, and, in the case of equity security loans, holds the voting rights associated with the shares. Voting rights remain with the borrower until the loaned securities are returned. Usually, the terms of the loan stipulate that dividends and interest payments must be passed back to the beneficial owner in the form of so-called “substitute dividends.” Because dividends and substitute dividends are sometimes taxed differently, an investor for whom a substitute dividend is taxed at a lower rate than a dividend may loan its shares to an investor for whom dividends are taxed at a lower rate than substitute dividends.

Several commenters pointed out additional reasons for borrowing securities. One commenter stated that there are many reasons for securities lending that are unrelated to short selling. Another commenter purported that securities lending can be used to gain influence in the voting of equity securities.

However, the Commission expects that the majority of equity securities lending, particularly in the Customer market, occurs to facilitate short selling. Investors' ability to use securities loans for purposes other than short selling is limited by the “permitted purpose requirement” of the Board of Governors of the Federal Reserve System's Regulation T. Regulation T broadly governs the lending activities of broker-dealers, and specifies that a broker- dealer may generally borrow or lend U.S. securities from or to a (non-broker-dealer) customer solely “for the purpose of making delivery of the securities in the case of short sales, failure to receive securities required to be delivered, or other similar situations,” unless an exemption applies. This limitation results in a close correlation between information about aggregate Customer loan sizes and short interest.

2. Current State of Transparency in Securities Lending

As described above, currently available securities lending data are produced by commercial data vendors, and are based on voluntary data contributions, either using a give-to-get model or from Customer market surveys. Some commenters expressed their belief that the Proposing Release did not adequately analyze the impact of Customer market survey data. This baseline, and the discussion following have been expanded to provide a more detailed discussion of the Customer market survey data and its impact on securities lending transparency. The available data have limitations. Specifically, they lack comprehensiveness in that they are based on voluntary data contributions. The reliance on voluntary data contributions increases the likelihood that data are missing in a non-random manner which can introduce biases into the data. Lastly, not all market participants have access to all existing commercial datasets resulting in information asymmetries between market participants.

Some commercial data vendors obtain information on Wholesale loans using a give-to-get model. Access to data collected using a give-to-get model is largely restricted, as only certain entities can purchase the data. The Commission understands entities with access to the give-to-get Wholesale market data access that data using various means such as an application programming interface (API), spreadsheet add-in applications, file downloads, or directly from the distributor's website. It is the Commission's understanding that some large institutional investors who would like the give-to-get data, such as hedge funds, cannot access it, even for a fee, because they do not provide lending data to the commercial vendors and so are not qualified to purchase the data. Additionally, distributing the data to these investors may discourage other market participants from contributing their data to the data vendors. Because securities loans are often entered into to facilitate various trading and hedging strategies, if the give-to-get data providers expanded access to their data, some securities lending market participants may be discouraged from contributing data. Consequently, if sophisticated traders such as hedge funds can access the data, then some market participants may be leery of contributing data to the give-to-get data vendors for fear of hedge funds learning about their trading or hedging strategies. Additionally, while some data vendors do allow non-lending market participants, such as academics and regulators, to access the data for a fee, they sometimes place usage restrictions on the data that make it unusable for regulatory and some academic functions. The Commission believes, based on conversations with industry participants and our staff's use of some of the data, that the coverage and timeliness of the three biggest give-to-get Wholesale data vendors are roughly comparable.

Other commercial data vendors provide a different approach to securities lending data by surveying asset managers and others in the Customer segment of the market about their borrowing experience, such as their costs to borrow, from which they provide estimates of lending fees. One commenter stated that the Commission “offers no reason to believe that the [Customer market] survey and `give-to-get' data are materially different.” Customer market survey data have some characteristics that make them distinct from the give-to-get data provided by the vendors of Wholesale market data discussed above. First, Customer market survey data vendors generally aggregate information from respondents in the Customer market—that is, the end borrowers that engage in securities borrowing to facilitate short selling—and thus cover a different segment of the securities lending market. Second, while Wholesale market data are mostly available only on a restricted basis, Customer market survey data are generally available to any purchaser. However, despite these differences, Customer market survey data face a similar issue to the give-to-get data, in that they only contain information about the subset of those end borrowers that choose to provide data. Since it is unlikely that the full universe of lending programs and borrowers contribute all data to any given data vendor, the Commission believes that both give-to-get and Customer market survey data lack comprehensiveness.

Furthermore, as stated in the Proposing Release and restated here, the Commission believes that the give-to-get data are likely biased due to non-random omissions that result from the voluntary nature of data submissions. The Commission also believes that, since Customer market survey data are also based on voluntary submissions, these data also suffer from bias due to non-random omissions. As will be further detailed below, data based on voluntary submissions can contain biases because market participants that choose not to disclose their data likely make that choice because it is in their strategic interest not to disclose, resulting in non-random omissions.

One commenter stated that the Commission's statements in the Proposing Release that give-to-get data could be systematically biased has “no basis beyond mere speculation.” The Commission disagrees. It is true that it is not possible to empirically analyze or quantify the extent of this bias using existing data because there currently does not exist a complete dataset with which to compare the existing data, and comparing a biased sample to an unbiased sample is generally the only data-based means of determining the extent of the bias in a non-random sample. Instead, the determination that data are likely to not contain a representative sample is routinely made in economic studies (as well as data analysis generally) based on an understanding of how the sample is created, which alone can be sufficient to determine the likely validity of the sample. The fact that non-random data omissions result in biased data is well established in statistics and data science, and simply stems from the logic that, if the reporting subsample is systematically different from the population, then statistics based on the reporting sample will only reflect the reporting sample, and will not be reflective of the whole population.

In the context of commercial securities lending datasets, the Commission believes that, because market participants can choose whether or not they want to contribute data to commercial datasets, it is likely that the sample of securities lending data that is reported to commercial data vendors is, on average, different from the sample that is not reported. As such, there are well-established theoretical reasons as to why these data are likely biased. These reasons are largely based on the fact that market participants that voluntarily contribute data to commercial datasets “self-select” the data that they would like to be included in the dataset. The practice of having entities under study “self-select” into the dataset used to study them is widely acknowledged in the empirical economics literature to be very likely to lead to biased data. This is because the incentives to provide data are likely connected to the incentives that drive the behavior under study, such that the behavior of those market participants that do contribute data are likely to be systematically different from the behavior of those that do not contribute. Trying to draw inferences from such a dataset about the entire population of market participants would thus result in a biased view of the market.

The Commission has identified examples of such incentives that would likely introduce such “self-selection biases” into commercially available securities lending datasets. First, there may be competitive reasons for market participants to selectively choose which loans to contribute to a securities lending dataset. For example, to attract more beneficial owner customers, lending programs would like their average terms to appear better than the benchmark average terms. Thus, they are incentivized to voluntarily contribute information about their lending activity for loans with higher-than-average borrowing costs—skewing the average cost to borrow in the resulting dataset up and making the lender's performance appear better than it actually was. This could lead the reported borrowing costs in commercial datasets to be systematically higher than the actual borrowing costs in the securities lending market.

Similarly, some market participants may choose not to voluntarily contribute data to securities lending datasets for their own strategic reasons, such as a desire to keep their activity private. The decision to not to participate is, in itself, indicative that such market participants are likely not reflective of those who do choose to participate. For example, market participants that choose not to contribute data may be more likely to be sophisticated investors with greater incentives to protect their trading and hedging strategies. This matters because the terms of securities loans are influenced by the characteristics of the parties involved, and so it follows that the borrowing costs for market participants that do not voluntarily contribute data are systematically different than the borrowing costs for market participants that do contribute data.

In addition to the issue of self-selection bias, currently there is no securities lending dataset that contains information about both the Customer and Wholesale segments of the market. The give-to-get datasets lack significant coverage of the Customer segment of the market, while the Customer market survey data lack significant coverage of the Wholesale segment of the market. While some, but not all, market participants could potentially gain access to data on both the Wholesale markets and Customer markets by subscribing to both give-to-get and Customer market survey datasets, this would generally be possible only for market participants that contribute data (and thus have access) to give-to-get datasets and would also not surmount the problems of incomplete or potentially biased data. Different types of data may also be cumbersome to combine. There are also limitations to using the Customer market survey data to assess conditions in the Wholesale market. Furthermore, the need to subscribe to multiple datasets increases market participants' overall costs of data access.

As an additional source of securities lending data, some market participants that have many connections within the securities lending market, such as large broker-dealers and lending programs, can query other market participants directly about current conditions. These centrally connected market participants are also likely to have access to both give-to-get Wholesale data and customer market survey data. Consequently, the largest and most centrally connected broker-dealers and lending programs likely have access to better information about the current state of the lending market than other participants, including their customers, the beneficial owners, and end borrowers. This asymmetric information between those on the periphery of the lending market and those in the center, who have access to both Wholesale and Customer market commercial datasets as well as queried data from their connections, may lead to inferior terms for those on the periphery in the form of lower performance and less favorable prices for beneficial owners and end borrowers. Because of the above-described issues with the comprehensiveness and accessibility of existing commercial datasets, the availability of commercial data products for the securities lending market does not alleviate this information asymmetry.

One commenter disagreed with the Commission's view that there are information asymmetries between those in the center of the lending markets and beneficial owners and borrowers on the periphery, because beneficial owners benefit from the information obtained by their lending agents. However, the Commission believes that, even if lending agents provide beneficial owners with benchmark reports based on commercial securities lending data, this would not eliminate all information asymmetries. The lending agent, or other centrally located entity, has access to the raw data, as well as the potential to influence that data through their own contributions of data to commercial data vendors, whereas the beneficial owner would need to rely on a subset of the processed data that their lending agent chooses to provide to them, if any.

In addition to the specific problem of information asymmetry between various market participants, the lack of comprehensive and widely available data on securities lending activity likely means that the prices at which securities loans take place are not efficient, relative to the hypothetical case in which complete information about securities lending activity were widely available. Asymmetric information in general deters outsiders from entering the market, as they anticipate not being able to transact on the same terms. This limits both liquidity (because fewer participants enter to transact) and price discovery (because not all information enters prices). Moreover, even lenders with many connections in the securities lending market lack a complete picture of the lending market, implying that the loan prices that they quote may not be completely efficient.

Commenters stated that the lending market for U.S. Government securities is already transparent “because there is sufficient liquidity and demand for loans of these types of securities on platforms and venues that have a high degree of transparency.” However, the commenters did not provide additional explanation as to why a high degree of liquidity would entail a transparent market, or additional details regarding the transparency of the platforms and venues that trade in U.S. Government securities.

The Commission believes that information asymmetries may still be present in the lending market for U.S. Government securities. Some information about U.S. Government securities loans is available from the commercial securities lending datasets described above. In addition, the Commission understands that information about loans of U.S. Government securities that are awarded to primary dealers through the Federal Reserve Bank of New York's System Open Market Account (SOMA) portfolio is available from the Federal Reserve Bank of New York. While volume and rate information from these auctions may be informative to participants about benchmark lending rates, these data do not contain information about activity in the broader U.S. Government securities lending market. Furthermore, certain aggregate data from the Federal Reserve's FR 2004 reports, which collect weekly and daily position, transaction, financing, and fails data of primary dealers in U.S. Government securities and other selected fixed income securities, are published in the Federal Reserve Bank of New York's press release, “Weekly Release of Primary Dealer Transactions.” This dataset contains aggregate volume information about securities lent and borrowed by primary dealers. It does not contain information about lending rates, and also does not contain information about the U.S. Government loans of market participants other than primary dealers.

Furthermore, some market participants may be able to infer some information about the securities lending market from the market for repos; this is particularly the case for securities with a very active repo market, such as U.S. Government securities. A fixed- term cash collateralized securities loan is economically similar to a repo. Consequently, an investor wishing to gather information about fixed-term loans in the securities lending market could use prevailing terms in the repo market to infer information about borrowing costs for the same securities. However, as majority of securities loans do not have a fixed term, information from the repo market would only be useful for a small subset of securities loans.

3. Characteristics of the Securities Lending Market

The value of securities available to be loaned generally far exceeds the total value on loan. The OFR Pilot Survey documents that in 2015 only about 10 percent of the value of securities available for lending were on loan. However, for a specific security it is not always the case that shares available to loan far exceeds shares on loan. For some securities, particularly highly shorted securities, it can be extremely difficult and expensive to find securities to borrow. Securities that are difficult to borrow are said to be “on special” and can have average borrowing costs many times higher than a security that is not on special.

Securities loans also exhibit a wide range of borrowing costs for the same security on the same day. The range of borrowing costs for the same security can be influenced by a number of characteristics, such as the creditworthiness of the borrower, the type of collateral used, and the terms of the loan. As discussed in further detail below, the range of borrowing costs likely also represents asymmetric information between the parties to the loan negotiation, such that one party is able to charge a higher price than would be possible if the other party had better information regarding the current loan prices for that security. It may also represent a general lack of price efficiency stemming from incomplete information, as market participants operate without a clear view of the market as a whole.

Table 1 provides descriptive statistics illustrating these characteristics of the securities lending market. The data come from Fidelity National Information Services, Inc. (FIS) and reflect conditions in the Wholesale market for U.S. equity loans on the same days as the OFR Pilot Survey, for the sample of lenders that provide data to FIS.

Panel A of Table 1 provides the distribution of utilization rates (defined as the percent of shares currently on loan relative to the total number of shares available for lending). This panel highlights that utilization rates are highly positively skewed. For most stocks supply significantly outstrips demand with median utilization rates of approximately 12 percent. For stocks at the 90th percentile, utilization rates are near 70 percent, implying that an investor seeking to find shares of such a stock to borrow may have a difficult time doing so.

Table 1 only presents utilization rates for equity lending transactions. However, at least one commenter states that utilization rates vary significantly across asset classes, stating that data from the OFR Pilot Survey show that utilization rates are lower for more individualized securities, such as corporate bonds. Meanwhile, referencing the OFR Pilot Survey, the commenter pointed out that the utilization rate for U.S. Treasuries and agencies is higher than that of equities.

Panel B of Table 1, which presents borrowing costs in term of lending fees, shows that the lending fees paid for equity loans exhibit a wide range. Stocks that are on special can have fees many times higher than the median stock. Specifically, stocks at the 90th percentile of lending fees have an average lending fee of 7 percent per year while the median stock has a lending fee of about 0.6 percent per year. Even when loans involve the same stock, and on the same day, there can be a significant range in fees paid to borrow securities.

Panel C of Table 1 highlights the range of lending fees charged for the same stock on the same day. The range in fees is defined as the difference in the maximum and minimum fees reported to FIS for loans of the same stock on the same day. This range can be quite substantial. For the median stock the range is about three percentage points, or approximately five times the median fee charged for securities lending transactions.

One commenter argued that the Proposing Release offered “no evidence to suggest that lending fees broker-dealers charge are so meaningfully different (after controlling for various commercial factors).” As the Commission acknowledged in the Proposing Release and continues to acknowledge here, there are a number of factors that can drive variations in borrowing costs including, e.g., the creditworthiness of the borrower. However, the Commission is unaware of the existence of data that would allow for a full analysis of all possible factors contributing to borrowing costs, and the commenter did not provide any roadmap for performing such an analysis accurately. That the observed range in borrowing costs is likely driven, at least in part, by asymmetric information and a general lack of price efficiency in the securities lending market is supported by academic literature, which finds that asymmetric information drives dispersion in prices, particularly in opaque markets with limited transparency. One implication of the results from this literature is that price dispersion should be higher when transparency is lower. Indeed, there is evidence that the dispersion in borrowing costs is higher in securities lending markets with less transparency, in which information asymmetry is expected to be higher. For example, the OFR Pilot Survey shows that the range of lending fees is higher in the U.S. equity lending market than it is in the U.S. Treasury/Agency lending market, where transparency is arguably higher.

Furthermore, the Commission views it as unlikely that one of the pricing factors frequently mentioned by commenters, namely the creditworthiness or counterparty risk of the borrower, can fully explain the large range in borrowing costs. Table 1 shows that the median range of fees is about 400 percent larger than the median fee itself—implying that some investors were paying at least four times as much to borrow the same security on the same day as other investors. Based on staff experience, securities loans are virtually always fully collateralized, and so counterparty risk is unlikely to be sufficient to explain such a large dispersion in borrowing costs.

Other commenters expressed support for the notion that price dispersions in the securities lending market were due to information asymmetries and a lack of informational efficiency. While at least one commenter questioned the need to improve transparency in the corporate bond and U.S. Treasury/Agency lending market due to structural differences between those markets and the equity market, there is evidence that there is a dispersion of borrowing costs within these asset classes as well. Information from the OFR Pilot Survey indicate a large dispersion of lending fees and rebates across all asset classes, including for U.S. Treasury/Agency and corporate bond securities. For the reasons discussed above, the Commission believes that the observed variation in borrowing costs in these markets is likely also at least partially driven by asymmetric information and a general lack of price efficiency in the securities lending market.

4. Structure of the Securities Lending Market

The securities lending market is made up of borrowers, lenders, and intermediaries that can facilitate a transaction. End borrowers can borrow securities either through their broker-dealers, or by themselves if they maintain their own relationships with lending programs. If they borrow through their broker-dealer, then they transact in the Customer market. If they maintain their own relationships and borrow directly from lending programs, then they transact in the Wholesale market. Beneficial owners can either supply shares to the lending market by contracting with a lending program, or they can run their own lending program and lend directly to entities such as large hedge funds with which they maintain relationships. In either case, such a transaction occurs in the Wholesale market. Lenders can also be broker-dealers who lend to end borrowers. These lenders transact in the Customer market.

A market participant such as a short seller wishing to borrow shares usually does so through its broker-dealer, who offers to find shares to borrow as part of its suite of services offered to customers. A broker-dealer may start by providing a security loan to its customer with shares from its own inventory or out of another customer's margin account. The Commission understands that in order to facilitate the amount of borrowing customers wish to do, a broker-dealer will sometimes have to find other sources of shares. To that end, broker-dealers maintain securities lending relationships with customers in fully paid accounts as well as relationships with various external lending programs.

Additionally, some large institutions, such as banks, credit unions, pension funds, and hedge funds, choose to maintain their own relationships with lending programs. These entities bypass broker-dealers to search for borrowable shares themselves. This option is not feasible for smaller institutions, who lack both the scale to make it cost effective, and the creditworthiness to be an acceptable counterparty for the lending programs in the absence of an intermediary, e.g., a broker-dealer.

The OFR Pilot Survey estimated that there were approximately $1 trillion of shares on loan. The OFR primarily focused on the Wholesale market, and consequently the overwhelming majority of borrowers were broker-dealers; the OFR Pilot Survey does not provide much insight into who the end borrowers are for securities lent by broker-dealers. Figure 1 provides the fraction of total securities on loan by type of borrower based on the OFR Pilot Survey across three trading days in 2015.

Source: OFR Pilot Survey of Agent Securities Lending Activity

There is currently no common source that those seeking security loans can use to find shares available to lend, which is why broker-dealers rely on relationships with lending programs to secure loans. This situation has contributed to high search costs in this market. High search costs imply that transactions cannot take place without a costly effort to find a favorable counterparty. The need for such costly effort can inhibit market efficiency.

One commenter stated that the level of opacity in the securities lending market does not lead to high search cost or inhibit the efficiency of the securities lending market, because the level of detail in data provided by data vendors has improved drastically, the number of data vendors in the market has been growing, and the vast majority of parties in securities lending markets have access to one or more sources of data. While these facts are generally associated with declining search costs and increased efficiency, they do not, by themselves, imply low search costs or an efficient market. As discussed above, the Commission continues to believe that the datasets provided by commercial data vendors are characterized by incomplete and potentially biased data. As a result, market participants that want information on whether a given loan price or quote is competitive, or how a given lending agent or broker-dealer is performing relative to a benchmark, may need to consult multiple datasets, if they have access to such data, or request and receive multiple quotes in order to confirm that a given quote is consistent with market conditions. A need to consult multiple sources of information implies high search costs. When the cost of acquiring additional information is high, it may not be economical to acquire the information, resulting in less efficient terms for loans. The Proposing Release also cited academic research characterizing the securities lending market as having high search costs, which was not rebutted by the commenter. Consequently, the Commission continues to believe that, despite the availability of commercial datasets, opacity in the securities lending market contributes to high search costs and inefficiencies in that market.

Broker-dealers possess some market power over their customers. Generally, broker-dealers assist investors in finding shares to borrow as part of a suite of services and the cost of switching to a new broker-dealer can be high. This relationship can make it difficult for investors to change broker-dealers if they underperform in one area because it is not just a securities lending relationship that would be changed, but the whole suite of broker-dealer services that would be affected.

Additionally, the relationship nature of the lending market favors larger broker-dealers who can maintain high-volume relationships with more lending programs. The limited availability of comprehensive data makes it difficult for customers to evaluate the performance of broker-dealers. Customers as well as lenders thus rely on relationships and reputation, a situation that also tends to favor larger broker-dealers.

The primary sources of shares to loan in the Wholesale market are long-term investors such as investment firms, pension and endowment funds, governmental entities, and insurance companies. These entities generally make their shares available to lend either through a lending program run by a lending agent or by running their own lending program. Broker-dealers borrow shares from lending programs in order to facilitate clearing and settlement on a net-basis, though they may additionally source shares for this purpose from their own inventory, from fully paid shares, and from customer margin accounts.

As described above, a beneficial owner seeking to lend shares will generally provide those shares to a lending agent, who runs a lending program. There are two broad categories of lending programs: custodian banks and third-party lending programs. In the case of custodian banks, the lending program is generally offered as part of their general custodian services.

Both types of lending programs will generally pool shares across accounts with which they have lending agreements to create a common pool of shares available to lend. As shares are lent out, the revenue earned from the pool of shares is generally distributed across all accounts contributing shares to the pool of shares on loan on a pro-rata basis. Pooled lending programs generally split the fees generated from lending with the beneficial owners. Based on the staff's experience, the Commission believes that the lending program will usually take about a third of the fees earned. In the case of custodian banks, the custodian bank may, rather than return the lending revenue directly to the beneficial owner, instead apply the beneficial owner's portion of the lending revenue to other fees charged by the custodian bank for other services.

Lending programs typically indemnify the beneficial owner from default by the borrower. This indemnity gives the lending program an incentive to ensure the creditworthiness of the borrower, and a lending program may assess higher fees to borrowers it deems as less creditworthy.

Over the past two decades, auction-based security lending has become an alternative for lender-borrower interactions. In this setting, unlike with the directed lending programs, positions of different beneficial owners are not pooled to cater to security-specific demand from borrowers. Instead, after determining the desired income streams, the lender's entire portfolio, or its segments, are offered via blind single-bid auctions.

In some cases, a beneficial owner may choose to set up its own lending program. This course is more common among very large funds that have the resources to build up the expertise necessary to operate a lending program.

The current relationship and network structure of lending programs and broker-dealers favors larger lending programs that have the resources to maintain relationships with more and larger broker-dealers. Thus, the Commission believes that the market for lending services is likely dominated by a few large lending programs, including those run by the large custodian banks.

The OFR Pilot Survey estimated that as of the latter part of 2015 there were approximately $9.5 trillion worth of shares available for lending. Figure 2 provides a breakout of the percent of shares available for lending provided by the various entities.

Source: OFR Pilot Survey of Agent Securities Lending Activity

5. Structure of the Market for Securities Lending Data and Analytics

The market to collect and disseminate securities lending data is an outgrowth of the market for securities lending market analytics. This market consists of a few established vendors that specialize in geographic areas ( i.e., U.S. and non-U.S.) but seek to compete in all geographic areas. Most vendors collect the data to support the analysis business in which they provide data-based service to institutions and other lending programs. Others collect data through their facilitation of security loans. As such, the data vendor business is often an outgrowth of another business. The Commission understands that the current practice by commercial data vendors is to provide preliminary statistics on the same day based on the intraday data collected by the vendors, while the main data are disseminated one day later.

The Commission believes that the data provided by the various data vendors operating in the give-to-get Wholesale data market are largely comparable. However, the entities providing data to the vendors are also their customers. This relationship limits the market power of the vendors with respect to their clients who provide data but results in the clients' incentives limiting the competitiveness of the market. This results in the market being largely inaccessible for many entities that could use the data for their own benefit or the benefit of the market as a whole.

The give-to-get model for securities lending data is a significant barrier to entry to any firm seeking to provide analytics services regarding Wholesale loans. Firms cannot provide analytics services without data, and the biggest three data vendors have established relationships with data contributors to collect data. Such data contributors have an incentive to also control who can access that data. Consequently, the Commission understands that the market for securities lending data and securities lending analytics in the Wholesale space is largely concentrated among the three biggest data vendors.

One commenter stated that the give-to-get model is “is not the industry standard” in the securities lending data market and pointed out that there are some commercial securities lending datasets that are available to all subscribers. The Commission understands that the collection of these data largely relies on surveying Customer market participants about their borrowing experiences. Data vendors then aggregate and sell these data, or derivative products that combine the survey data with other data sources to provide derived metrics in areas such as short selling. For example, the Commission staff understands that, based on interactions with vendors of such data, that the customer market survey data are used largely as an input to proprietary algorithms that combine information from multiple data sources—such as FINRA's bimonthly short interest—to produce a short selling metric that is then used by consumers. Building relationships with a large number of Customer market participants is a significant barrier to entry and thus the Commission understands that data provision and analytics regarding customer market survey data is concentrated among the top two biggest data vendors.

6. Short Selling Transparency

Multiple commenters stated that granular securities lending data generally provide information about investors' short selling positions and strategies. This section provides information about existing sources of short selling data to facilitate an analysis of the impact of the Rule on short selling transparency.

Securities Lending Data

Because of the need to borrow securities to facilitate a short sale, securities lending data can be used to gain insight into short interest and short sentiment, and market participants use both give-to-get Wholesale data and customer market survey data on Customer loans to gain insights into short selling. For example, some academics have used securities lending data to study short selling in published research. Additionally, while access to Wholesale market data is generally limited, the Commission understands that some third-party data providers combine Wholesale market data with other sources of short selling data, such as bimonthly short interest data, to provide retail clients with estimated short selling metrics. The Commission also understands that commercial data vendors use securities lending data as an input to proprietary algorithms that combine information from multiple data sources—such as biweekly short interest—to produce a short selling metric that is then used by consumers.

Data on Wholesale securities loans provide only a noisy approximation of short interest at the security level. This is because Wholesale loans are made largely to facilitate clearing and settlement on a net basis at a clearing broker, rather than by transaction or position. Thus, Wholesale loans are not traceable to individual short sellers. Further, the Commission understands that broker-dealers will usually source shares to meet their net clearing and settlement requirements from other sources before engaging in Wholesale loans.

The Commission believes that information about Customer loans is more likely to be informative about short selling activity than information about Wholesale loans. This is because, unlike Wholesale loans, Customer loans are used to facilitate individual short positions. The size of a Customer loan when first created is the same as the size of a short position as of the end of the day on which the short position was created. Should the customer increase or decrease the size of their short position, the Commission understands that the original loan would be modified accordingly. Thus, individual Customer loans can be tracked to reveal the size and changes in individual short positions. However, existing Customer market datasets, like Wholesale market datasets, lack comprehensiveness.

Most commercially available datasets are currently produced with a one-day lag. As a result of the difference in settlement cycles between the equity and securities lending market, there is therefore typically a three-day lag between the availability of the securities lending data and the change in short interest that corresponds to the securities loans.

Bimonthly Short Interest Data From FINRA

One of the primary data sources for information about aggregate short positions is the bimonthly short interest data collected by FINRA. FINRA collects aggregate short interest information in individual securities on a bimonthly basis as the total number of shares sold short in a given stock as of the middle and end of each month. In the case of U.S. exchange-listed stocks, FINRA shares the data with the listing exchange. FINRA then publishes short interest data for all exchange-listed and OTC equity securities on its Equity Short Interest Data page free for the investing public. Specifically, aggregate positions in each security are provided for publication on the seventh business day after the reporting settlement date.

FINRA computes short interest using information it receives from its broker-dealer members pursuant to FINRA Rule 4560 reflecting all trades cleared through clearing broker-dealers. FINRA Rule 4560 generally requires that broker-dealers that are FINRA members report “short positions” in customer and proprietary firm accounts in all equity securities twice a month through FINRA's web-based Regulation Filing Applications (RFA) system. FINRA defines “short positions” for this purpose simply as those resulting from “short sales” as defined in Rule 200(a) of Regulation SHO under the Exchange Act. Member firms must report their short positions to FINRA regardless of position size.

These short interest data are widely available and are used by academics and other market participants. These short interest data are found to predict future stock and market returns over the monthly and annual horizons, suggesting that the bimonthly short interest data capture short selling strategies based on fundamental research. However, the transparency offered by these data is limited in at least two ways. First, the information content is delayed by at least seven business days, and also does not provide insight into the timing with which short positions are established or covered within the two-week reporting period. This precludes the possibility of understanding the behavior of aggregate short selling in the two weeks leading up to the reporting date of the positions. Second, given that the data are aggregated at the security level, they prevent the public from understanding certain aspects of the underlying proprietary short selling strategies. For example, the data can't inform on whether short positions are distributed across a large or small number of market participants.

Short Selling Volume and Transaction Data From SROs

This section discusses data that exist on daily and intraday short volume data, which provide different information relative to the bimonthly short interest data discussed above. The bimonthly short interest data provides an aggregate measure of outstanding short interest whereas the data discussed in this section provides information about the flow of short sales in the market. It is less useful in determining total short interest outstanding because it is not accompanied by data regarding when short sales are covered but allows market participants to observe the actual trades of short sellers.

Since 2009, many SROs have been publishing two short selling datasets, including same-day publication of daily aggregated short sale volume in individual securities and publication of short sale transaction information with a delay of up to two months. Some SROs make the historical daily short volume data available to market participants at a fee.

Despite offering higher granularity, there is a concern that the SRO short volume data may over-represent the total volume of short sales occurring in the market. This is because Regulation SHO provides specific criteria regarding what is a long sale. If a market participant is unclear whether their trade would meet all the requirements at settlement to be marked a long sale, then they may choose to mark the trade as short to not run afoul of Regulation SHO requirements, even if the trade is likely economically equivalent to a long sale. For instance, the market participant may be deemed to own the security but does not reasonably believe they can deliver the security in time for settlement, and thereby must mark the order as short regardless of their ownership of the security. Aggregate short selling volume data and short selling transactions data typically have different lags with which they are available, though aggregate short selling volume data can be disseminated as early as the same day of the short sale.

B. Economic Effects of the Final Rule

1. Benefits of Increased Transparency in the Securities Lending Market

The Commission believes that the primary impact of final Rule 10c–1a will be to increase transparency in the securities lending market through improvements to the comprehensiveness, breadth, accuracy, and accessibility of securities lending data. These impacts will reduce information asymmetries in the securities lending market and improve informational efficiency leading to a more efficient securities lending market.

Final Rule 10c–1a will reduce information asymmetries and improve informational efficiency in three ways. First, the data provided by final Rule 10c–1a will be more comprehensive than the data currently offered by commercial data vendors because it will not rely on voluntary data submissions but will result from mandated disclosure from both the Wholesale and Customer segments of the market. Second, Rule 10c–1a data will improve informational efficiency by including certain data fields that are not currently offered by commercial data vendors, contributing to the breadth of available securities lending data. In addition, Rule 10c–1a data will contain detailed security loan modification information. Third, the final rule will expand the accessibility of the data by allowing all market participants to access Rule 10c–1a data, possibly subject to a fee.

The increased transparency from final Rule 10c–1a will result in several notable economic benefits. First, the final rule will reduce information asymmetries, which will benefit investors by reducing borrowing costs, increasing price efficiency, and increasing competition between broker-dealers and lending programs. Second, by reducing the borrowing costs for some securities, the final rule will lower the cost of short selling these securities and improve market efficiency. Third, improvements in the information available to various market participants will lead to a variety of benefits for these participants, including increased profits and reduced costs of business due to easier access to information and more informed decisions.

Reduction in Information Asymmetry

The Commission believes that the transparency created by final Rule 10c–1a will reduce information asymmetries between more centrally connected securities lending market participants and those on the periphery. Specifically, it will reduce the information asymmetries between broker-dealers and end borrowers, and between beneficial owners and lending programs. This will result in increased competition between dealers and between lending programs, ultimately leading to better loan terms for some securities.

The Commission believes that the transparency created by the final rule will benefit end borrowers by reducing the information disadvantage they have with their broker-dealers when borrowing shares. Most security loans obtained by end borrowers are facilitated through broker-dealers. Rule 10c–1a data will allow end borrowers to determine the extent to which their broker-dealers are obtaining terms that are better, worse, or consistent with current market conditions for loans with similar characteristics. It will facilitate this comparison by providing comprehensive transaction-by-transaction information about the cost to borrow and other loan characteristics that are currently mostly unavailable to end borrowers. For example, end borrowers will be able to compare the lending terms provided by their broker-dealers to the terms of similar loans, such as loans with the same collateral and borrower types. End borrowers will also be able to compare loans of similar sizes, though with a delay of 20 business days. The Commission believes that, despite the delay in dissemination, loan size information will still be useful for end borrowers, who can use this information to compare the prices of their loans to other loans with similar terms ( e.g., size, collateral type, loan type, etc.), that were effected approximately one month prior. Additionally, to the extent that an RNSA publicly disseminates daily volume-weighted cost-to-borrow statistics as part of its requirement to disseminate daily aggregate information, these statistics will provide end borrowers with access to information about loan prices that incorporates information about loan size information even prior to the public dissemination of loan sizes.

Similarly, the Commission believes that the final rule will benefit beneficial owners by reducing their information disadvantage with respect to their lending programs. By allowing beneficial owners to more easily benchmark their lending programs through access to data on loan prices and other characteristics of recently transacted security loans, as well as the statistics provided by an RNSA on the next business day, the final rule will provide beneficial owners with an improved ability to determine the quality of the loans that their lending program executes on their behalf relative to other loans with similar characteristics. The ability to view the cost to borrow in both the Customer markets and Wholesale markets will also provide beneficial owners with additional information that they can use to benchmark the performance of their lending programs.

Financial institutions such as banks and broker-dealers use the securities lending market in order to manage collateral needed for other transactions. These entities can face the same asymmetric information concerns as do end borrowers and beneficial owners, and thus an increase in market transparency may lead financial institutions to receive better loan terms, and thus improve their ability to manage collateral. Banks also borrow securities to manage their balance sheets, and the Commission believes that this too will become easier as a result of the final Rule 10c–1a due to a more competitive lending market, leading to the benefit of improved balance sheet management by banks.

At the same time, as access to securities lending data increases, competition for securities lending analytics may increase as well. Increased competition for and availability of securities lending analytics would further reduce information asymmetries by reducing the cost of and increasing access to securities lending analytics for both end borrowers and beneficial owners.

If an end borrower or beneficial owner believes that a particular broker-dealer or lending agent is consistently underperforming, the final rule will provide that market participant with the tools to identify such underperformance and address it with their broker-dealer or lending agent, or to find a new broker-dealer or lending agent. For example, a beneficial owner or end borrower could use the transaction-by-transaction Rule 10c–1a data to create a bespoke benchmark of recently transacted loans that are similar to a loan they wish to effect, and compare this benchmark to the terms offered by their lending agent or broker dealer for that loan. A beneficial owner or end borrower could also use the daily information disseminated by an RNSA about the distribution of loan prices to assess how the terms offered by their broker-dealer or lending agent compare more generally to the distribution of recently transacted loans in the same security. An increased ability to compare loan performance will increase competition between broker-dealers and between lending agents, which will ultimately improve lending terms for both end borrowers and beneficial owners.

The Commission expects improved terms to be concentrated among securities that are “on special” and have higher borrowing costs. This is because general collateral securities lending is a low margin business and lending supply for these securities far outstrips lending demand. Combined, these two factors mean that there is likely not much room for fees to improve for general collateral securities. As securities become on special ( i.e., harder to borrow), lending rates and utilization rates increase. Higher utilization rates mean that there are fewer shares available to borrow and lend. In this environment, high search costs and asymmetric information are significant contributors to both the level and dispersion of fees, and reducing asymmetric information has a better chance of leading to better terms. Beneficial owners that currently lend shares of hard-to-borrow securities for rates that are consistently below the market average could receive higher lending rates for these securities as their lending agents both have better access to information and become more accountable for their performance due to beneficial owners also having access to the same information. Similarly, end borrowers whose borrowing costs are currently higher than the market average for hard-to-borrow securities may see their borrowing costs decrease as both their broker-dealers' information access and their ability to monitor to their broker-dealers improve. These two effects suggest that the dispersion of borrowing costs for hard-to-borrow securities will diminish.

That transparency can result in improved market quality is consistent with the experience in other markets. The implementation of TRACE improved transparency in the corporate bond market, and research has shown that TRACE lowered the average cost of transacting and increased competition between dealers in this market. Additionally, recent research from Brazil has shown that improving securities lending transparency led to lower fees, increased liquidity, and increased price efficiency in that country. In both cases, researchers have identified reduced information asymmetry as a key mechanism leading to improved market outcomes.

While some commenters supported the applicability of evidence from TRACE, other commenters questioned the applicability of evidence from TRACE and from the Brazilian policy change. Specifically, commenters argued that, because bond transactions are irrevocable and fungible, the identity and characteristics of the counterparty are less relevant to prices in bond transactions than in securities loans, and thus the Commission's reliance on TRACE studies to draw inferences about the securities lending market is misplaced. At least one commenter disputed the Commission's use of the Brazilian policy change to draw inferences about the potential benefits of proposed Rule 10c–1 because the Brazilian policy “did not involve the trade-by-trade disclosure of stock lending terms” and is thus not comparable to the Rule. The Commission continues to believe that both events provide meaningful information about the potential impacts of the final Rule 10c–1, as evidence from both events illustrates how a reduction in asymmetric information between market participants has the effect of improving market quality.

First, the introduction of TRACE, which introduced transaction-by-transaction transparency into a market that previously did not have such transparency, represents a policy change that is markedly similar to that under the final Rule 10c–1a. The Commission acknowledges that the market setting for TRACE is different from that of final Rule 10c–1a, for example, because of the importance of counterparty risk in securities loans prices as compared to prices of corporate bonds. However, the fact that corporate bond prices are less sensitive to the identities of the counterparties than securities loans does not invalidate using TRACE to illustrate the effect of decreasing information asymmetry. Instead, it simply implies that the baseline dispersion in corporate bond prices prior to TRACE was likely less than it currently is in the securities lending market.

Second, while the Brazilian study takes place in a similar market setting as Rule 10c–1a ( i.e., the securities lending market) it is true that the policy details differ. In particular, the Brazilian policy change involved an increase in transparency not through the creation of a transaction-by-transaction report, but through an increase in the informativeness of an available loan fee “benchmark.” However, despite the differences in details, the policy had a similar target to that of Rule 10c–1a: namely, an improvement in the transparency of the securities lending market. The Brazilian study shows that this increase in transparency improved market quality through the economic channel of reduced information asymmetry. This study is thus directly relevant for final Rule 10c–1a, as the Commission expects that many of the economic effects of the rule will be realized through the same economic channel ( i.e., a reduction in information asymmetry).

Additionally, in support of their concern over the applicability of TRACE research, one commenter stated that, “while there is no reason for the supply of corporate bonds to decrease in response to increased trade transparency, greater transparency in the securities lending market may well reduce the lending supply.” This commenter cited industry research showing that increased short selling transparency can decrease beneficial owners' willingness to lend shares, as well as academic research calling into question the applicability of TRACE to securities lending “since greater loan fee transparency could reduce the lending supply.”

The Commission believes that the industry study cited by the commenter has empirical limitations that make it difficult to extract robust empirical conclusions from the study. These limitations are acknowledged by the author of the study; specifically, the author states that, during the sample period, which occurs during the height of the 2008 financial crisis, “equity markets were in a relatively disordered state,” and that “such extreme change in markets has the consequence of making it complicated to establish a control group of stocks such that all variables, except regulatory variables, remain constant. This, in effect, makes it difficult to attribute causality solely to the regulatory variables.” The Commission agrees that the extreme market volatility during 2008–2009 increases the likelihood that the decrease in beneficial owners' appetite to lend out shares documented by the author may have been driven by other market events concurrent to but distinct from changes to regulatory regime for short selling and, as a result, “causality is still difficult to prove.”

The Brazilian study, also cited by the commenters, occurred during a less volatile period rendering potentially cleaner empirical results. The authors of that study indeed suggest that the loan supply could decrease “if the lower loan fees received by lenders were not sufficient to cover the potential losses from not selling a stock when shorting activity increases.” However, later in that same paragraph, the authors reject this hypothesis and conclude that their results were not consistent with increased transparency affecting liquidity in the lending market.

Improved Market Quality Due to Lower Short Selling Costs

The Commission expects that final Rule 10c–1a will produce countervailing effects through its impact on short selling, but ultimately that the net result will be positive. On one hand, the final rule will benefit short sellers by lower borrowing costs for securities that are hard to borrow, which will improve market quality through increased price efficiency, managerial oversight, and liquidity. On the other hand, the final rule could potentially harm market quality by making it easier for other investors to discern short sellers' trading strategies, thereby discouraging the costly fundamental research that underlies some short selling strategies. On balance, the Commission expects that the final rule, which delays the dissemination of loan volume information by 20 business days, is not likely to significantly expand market participants' abilities to discern short selling strategies.

The Commission expects that final Rule 10c–1a will lower short selling costs by decreasing borrowing costs, especially for hard-to-borrow securities. Academic research indicates that, when short selling costs diminish, investors increase their fundamental research because it is easier to trade on any negative information that they uncover ( e.g., poor earnings). This increase in fundamental research may in turn lead to better investment decisions by these investors. Additionally, by facilitating more research, the final rule will benefit market participants by improving price discovery. Academic research shows that short sellers, through their research, contribute to price efficiency by gathering and trading on relevant private information.

Short sellers also serve as valuable monitors of management. Extant research has demonstrated that when management knows that short sellers may be studying their firms, they are less likely to engage in inappropriate or value-destroying behavior. Research also indicates that when short selling becomes easier the effectiveness of short sellers as monitors increases.

Reducing the costs of short selling may also have the benefit of increasing liquidity in the underlying securities markets. Short sellers are key contributors to liquidity in both equity and options markets and existing research shows that, when short selling is constrained by tightness in the securities lending market, the stock market is less liquid. Lower short selling costs could have potential benefits for options market liquidity as well. Securities lending affects liquidity in the options market through its impact on how easily options market makers can delta hedge. Less costly delta hedging may therefore increase liquidity in the options market. Also, since some price discovery occurs in the options market, to the extent that the final rule increases the ease with which investors can trade in options, the proposal may further enhance price efficiency in the options market.

Commenters to the proposed Rule 10c–1, which did not include a delay in the dissemination of any information that would be collected under the rule, expressed concern that the dissemination of Rule 10c–1a data could harm short sellers, and thus U.S. financial and capital markets in general, and that this outcome was not analyzed in the Proposing Release. Some commenters expressed concern that information about Customer loans collected under the rule in particular could be used to determine individual short positions with a high degree of accuracy, and that this could result in harm to short sellers, resulting in negative outcomes for market quality, such as reductions in liquidity, price efficiency and shareholder engagement.

The Commission acknowledges these commenters' concerns about the effect of proposed Rule 10c–1 on short sellers, and believes that these concerns are mitigated by the final rule's inclusion of a delay in disseminating loan size information. Specifically, the Commission acknowledges that activity in certain segments of the securities lending market are tightly linked to short selling positions, in particular the market for Customer loans. Thus, the sum of loans identified in Rule 10c–1a data as being to “a customer (if the person lending securities is a broker or dealer)” could give a strong indication of aggregate short interest. However, under the final rule, this information will only be directly available to Rule 10c–1a data consumers after a delay of 20 business days, which significantly reduces the novelty of this information compared to the proposed rule. In particular, this delay means that the loan size information, which is the portion of the data most directly related to short selling activity, disseminated under final Rule 10c–1a will generally be less timely than pre-existing sources of short selling transparency, such as FINRA's bimonthly short interest data.

Additionally, the Commission does not believe that market participants will have more timely access to indicators of aggregate short interest from the daily information disseminated by an RNSA pertaining to the aggregate transaction activity for each reportable security. This is because the term “aggregate transaction activity” refers to information pertaining to the absolute value of transactions, and excludes information that could reveal net transaction activity information. Therefore, it would not be possible to use this information to discern information about, for example, changes in net short sale positions.

The loan amount data disseminated under final Rule 10c–1a will provide some novel information about short positions, although only on a look-back basis after the 20-business-day dissemination delay has lapsed. In particular, since each loan likely relates to a unique market participant, the public dissemination of such transaction-by-transaction volume data under final Rule 10c–1a, albeit delayed, will provide an indication of the distribution of short sentiment, i.e., whether short interest is concentrated among a few short sellers with large positions, or whether it is spread out over many short sellers. It can also give an indication about when individual market participants increased or decreased their short positions by examining the change in the size of a loan from the reported data. As most currently available sources of short selling information only contain information about aggregated short selling activity, this could represent an increase in the granularity of market participants' information about past short positions and, in this way, could provide information about short sellers' strategies. Specifically, market participants could examine the historical securities lending data to try to identify factors that may be indicative of short selling activity. However, it is not clear that such an analysis, which would be inherently noisy, would provide actionable insights into future short selling activity that could harm short sellers' abilities to profit from negative information.

Economic Effects From Improved Securities Lending Data Quality

The Commission believes that final Rule 10c–1a will increase the information about the state of the securities lending markets that is generally available to market participants. As discussed in this section, increased information will result in benefits in the form of better decision-making by investors, beneficial owners and other market participants, reduced costs of business for broker-dealers, improved performance and reduced costs for lending programs, new business opportunities for data vendors, improvements to shareholder monitoring and, ultimately, improved market stability and price discovery both in the securities lending market and the market for the underlying security.

Overall, the data provided by the final rule will significantly improve market participants' views into lending rates for various securities. It does so first through the comprehensiveness of the data. By mandating the disclosure of certain information about all securities loans, the data provided by the Rule will be comprehensive and thus not prone to biases that occur due to non-random observations. Comprehensive reporting also means that the data provided by the final rule covers both the Wholesale and Customer segments of the market, allowing market participants to compare trends in both markets simultaneously. The data will also be more granular than existing data. In addition to information about the size of the loan and the loan fee, the data provided by the final rule contains a number of other data fields which allow market participants to parse the data to analyze the subset of transactions that are most relevant to their needs.

There are a few dimensions where data provided by commercial data providers may be similar to or provide information not provided by the Rule. The data provided by the final rule is disseminated next day, which is similar to the timeliness of much of the data provided by the commercial data providers except that some commercial data providers offer subsets of intraday transaction data that is timelier than the data provided by the final rule. Additionally, the data provided by the Rule masks loan size for 20 business days. Consequently, existing data providers' information regarding shares on loan will be considerably timelier than the data provided by the Rule. Lastly, the Rule does not provide information about utilization rates, and so market participants wishing to observe utilization rates will need to maintain access to commercial datasets.

First, the improvement in securities lending rate information and the ability to create bespoke benchmarks will improve the quality of information that market participants rely on to make decisions regarding investment strategies that require borrowing securities and the cost of those strategies. An increase in the quality of information regarding the costs of borrowing a security may decrease risk and thereby increase the risk-adjusted profits of pursuing investment strategies that require borrowing securities, such as short sales. For example, prior to a short sale transaction, the end borrower will be able to get a better sense of the likely costs associated with such an investment strategy by examining the data regarding recently transacted securities loans, as well as information provided by an RNSA about aggregate transaction activity and cost to borrow.

Access to Rule 10c–1a data may also benefit investors by enabling them to make more informed decisions about whether to buy, hold, or sell a given security. This will occur as they will have increased certainty regarding lending rates and which securities are most likely to be profitably lent or not, leading to better risk allocations, investment decisions, and ultimately risk-adjusted returns. Extant research has demonstrated that securities lending data have information relevant to the prices of the underlying security. By making securities lending information both more granular and more accessible for market participants, final Rule 10c–1a will enable investors to utilize data to gain insights into the underlying security. As investors become more informed, their investment decisions are expected to improve.

Additionally, an improved view of current lending market conditions for various securities may help inform beneficial owners in making better decisions concerning which shares to make available for lending, potentially leading to more profitable lending. For example, to the extent that beneficial owners do not currently have a clear means of determining which securities have high average lending rates Rule 10c–1a data may alert them about securities with high borrowing costs, which would enable them to better optimize which shares in their portfolio to make available for lending. Furthermore, as a result of having better access to information about the securities lending market, financial institutions will be able to improve their collateral and balance sheet management.

Second, a clearer understanding of lending market conditions, and specifically lending market rates, facilitated by the dissemination of Rule 10c–1a data may benefit broker-dealers by decreasing the search costs incurred to obtain a locate in order to facilitate a short sale on behalf of a customer. The increase in transparency under the final rule will allow broker-dealers to better ascertain current prevailing lending rates for security loans with certain characteristics prior to calling lending programs to get competing quotes. Broker-dealers tend to find loans for their customers through the network of lending programs with which they have relationships, after they have exhausted their own inventory and customer margin accounts. Rule 10c–1a data will enable them to determine with greater ease whether a quote from a lending program is competitive. It is possible that new broker-dealers may choose to enter the market for lending services because of this reduction in cost, which may further increase competition between broker-dealers.

Third, final Rule 10c–1a will benefit lending programs by providing a means by which they may improve the performance of their lending. Rule 10c–1a data will provide lending programs with a source of securities lending market data that is more comprehensive than existing commercial data. With these data, the lending programs will have an improved ability to determine prevailing market conditions as they compete to lend shares, which may improve their lending performance.

Fourth, more comprehensive, granular, and accessible securities lending information may also lead to additional business opportunities for commercial securities lending data vendors. In addition to data products, commercial data vendors also provide analytics to their customers, and may be able to support these analytics data services with the data provided by the final rule. Further, because the commercial data vendors would be less dependent on their data providers for data, they may be able to provide analytics to more market participants. This may result in increased competition for data analytics services as the barriers to entry for providing analytics services decline and new entrants compete to provide analytics services. While this effect may lower what the data vendors can charge for analytics services, to the extent that the commercial data vendors offer their customers other securities lending services, such as execution services, the final rule may enhance their other business lines by providing more comprehensive data to support other securities lending market services. To the extent that there is a demand for covered persons to contract privately with third party vendors to assist in reporting under final Rule 10c–1a, this would likely also lead to additional business opportunities for commercial securities lending data vendors, as these vendors already have experience with handling and disseminating securities lending data and could leverage that experience to offer such services to customers.

Ultimately, the improved information access will result in improved market quality in the security lending market. More informed investment decisions facilitated by the final rule may improve market stability by allowing investors to better manage risk. Furthermore, as all participants in the securities lending market will be able to obtain better data on that market, utilize the insights contained in the data, and then improve their decisions based on it, the price discovery process will improve. This will lead to more efficient prices for securities loans. This improved information access may also improve price discovery in the market for the securities underlying the security loans, as information about the underlying security will have a greater opportunity to incorporate into the price of the underlying security.

The requirement to report all of the data elements under paragraph (c) the first time the modification of a “day-one” loan (that is, loans in existence prior to the final rule's reporting date for covered persons) occurs will avoid the exclusion of certain loan information during the early phases of implementation. This will enhance the extent to which Rule 10c–1a data improves informational efficiency during the earlier phases of implementation. In the Wholesale market lending largely facilitates clearing and settlement, and for commonly lent securities it is possible that a broker dealer could have a continuous need to borrow the security. In this case the broker dealer likely would enter into a loan that is kept open for an extended period of time, but with frequent size and/or rate modifications to match their ongoing settlement needs and market conditions. Therefore, absent such a requirement, all the terms of such a lending arrangement would not be fully reportable unless the existing loan is closed out and a new one is opened. It could take a considerable amount of time for this to occur and thus the data collected by the Rule would lack the full context for such loans until they were closed and then re-entered into. The lack of information about these loans would harm data quality because it would render the data less complete and would thus limit market participants' ability to determine conditions in the lending market and would thus mitigate the benefits described in this section. Ensuring that these loans are included upon their first modification improves the quality of the data in the earlier phases of reporting which would in turn speed up the time when the benefits articulated will become fully available to market participants.

Potential Limits to Benefits and Sources of Uncertainty

Commenters mentioned a number of factors that may limit the extent to which the rule increases transparency in the securities lending market, and thereby limit the associated benefits as well. However, despite these potential limitations, the Commission expects that final Rule 10c–1a will improve the transparency and efficiency of the securities lending market, which will improve market participants' access to information and lower their information asymmetry.

Commenters noted that loan fees are determined by a variety of factors, some of which are not directly included in Rule 10c–1a data. Though the Commission believes that the final rule will improve market participants' ability to compare loans, the Commission recognizes that the data provided will not include information about all of the factors that are relevant to the pricing of securities loans. As such, two loans may appear to be similar based on the Rule 10c–1a data but may not have the same fees due to factors not recorded in the data, such as the counterparty risk or the stability of the portfolio. Some commenters expressed concern that failing to provide all information relevant to pricing would lead the data to be either not useful, potentially misleading, or even harmful.

The Commission recognized in the Proposing Releases and continues to acknowledge that Rule 10c–1a data will not contain all information needed to perfectly price loans. However, compared to the baseline level of information available to market participants, the more granular, comprehensive, and accessible data provided by the final rule will improve market participants' abilities to compare loans and provide numerous benefits to market participants. The Commission does not believe that this information will be misleading or harmful because, as the commenters make clear, it is well understood in this market that the pricing of loans is determined by many factors. Knowing this, beneficial owners and end borrowers will be able to use this information to create, for example, an expected range of borrowing costs and use these data to facilitate conversations with their lending agent or broker dealer about why certain loans have prices that they do, or why a specific loan falls where it does in the distribution of similar loans. Further, to the extent that the distribution of borrowing costs provides information that is relevant to stock prices, this information alone could improve price efficiency in the underlying securities market.

One commenter questioned the utility of information about borrowing costs, asserting their belief that the cost to borrow is not as important to the execution of short sale strategies as whether there are shares available to borrow. The Proposing Release acknowledged that information on shares available is useful to markets. Rule 10c–1a data will not specifically provide information on shares available. Market participants seeking such information may need to contract with current commercial data vendors, if they can, for estimates of shares available and utilization rates. However, since borrowing costs are correlated with the ease of locating securities to borrow, the improved access to fee information provided by the final rule will provide market participants with increased information regarding the availability of shares to lend.

Some commenters expressed concern that the data may be too granular to be directly usable by market participants, and at least one commenter expressed concern that the final rule could increase information asymmetries if the complexity of the data are such that only the largest and most sophisticated market participants would be able to benefit. The Commission is mindful that the reports prepared according to the final rule will contain a large volume of statistical data, and as a result it may be difficult for some market participants to review and digest the reports. However, by requiring reporting entities to report transaction-level information in a uniform manner rather than aggregated data, the final rule will make it possible for market participants and other interested parties to make their own determinations about how to group securities or loans when comparing across loan transactions. Requiring more detailed data will also help ameliorate potential concerns about overly general statistics, or about the specific categorization of loans and selection of aggregated metrics, by allowing market participants and other interested parties to conduct their own analysis based on alternative categorizations of the underlying data. Should certain market participants not have the means to directly analyze the detailed statistics, third parties, such as commercial securities lending data vendors, likely will respond to the needs of investors by analyzing the disclosures and producing more digestible information using the data. Additionally, this concern is mitigated somewhat by the requirement that an RNSA provide information about the distribution of loan rates at the security level, which could be more easily used similar to current data models.

One commenter suggested that the inclusion of short positions data would lead to double counting and could confuse market participants. Short positions are not required to be reported, but loans from a broker-dealer to an end customer—such as might support a short position—will be marked in the data as loans to a “customer (if the lender is a broker-dealer).” The Commission does not believe that the inclusion of such data constitutes double counting because a loan from a broker-dealer to their customer ( e.g., to facilitate a short sale and where the broker-dealer is the lender) is a distinct economic activity from a loan that the broker-dealer may engage in as the borrower ( e.g., to meet their own clearing and settlement obligations). The Commission does not believe that such data will lead to confusion because these loans are marked in the data in a way that allows market participants to separate customer loans from other loans in their analysis.

One commenter suggested that including rebate data could lead to confusion because rebates are often fixed to benchmarks that can change from day to day, potentially leading to many, perhaps daily, modifications of the loan needing to be submitted and that the volume of these data could confuse market participants. The extent to which the commenter's concern is realized will be determined by how an RNSA chooses to structure the reporting of this variable. For instance, if an RNSA chooses to allow market participants to report a spread and a benchmark, then no modifications would be required to be reported from day to day unless there were a change in the negotiated spread or benchmark. However, if an RNSA chooses to require market participants to report the total fee, then market participants would be required to report changes to the fee if the benchmark changes, which could require daily revisions. However, the Commission does not believe that this will cause confusion. The Commission expects that market participants know that rebates can change regularly and thus revisions would not be unexpected. Further, gathering loan modification data is important to facilitate the accurate computation of statistics regarding the cost to borrow by an RNSA and other market participants.

At least one commenter suggested that some of the benefits of the Rule from reducing information asymmetries may be limited as a result of switching costs. The Commission acknowledges that the cost of switching to a new broker-dealer can be high, and that high switching costs may make it more difficult for some end borrowers to easily switch their broker-dealer for one that would offer them better terms on a loan. However, at the same time, by increasing end borrowers' access to comprehensive information about the securities lending market, final Rule 10c–1a will improve end borrowers' ability to determine which broker-dealers are offering better loan terms, as well as to negotiate better terms with broker-dealers. Both of these will increase the benefits to end borrowers from switching broker-dealers. Holding switching costs constant, final Rule 10c–1a could thus still result in more end borrowers finding it beneficial to switch to better-performing broker-dealers. Furthermore, as discussed in the Proposing Release, even for those borrowers for which switching broker-dealers would not be cost effective, the data would provide benchmark statistics that may enable smaller borrowers to select higher performing broker-dealers initially.

The benefits of the final rule may be limited if the rule results in a reduction in the supply of securities loans. A lower supply of securities loans, for example, could weaken or even counteract the rule's expected effect on lowering borrowing costs. As previously discussed, the Commission believes that the empirical evidence does not show an increase in securities lending transparency would result in a decrease in the supply of securities loans. However, another possibility that was pointed out by commenters is that the supply of securities loans could decrease if compliance costs are passed on to lenders and beneficial owners, who are then less willing or able to participate in the securities lending market as a result. The Commission recognizes that some market participants, including beneficial owners, may experience reduced revenue from securities lending as a result of the rule. However, the Commission expects that beneficial owners and lenders will for the most part benefit from the final rule, as a result of increased information about the securities lending market and reduced information asymmetry. This benefit, which could even encourage more beneficial owners and lenders to enter the securities lending market, would serve to mitigate an increase in cost and a subsequent decrease in lending, to the extent that it would occur.

One commenter stated that securities lending data would be less useful for corporate bonds and asset-backed securities (ABS) because, due to the “the greater diversity of bond characteristics, data from one set of bonds cannot be as easily used as benchmarks for others,” and as a result “there is little indication that data from corporate bond or ABS lending transactions would be sufficiently useful to justify the reporting costs.” As discussed above, the Commission acknowledges that the data provided will not include information about all of the factors that are relevant to the pricing of securities loans. It is possible that this limitation on the benefits may be higher for certain infrequently loaned securities, such as corporate bonds. However, the Commission believes that, even for less liquid lending markets, Rule 10c–1a will still reduce information asymmetries and improve pricing efficiency by facilitating better benchmarking than is currently possible, and thus will represent an improvement relative to the baseline similar to other assets discussed. For example, market participants will be able to use Rule 10c–1a data to pool information across similar securities and/or similar time horizons to create their own bespoke benchmarks for loan terms.

The commenter also stated that securities lending data would be less useful for U.S. Government securities because “the market for these products is already fairly transparent,” and thus “imposing reporting obligations would not provide sufficient additional data to justify the compliance burden.” As discussed in the Baseline, the Commission believes that information asymmetries in the lending market for U.S. Government securities loans are unlikely to be fully addressed by the currently available datasets, and thus this market will ultimately still benefit from an increase in transparency.

While the adoption of final Rule 10c–1a will lead to benefits from an overall increase in transparency, the Commission acknowledges that, in some areas, there is uncertainty as to how the information content of Rule 10c–1a data will compare to that of existing data. The requirement for an RNSA to disseminate daily information about aggregate transaction activity and distribution of loan rates will also provide market participants with information about the securities lending market. However, how the aggregate information disseminated by an RNSA will compare to the information available from current commercially available datasets, and the range of analyses that market participants will be able to perform using this information, will depend on the specific aggregate statistics that an RNSA chooses to publish.

The Commission bases its estimates on an RNSA providing daily information that is at least as informative as comparable statistics that are currently provided by the commercial datasets. This is because, consistent with section 19(b) of the Exchange Act, any changes to an RNSA rules required by final Rule 10c–1a, including its Rule 10c–1a information collection and dissemination practices, would have to be filed with the Commission pursuant to section 19(b) and Rule 19b–4 prior to implementation. Such rules would also be subject to public comment. This process offers market participants the opportunity to provide feedback on the potential specific statistical form of the aggregated daily information based on, e.g., their experience using information from commercial datasets. As such, the daily information provided by an RNSA could be at least as informative as those statistics provided by current data providers. Therefore, at a minimum, the Commission expects that the final rule will result in a positive benefit to market participants in the form of increased securities lending market transparency. This is because, while the aggregate information released by an RNSA may be the same or similar to what is currently provided by commercial datasets, the comprehensiveness and accessibility of the Rule 10c–1a data is expected to be better than that of commercial data, such that aggregate information produced using Rule 10c–1a data can be expected to be better than that produced using currently available commercial data.

2. Regulatory Benefits

Final Rule 10c–1a will improve upon current data sources by providing an RNSA ( i.e., FINRA) and the Commission access to securities lending information that identifies the parties to the loans, indicates when a broker-dealer loans its own securities to its customers, and indicates whether the purpose of such a loan was to close out a failure to deliver. Further, the improved access and comprehensiveness and reduced bias of the publicly available data will also accrue to FINRA and the Commission, as well as any other regulators using these data. This access will benefit investors by enhancing regulatory tools employed to promote fair and orderly securities markets. In particular, investors may benefit from improved surveillance and enforcement uses, market reconstruction uses, and market research uses.

Surveillance and Enforcement Uses

The party identities and purpose information may facilitate better surveillance by FINRA for regulatory compliance by its members, and may improve its ability to enforce such regulations. Additionally, FINRA will be able to notify another regulator as permitted.

For example, for FINRA, the information on whether the security is loaned from a broker-dealer's securities inventory to its customer may assist FINRA in determining whether a broker-dealer is charging lending fees or paying rebates commensurate with the market. Thus, beneficial owners and end borrowers, who engage in securities lending transactions, will be better protected against potential unfair pricing of securities loans by broker-dealers. In addition, FINRA can use the data more generally to assist in its surveillance of FINRA Rules 4314, 4320, and 4330 regarding securities lending and short selling that primarily intend to reduce information asymmetry in the securities lending markets. For instance, the final rule can help FINRA identify broker-dealers who tend to lend to or borrow from non-FINRA members to examine compliance with provisions of FINRA Rules 4314 and 4330 that entail agreement, disclosure, and other requirements for this activity. In addition, the information on how much borrowing particular FINRA members engage in can assist FINRA in identifying which broker-dealers to examine for compliance with FINRA Rule 4320, which contains short sale delivery requirements. These types of activities will better protect investors by helping to ensure that entities engaging in certain securities lending transactions are authorized to do so and are in compliance with applicable regulations. FINRA can also use the information to monitor when broker-dealers are building up risk, thereby protecting broker-dealers' customers against potential instabilities. FINRA can use data on the identity and activity of its members to provide an early warning with regard to the behavior of its members during a short squeeze.

Additionally, the information on whether the loan is being used to close out a fail to deliver is relevant to Rule 204 compliance. Importantly, being able to estimate the securities lending revenues and costs of particular participants may help to fine tune disgorgement estimations. The Commission and FINRA can also use Rule 10c–1a data to oversee broker-dealer compliance with Exchange Act Rule 15c3–3.

The Commission believes that the requirement pursuant to Rule 10c–1a to identify all parties to securities lending transactions with LEIs, if such parties have LEIs, will facilitate the Commission and FINRA's efforts to monitor securities lending. Reporting of LEIs by legal entity securities loan participants that have LEIs will help provide a more precise and consistent means of identification for loan participants as compared to using loan participant names and, if available, CRDs, IARD Numbers, or MPIDs. In that regard, obtaining an LEI requires that an entity's identity be verified by a third party upon issuance of the LEI and upon annual renewal of the LEI. Additionally, LEIs contain “Level 2” information about the linkages between the entities being identified and their parent and child entities, which can better enable Commission staff and market participants to understand the relationships between various firms with an eye toward potential aggregations of risk. Furthermore, requiring LEI disclosure for loan participants that have LEIs also will facilitate the linkage of data reported about securities loans with any relevant data, such as data on short sales or positions, from other sources. However, because the Commission is not requiring loan participants without LEIs to obtain and report them, the aforementioned benefits will only arise for reported securities loans in which participants have LEIs.

One commenter expressed concern that, to the extent that there is a wide variance in interpretation of data, the rule could result in “false positive” enforcement actions. Conversely, the Commission believes that improvements in securities lending data will enhance the ability of FINRA and the Commission to oversee the securities lending market and more efficiently detect potential rule violations, such as those described above. This will result in more targeted actions, which will benefit market participants by resulting in more efficient oversight.

Market Reconstruction Uses

Final Rule 10c–1a may help regulators reconstruct market events. For example, in January 2021, trading in so-called “meme” stocks led to many questions about securities lending being asked by lawmakers, investors, and the media as well as calls by some for increased regulation in some areas. Rule 10c–1a data will allow for more detailed evaluations of such events in the future than was possible with existing data during January 2021. For example, January 2021 information on market participants' securities lending activity would have provided FINRA and Commission staff a more timely and comprehensive view of who was entering into new loans and who was no longer borrowing securities. This would have facilitated a deeper understanding of how the events were or were not impacting market participants. Such analyses can help determine if further regulatory intervention in markets is warranted and can inform the nature of any intervention.

Market Research Uses

Greater access and more comprehensive data on the securities lending market will improve the quality and expand the scope of research by both academics and regulators, which will better inform the regulators. In particular, improving the information available for their policy decisions will promote fair, orderly, and efficient markets and the protection of investors. For example, the data could facilitate research on the effectiveness of regulations such as Regulation SHO or FINRA Rules 4320 and 4330. Additionally, research conducted by academic researchers and market participants could also improve the value of public comment letters on Commission and FINRA proposals, which will also better inform policy decisions.

3. Direct Compliance Costs

Final Rule 10c–1a will require various entities to enter into contracts and develop recording and reporting systems to comply with the final rule. This section provides estimates of those costs. We note that the Commission has provided certain estimates for purposes of compliance with the Paperwork Reduction Act of 1995 (“PRA”), as further discussed below, in Part X. Those estimates, while useful to understanding the collection of information burden associated with the final rules, do not purport to reflect the full economic costs associated with making the required disclosures.

Table 2 shows that the Commission believes that final Rule 10c–1a will impose a one-time cost of $3.88 million and ongoing expenses of $2.76 million on FINRA, the only RNSA. An RNSA will incur these costs to develop systems to take and disseminate data required by the final rule. These include larger costs associated with creating and maintaining the infrastructure to enable providing covered persons and reporting agents to provide an RNSA with Rule 10c–1a information and entering into written agreements with providing covered persons and reporting agents, as well as smaller costs associated with providing such information to the public. These costs will be somewhat mitigated by the fact that final Rule 10c–1a requires an RNSA to disseminate the data “as soon as practicable,” which gives an RNSA some flexibility to build systems that mitigate costs.

Table 2 also shows that covered persons and reporting agents will, in aggregate, incur roughly $523 million in initial costs and $233 million annually in ongoing costs to comply with the final rule. These costs come from costs to develop and maintain systems and from costs to enter into agreements. Tables 3 and 4 break these costs down by those incurred by reporting agents, and by covered persons based on the decision by covered persons to self-report or use a reporting agent.

Table 3 shows that covered persons and reporting agents would incur an aggregate of roughly $517 million in initial costs and $233 million annually in ongoing costs to develop and maintain systems for reporting securities lending information. These include larger costs associated with developing and reconfiguring their current systems to capture the required data elements, as well as smaller costs associated with implementing changes and monitoring systems, most of which would be incurred by covered persons who provide the Rule 10c–1a information to an RNSA instead of relying on a reporting agent to do so. It is possible that some providing covered persons could privately contract with third party vendors to assist them with their reporting obligations. The potential for third party vendors, such as existing securities lending data vendors, to leverage their existing experience with collecting and disseminating securities lending data as well as economics of scale could result in lower compliance costs. If this is the case, the compliance costs for providing covered persons could be lower than what is reflected in Table 3.

Table 4 shows that covered persons and reporting agents will incur an aggregate of $5.25 million in initial costs and $0 annually in ongoing costs to enter into agreements for reporting securities lending information. These include costs associated with drafting, negotiating, and executing agreements with counterparties, most of which will be incurred by covered persons that directly employ a reporting agent. There will not be ongoing costs because, once an agreement is signed, there will be no need to modify the written agreement or take additional action after it is executed.

One commenter expressed concern some of the compliance costs incurred by lending agents could be passed along to lenders and beneficial owners, including funds and their shareholders. The Commission acknowledges that it is possible that some compliance costs may be passed along to beneficial owners, for instance, in the form of higher fees charged for lending services or lower pro-rated lending revenue. Compliance costs may also be passed along by broker-dealers to end borrowers in the form of higher borrowing costs. Ultimately, the extent to which either beneficial owners or end borrowers may bear some of the costs of complying with final Rule 10c–1a will depend on the extent to which covered persons, including lending agents and broker-dealers, pass on their compliance costs to their customers and/or split these costs across different types of customers. However, this effect is expected to be mitigated by the fact that the projected compliance costs of the final rule as a fraction of an estimate of the total lending fee revenue generated is likely small enough to be within the bounds of a positive profit margin.

In addition to the above enumerated costs, the estimated 361 reporting entities ( i.e., providing covered persons and reporting agents), as well as subscribers to Rule 10c–1a data, may be required to pay fees to an RNSA. The fees that these entities may be required to pay will depend on a number of factors, including the number of reporting entities and subscribers, along with how an RNSA will choose to split any such fees between these different types of entities.

Some commenters expressed concern that an RNSA would have monopoly pricing power over the resulting data. However, any fee that an RNSA would charge must be consistent with the Exchange Act and rules thereunder, and are subject to Commission review, notice and public comment. Consequently, the Commission believes that an RNSA is unlikely to exert monopoly pricing in terms of fees for the data, and that their fees will be reasonably related to costs. As shown in Table 2, the Commission expects an RNSA to incur ongoing costs of $2.76 million per year. If the 361 providing covered persons and reporting agents were the only entities to subscribe to the data, dividing the cost incurred by an RNSA by the 361 entities results in an annual fee per reporting entity of approximately $7,600, or approximately $633 per month. This cost would decrease to the extent that entities other than covered persons choose to subscribe to Rule 10c–1a data, which would allow an RNSA to spread its costs across more entities. At the same time, this estimate represents a lower bound on the estimated fees levied by an RNSA as an RNSA likely will need to recoup some of the initial fixed costs associated with administering the data.

One commenter stated that, by considering proposed Rule 10c–1 “in isolation” from other recent Commission rules, “the Commission is not providing a comprehensive picture of the compliance and other direct costs,” and that “these costs will aggregate and will burden market participants with higher costs when considered jointly.” But, consistent with its long-standing practice, the Commission's economic analysis in each adopting release considers the incremental benefits and costs for the specific rule—that is, the benefits and costs stemming from that rule compared to the baseline. In doing so, the Commission acknowledges that in some cases resource limitations can lead to higher compliance costs when the compliance period of the rule being considered overlaps with the compliance period of other rules. In determining compliance periods, the Commission considers the benefits of the rules as well as the costs of delayed compliance periods and potential overlapping compliance periods.

We considered here whether recently adopted rules identified by commenters that affect market participants subject to the final rule have overlapping implementation timeframes with the final rule. We found, however, that the compliance dates for these rules do not significantly overlap with the expected industry compliance dates of final Rule 10c–1a, and therefore we do not expect significant effects on compliance costs arising from overlapping compliance periods. We acknowledge that to the extent such overlap occurs, there could be costs, but we do not expect these to be significant costs.

The Commission does not believe that loan participants will incur costs associated with obtaining and renewing LEIs, because only those loan participants that have non-lapsed LEIs will be required to report them. One commenter stated that borrowers should be required to obtain an LEI if they do not already have one, because the lack of an LEI would make it more challenging to identify entities. We agree that wider LEI adoption and reporting would likely lead to more consistent and precise identification of legal entities, which could enhance analysis of the reported information. We also recognize that any mandatory LEI reporting requirement for LEI-eligible loan participants would also generate additional costs for loan participants that do not currently have them, although such costs would likely be modest.

4. Other Costs

Increase in Borrowing Costs

The Commission acknowledged in the Proposing Release that increased compliance costs could lead to some consolidation in the securities lending market and impose some costs on market participants, including potentially higher prices associated with reduced competitive pressures. However, the Commission believes that this effect by itself is unlikely to lead to higher overall borrowing costs due to the fact that the value of securities available to be loaned generally far exceeds the total value on loan. For general collateral securities, which make up the majority of loans, a modest decline in shares available to lend would not eliminate the slack in the market and thus would not likely increase fees. For securities “on special” with less available supply relative to demand, it is possible that a decline in shares available to lend could result in higher borrowing costs for some securities. However, the Commission believes that, to the extent that it occurs, these fee increases would likely be small and have minimal downstream economic effects. This minimal effect is expected for two reasons. First, the projected compliance cost of the final rule is a relatively small fraction of the likely total lending fee revenue generated, which makes it likely that most broker-dealers and lending programs would continue to earn a positive profit margin after accounting for these costs. Second, academic research suggests that very small increases in lending fees are unlikely to result in significant downstream economic effects. As a result, the Commission expects that any increase in borrowing costs that may result from consolidation among broker-dealers or lending programs would be minimal, and therefore likely offset by a simultaneous decrease in borrowing costs due to improved transparency and efficiency in the securities lending market which, as described above, are expected to concentrate among stocks that are on special.

Lower Revenues From Securities Lending

Commenters expressed concern that, to the extent that lenders are not able to pass on compliance costs to borrowers through increased fees, the rule could lead to decreased profit from securities lending for some market participants. However, the Commission estimates that, as the projected compliance cost of the final rule is a relatively small fraction of the likely total lending fee revenue generated, most market participants will continue to be able to earn a positive profit margin from securities lending.

In addition, as acknowledged in the Proposing Release, a reduction in information asymmetry may result in reduced revenue for some broker-dealers and lending programs in at least two ways. First, a reduction in securities lending revenues could occur because end borrowers and beneficial owners will have more information about the state of the lending market. As a result, broker-dealers and lending programs who consistently underperform, the market may lose customers to better-performing broker-dealers and lending programs or begin offering better terms to their customers. Both possibilities represent a reduction in revenue for some broker-dealers and lending programs. It is possible that some broker-dealers and lending programs may choose to exit some or all of the market for lending services as a result of this loss of revenue. The loss of revenue will in part be a transfer to the end borrowers and beneficial owners that switch to better-performing lending programs and better-performing broker-dealers, as well as the better-performing lending programs and broker-dealers that attract additional customers. The Commission is not able to quantify the potential loss in revenue that could occur for underperforming broker-dealers and lending programs because the Commission is not able to quantify the extent to which these broker-dealers and lending programs' current revenues are driven by asymmetric information.

Second, lending programs may also experience reduced profitability through the lower rates offered by broker-dealers to their customers. If a given lending program has become skilled in cultivating relationships with broker-dealers currently willing to pay higher fees, then the increased competition that broker-dealers face as a result of the rule may lead to lower overall fees being charged for security loans—lowering the total lending revenue produced by securities lending. While the Commission expects that beneficial owners will benefit from the final Rule as a result of the reduction in information asymmetry, in some cases lower lending revenues and potentially increased costs may reduce the revenue earned by some beneficial owners. As acknowledged in the Proposing Release, this would represent a partial transfer from beneficial owners to the end borrowers who may receive better terms on average as a result of decreased information asymmetries. Ultimately, the Commission is not able to quantify the net impact of the rule on beneficial owners' lending revenues, as the Commission is not able to quantify, among other items, the extent to which end borrowers may receive better terms on average as a result of decreased information asymmetries.

Lower Revenues for Securities Lending Data Vendors

Commenters expressed concern that the increase in securities lending information provided by the rule will also result in costs in the form of lost revenue for commercial securities lending data vendors, which could lead some data vendors to pare back their product offerings. The Commission acknowledges that this may occur for a number of reasons. First, commercial data vendors may pare back their offerings if demand for their products decreases because their customers switch to using Rule 10c–1a data. Second, some market participants who currently contribute data under a give-to-get model may find it less worthwhile to do so when the rule provides an alternative source of data. This effect would diminish the quality of the give-to-get data currently provided by commercial data vendors and may lead to fewer market participants being willing to purchase the data. Both of these effects could result in lower revenues. To the extent that data vendors do pare back their product offerings and/or if the quality of their data products are diminished, then this could reduce transparency in the securities lending market in cases where these datasets contain information that is not collected and disseminated by the final rule, such as utilization rates and shares available to lend.

The Commission believes that a potential mitigating factor that may reduce or even offset the severity of this loss in revenue will be that commercial data vendors may offset some of the impact of lowered demand for their data by enhancing their related data analytics businesses using Rule 10c–1a data. Furthermore, to the extent that customers value the availability of information about securities available to lend and utilization rates, the fact that commercial data vendors will continue to have superior access to this information will likely mitigate or even offset a decrease in demand, as investors would continue to purchase commercial datasets in order to gain access to this information. The Commission is unable to quantify the potential impact of the final rule on securities lending data vendors' revenues because this would require knowledge of the baseline level of such revenues, which the Commission does not have access to and about which commenters did not provide information.

Miscellaneous Costs

Some commenters expressed concern that the rule requires the reporting of sensitive information and thus could present a risk to individuals in the case of a data breach. The Commission recognizes that the final rule collects sensitive information and that the costs of a data breach could can be substantial. These costs include, but are not limited to, the following: (1) trading losses that could occur due to the revelation of private trading strategies or economic positions which may enable identifying and trading opportunistically around such strategies, such as facilitating a short squeeze; (2) business disruptions that could occur if the data breach results in temporary system down time; (3) data breach response costs as market participants must devote resources to determining how to respond to the data breach; and (4) reputational harm to an RNSA. While the potential costs of a breach, to the extent that one occurs could be severe, RNSAs, as well as ATSs and SROs, are currently subject to existing regulations that aim to improve the resiliency and oversight of securities market technology infrastructure, such as Regulation Systems Compliance and Integrity (“Regulation SCI”). Adherence to regulations that seek to ensure the resiliency and integrity of technology systems can reduce the probability of a data breach and mitigate the costs associated with a breach, should it occur.

Some commenters expressed concern that the rule could inhibit innovation in securities lending, such as the potential adoption of blockchain technology, by diverting resources away from innovation and towards compliance. While such diversion is possible, we expect that market participants will continue to pursue innovation that creates a competitive advantage or for which there is a market demand. One commenter suggested that innovations such as blockchain technology could obviate the need for transaction reporting systems. The Commission acknowledges this possibility; however, such innovations would need to overcome the market failures described earlier in this analysis.

One commenter expressed concern that, “with lenders dependent upon their agents for reporting to the RNSA, the rule could make it more difficult for lenders to switch agents.” To the extent that there would be any bespoke reporting services offered by lending agents, it is possible that the final rule could increase a lender's dependence on their lending agent, and thus marginally increase their costs of switching agents. However, the Commission expects this increased cost, to the extent it occurs, to be minimal. The Commission expects that many covered persons, including lending agents, will rely on reporting agents to provide the Rule 10c–1a information directly to an RNSA. This would generally result in an unbundling of lending and reporting services, such that an increased dependence on a lending agent for their reporting services would be unlikely. To the extent that a lender requests or requires bespoke reporting services from their lending agent, that lender could examine such a service among the package of lending services—including reporting services—that a prospective agent offers, and decide accordingly while taking into account the potentially higher costs of switching to a new lending agent.

To the extent that there are entities that would like to serve as reporting agents but are not currently registered as broker-dealers, the cost of registering as a broker-dealer may be another cost associated with the final rule. However, the final rule does not require any entity to serve as a reporting agent, and as such this cost would only accrue to those entities that make the business decision to become a reporting agent. It is likely that an entity that is not currently registered as a broker or dealer will only make the business decision to become a reporting agent if they believe that the cost of registering as a broker-dealer is more than offset by the revenue that they will earn as a reporting agent.

5. Reduced Benefits From Alternative Arrangements

Economically Similar Arrangements

The Commission recognizes a risk that the comprehensiveness of Rule 10c–1a data, and hence the benefits that accrue due to the comprehensive nature of the data, could be diminished to the extent that market participants choose to use arrangements that are economically similar to securities lending agreements. For example, market participants may substitute repo for securities lending agreements. This substitution may occur because a fixed term cash collateralized securities loan is economically very similar to a repo. While the Commission is unaware of short sales of equities currently being facilitated by repo contracts, the Commission understands that in fixed-income it is fairly common for entities wishing to short sell a bond to facilitate that transaction with a repo instead of a securities loan.

The Commission believes that this risk varies across asset classes. In equities, the Commission believes that the current risk of such migration may be minimal because of the lack of a well-developed repo market for equities. However, this risk may increase if the market for equity repos becomes more developed in the future. Among fixed-income securities the risk is substantially greater due to a well-developed repo market for fixed-income securities and the established practice of using both securities loans and repo transactions to facilitate short sales of fixed-income securities. In all asset classes, if the final Rule 10c–1a leads to improvements in the functioning of the securities lending market, then the risk of migration may diminish to the extent that improved efficiency in the securities lending market diminishes the incentive to transfer activity to potentially less developed repo markets.

Should this substitution affect a significant volume of securities lending, certain benefits and costs discussed above may decline. The less comprehensive data could reduce the extent to which the Rule reduces any biases or gaps in the data. For instance, market participants who use the data to price securities loans will have a less accurate and potentially biased view of the market, which will limit the benefits from increased transparency. Additionally, regulators using the data to determine lending market conditions at the time of, for example, a Reg SHO violation will be using less precise data—limiting the benefits of Reg SHO enforcement. On the other hand, such substitution could reduce compliance costs for some. Obviously, those substituting into repo will incur lower compliance costs from the final rule, including one-time implementation costs if they replace all securities lending with repo. Further, a significant substitution will reduce the ongoing costs of an RNSA because an RNSA will not have to collect and process as many transaction reports.

Cross-Border Application

One commenter expressed concern that “an overly broad or ambiguous reach of Proposed Rule 10c–1 could reduce liquidity and place U.S. intermediaries . . . at a competitive disadvantage,” because “market participants—particularly those outside of the United States—might determine to limit or cease trading or interacting with U.S. intermediaries (or leave the U.S. markets all together) in order to prevent public disclosure of their transactions.” The Commission acknowledges some market participants could be incentivized to restructure their activities to avoid reporting. Specifically, in an effort to avoid having their lending activities fall within the reach of section 10(c), participants could relocate overseas any of their current U.S.-based conduct that constitutes effecting, accepting, or facilitating a lending transaction. However, the Commission believes that this risk is low. As pointed out by the same commenter, there are disincentives for market participants to move lending activity outside of the U.S. market, including “the depth and liquidity of the U.S. market,” as well as few incentives for “U.S. asset managers to prefer non-U.S. lenders that are themselves subject to similar reporting requirements in their home jurisdictions.”

C. Impact on Efficiency, Competition, and Capital Formation

1. Efficiency

In the securities lending market, the availability of Rule 10c–1a data for market participants will lead to more efficient prices for securities loans. The reduction in asymmetric information in the market for lending programs and broker-dealers may also make those markets more efficient. Additionally, the Commission believes that final Rule 10c–1a may have secondary effects that will increase price efficiency in the underlying securities markets and option markets. Also, the increased ease with which banks and other financial institutions will be able to manage collateral and balance sheets as a result of the final rule could lead to increased efficiency in their functioning and in those markets in which they play a role.

2. Competition

The Commission believes that the net impact of final Rule 10c–1a on competition is difficult to predict, in that some aspects will likely increase competition and some aspects will likely reduce competition. Competition will likely be impacted in the markets for broker-dealer services, lending programs, and securities lending data vendors.

The Commission believes that one effect from the final rule's reduction of information asymmetry between end borrowers and broker-dealers, and between beneficial owners and lending programs, will be to increase competition between broker-dealers and between lending programs. A reduction in information asymmetry will permit better monitoring of the performance of these entities by their respective customers, and will likely force these entities to do more to match the performance of their competitors, to the extent that they do not already do so. One commenter stated that the effect of the final rule on competition between broker-dealers may be attenuated by the presence of high switching costs. The Commission does not believe that switching costs for most market participants are high enough such that this is a likely outcome.

Furthermore, the increased ability for broker-dealers to monitor conditions in the lending market may encourage new broker-dealers to enter the market, further increasing competition for broker-dealer services. This same argument may be true for platforms that engage in securities lending. Improved data may allow for better evaluation of the performance of such platforms and may also lower barriers to entry for new platforms, thus enhancing competition among securities lending platforms.

Commenters expressed concern that competition in the securities lending market could decrease if lenders are driven out of the market because of compliance costs. The Commission does not believe that compliance costs are such that this would be a likely scenario because the projected compliance costs associated with final Rule 10c–1a are likely to be a relatively small fraction of total lending fee revenues. At the same time, the reduction in asymmetric information in the securities lending market that will result from the final rule may diminish broker-dealer and lending program profits to the extent that it reduces their current information advantage over their customers. To this end, some broker-dealers and lending programs whose profitability primarily depends on economic inefficiencies associated with asymmetric information may exit the market for facilitating securities loans.

The Commission also recognizes that, given the significant fixed costs of implementing the systems required by the final rule for lending programs to report to an RNSA, some smaller lending programs and broker-dealers may consolidate or exit the lending market. The Commission believes that a mitigating factor leading to less consolidation is that the current relationship and network structure of lending programs and broker dealers already favors larger lending programs and broker-dealers who have the resources to maintain relationships with more and larger securities lending counterparties. Consequently, the Commission believes that the market for lending programs and broker-dealer security borrowing services is already likely dominated by larger lending programs and broker-dealers that the Commission does not believe will cease operating as a result of these fixed costs. The Commission recognizes that smaller lending agents may have unique business models that are not currently offered by competitors, but the Commission believes a competitor could create similar business models if demand were adequate.

The Commission believes that the dissemination of Rule 10c–1a data collected under the final rule will change the competitive landscape for securities lending data analytics services by increasing opportunities for enhancing products and services that depend on securities lending data and lowering barriers to entry concerning who can provide those services. Increased competition in this space will likely lead to more options for consumers of analytics services, lower prices, and improved analytics services. The new information available through an RNSA as a result of the final rule will produce an alternative to existing data vendors' commercial securities lending data products. Existing data vendors may retain a competitive advantage, however, as they will likely continue to have access to information on utilization rates and shares available for lending, which are not provided by the final rule.

Some commenters expressed concern that the rule would cause lending activity to migrate overseas to avoid the reporting obligation under final Rule 10c–1a, which would make the U.S. securities lending market less competitive. While it is possible that some market participants could move their activity to other jurisdictions in attempt to circumvent the reporting obligation of final Rule 10c–1a, the Commission believes that it is unlikely that this would occur on a large scale. This is because the final Rule imposes a reporting obligation on any qualifying loan regardless of whether the person is located abroad or is a non-U.S. person when the transaction occurs.

In addition, as stated above, some commenters requested the Commission consider interactions between the economic effects of the proposed rule and other recent Commission rules, as well as practical realities such as implementation timelines. One commenter stated that because “[t]he resources of market participants are not infinite,” that commenter “believe[s that] unnecessary and aggressively timed regulatory changes will impact costs, complexity, competition and the ability of smaller market participants to enter into or remain in the markets.” As discussed above, the Commission acknowledges that simultaneous compliance periods may in some cases increase costs. This may be particularly true for smaller entities with more limited compliance resources. This effect can negatively impact competition because these entities may be less able to absorb or pass on these additional costs, making it more difficult for them to remain in business or compete. However, we determined that the rules highlighted by commenters either did not impose implementation costs on the same entities as this final rule and/or had compliance dates that did not significantly overlap with the expected industry compliance dates of this final rule, and therefore we do not expect these effects on competition to be significant. We acknowledge that to the extent such overlap (in scope or timing) occurs, there could be costs which could affect competition, but we do not expect these costs to be significant.

3. Capital Formation

The Commission believes that the impact of final Rule 10c–1a will have mostly positive effects on capital formation. In particular, improved price discovery in securities markets and improved balance sheet management by financial institutions could facilitate improvements in the provision of capital. In addition, to the extent that the final rule reduces the costs of short selling it may facilitate more effective discovery of negative information that in turn could lead to more efficient allocation of capital. However, to the extent that the Rule increases the cost of short selling for some stocks in some situations, it could also lead to less efficient allocation of capital in these instances.

D. Alternatives

1. Report Loan Sizes Without a Delay

The Commission considered an alternative requiring that an RNSA disseminate loan sizes at the same time as it disseminates all other loan information ( i.e., without a 20-business-day delay). This alternative would increase the timeliness of securities lending information available to the public, which could improve market quality by reducing information asymmetries as well as providing additional transparency to the securities lending market. However, it could also harm market quality by providing novel information about short positions with a greater timeliness than the short selling information that is currently available.

To the extent that the aggregate cost-to-borrow statistics that an RNSA publicly disseminates are volume-weighted, these statistics will provide end borrowers with access to information about loan prices that incorporates information about loan sizes even prior to the public dissemination of loan sizes. However, disseminating securities loan sizes as well as modifications of securities loan sizes on the next business day after a loan is effected or modified would allow market participants to produce their own custom cost-to-borrow metrics that are most tailored to their business needs. This would allow market participants to focus on the aspects of the market that are most relevant to their business needs. This effect would increase the benefits of final Rule 10c–1a from increased transparency in the securities lending market.

However, the information could also be used to estimate short selling positions with more accuracy than is possible under the final rule. A market participant wishing to gain insight into short selling sentiment could aggregate the sizes of all open loans that are associated with categories of borrowers that are likely to be short sellers. If information about these loans is available the day after the loans are effected then, given the settlement cycles for equity short sales and equity loans, this would be equivalent to three business days after an equity short sale took place. These data would be considerably timelier than existing short interest data such as the FINRA data. Additionally, market participants could follow individual short positions with a three-business-day lag for equity short positions. This is a capacity that is not possible using existing data. This could be accomplished by tracking individual loans that are marked as being to a customer if the lender is a broker dealer. While the identity of the borrower would not be available to the public, market participants could discern whether changes in short interest were due to changes across many positions or few.

This additional information that market participants could extract about short selling could enable copycat investing strategies that seek to mimic what the short sellers do. This could increase the costs of establishing a short position, thus making it costlier for short sellers to profit from their research. If the profitability of research diminishes, there is the risk that less market research could be performed and, if this occurs, there could be negative impacts on market quality and corporate monitoring.

2. Alternative Timeframes for Reporting or Dissemination

The Commission also considered an alternative requiring different timeframes for reporting or disseminating the securities lending transaction information.

First, the Commission considered requiring covered persons to provide an RNSA with Rule 10c–1a information earlier than the end of each business day on which a covered securities loan is affected. For example, the Commission could require covered persons to provide an RNSA with information about a loan within a fixed period after the loan has been effected, such as 15 minutes. This alternative would increase the timeliness of securities lending information, potentially increasing some of the benefits of final Rule 10c–1a, such as regulators' access to this information for regulatory and surveillance use. Furthermore, if an RNSA were required to make the information publicly available as soon as practicable after an, e.g., 15-minute reporting timeframe, this could increase the benefits of the final rule for transparency in the securities lending market by making the data available up to one business day sooner than under the final rule. It would, however, likely increase compliance costs to an RNSA as they would be required to build out systems capable of intraday dissemination.

At the same time, several commenters pointed out numerous costs associated with more timely and frequent reporting. For example, some commenters suggested many loan terms are not finalized until the end of the day, and so a 15 minute reporting requirement increases the risk that the data provided by the final rule could be less accurate and more prone to errors that would later have to be corrected, which would mitigate the usefulness of the final rule. This alternative would also require greater compliance costs because, as commenters also pointed out, securities lending systems would need greater adaptations to comply with a more stringent reporting time horizon.

The Commission could also require covered persons to provide an RNSA with Rule 10c–1a information later than the end of each business day on which a covered securities loan is effected, for example, by the end of the next business day after a covered securities loan is effected. This alternative may lead to lower compliance costs, as reporting entities would have more time to prepare the Rule 10c–1a information for reporting; however, the Commission believes that longer reporting horizons would likely not decrease the cost substantially due to the automated nature of the securities lending transactions and the need to build out systems regardless. Furthermore, a longer reporting horizon would delay the dissemination and availability of securities loan information, potentially reducing some of the benefits of the final rule.

In general, because securities loan terms cannot be disseminated until after they are reported, alternative reporting timeframes reflect different tradeoffs between the value of disseminating security loan terms close to the time that the loan is effected, and the compliance costs associated with reporting loans at shorter time horizons.

The Commission also considered alternatives requiring an RNSA to distribute the collected data required in paragraph (c) at different horizons, such as by the end of the following day. This alternative would allow an RNSA to process the data during regular business hours, potentially limiting the amount of data validation an RNSA could perform prior to distributing the data. However, this would delay market participants' abilities to benefit from Rule 10c–1a data by at least a business day. Furthermore, given the automated nature of the data, the compliance costs associated with this alternative are not likely to be significantly lower than those associated with the final rule. Alternative dissemination timeframes reflect different tradeoffs between price discovery and price efficiency benefits on one hand and harmful information leakage on the other, as well as the cost of reporting at a faster or slower horizon.

An alternative dissemination timeline could require delayed dissemination of size information about large loans ( i.e., loans larger than a certain size threshold) while disseminating size information about smaller loans the next business day. This alternative would provide more information to the market than the final rule because it would not mask the size of smaller loans, and thus could enhance the final rule's benefits due to increased transparency. However, this alternative would increase the amount of novel information about short positions that the market could learn from Rule 10c–1a data, which would increase the risk of negative effects due to over-exposing short selling information. This alternative would also increase the complexity of the Rule as determinations would need to be made regarding the optimal threshold size level, as well as how to handle loan modifications that could potentially move a given loan size above or below the given threshold.

3. Only Require Dissemination of Aggregate or Wholesale Statistics

The Commission also considered an alternative whereby an RNSA would produce only aggregate statistics and no transaction-by-transaction reports. Specifically, an RNSA could aggregate transaction data elements and produce various aggregated statistics at the end of the business day following the date that the loans were effectuated, without disseminating the raw transaction-by-transaction data.

This alternative would not change the reporting requirements for covered persons relative to the final rule and thus would have the same initial and ongoing compliance costs as the final rule for covered persons. This alternative would be somewhat less costly to implement for an RNSA than the final rule, which requires an RNSA to produce daily aggregate information, similar to this alternative, but also requires dissemination of loan-by-loan information, which this alternative would not do.

This alternative would increase the amount of information available to the public relative to the baseline and thus would mitigate information asymmetries relative to the baseline. Aggregate statistics would also make it more difficult for market participants to use the data to discern information about short positions, which would reduce the risk of exposing short selling strategies relative to the adopted Rule.

However, only having daily aggregate data would not provide the same transparency benefits as final Rule 10c–1a. Without comprehensive transaction data, it would be more difficult for market participants to study and understand pricing dynamics in the securities lending market. The alternative would also make it more difficult for end investors to determine if the terms that their broker-dealer offers are consistent with current market terms for similar loans, rendering it more difficult for investors to evaluate the performance of their broker-dealer. Similarly, without transaction data, beneficial owners would be hampered in their ability to determine whether the terms for loans secured by their lending agents were consistent with market conditions for loans with similar characteristics, rendering it more difficult for beneficial owners to evaluate the performance of their lending agents. A lack of publicly disseminated transaction data may also hinder broker-dealers from determining if the terms being offered by a lending agent for a loan are consistent with market conditions for similar loans. These effects would reduce the benefits of final rule for improved competition.

The diminished transparency of this alternative relative to the final rule may also lead to less improvement in the efficiency of the securities lending market, leading to fewer benefits for traders in the form of increased trading profits. This alternative would also hamper research into the securities lending market by some market participants and academics, relative to the final rule, by preventing researchers from performing analysis using individual transactions.

The Commission also considered several alternatives related to the applicability of the rule to market for Customer loans, including an alternative in which Customers loan are excluded from reporting requirements, and an alternative in which Customer loans are reported, but not disseminated to the public. One commenter stated that “the Commission should limit the Proposal to the `wholesale' market,” because including Customer market loans in a securities lending reporting regime would be akin to a “transaction-by-transaction short sale public reporting regime.”

The Commission believes that limiting the applicability of the rule for Customer loans would significantly limit the benefits of the rule for increasing transparency in the securities lending market. In particular, end borrowers would not benefit from a reduction in their information disadvantage vis-à-vis their broker-dealers in the securities lending market. As a result, the effects of the rule on competition between broker-dealers, which would be expected to lower borrowing costs for hard-to-borrow stocks and thus reduce costs for short sellers, would not materialize. Market participants would also not be able to use Rule 10c–1a data to compare trends in both the Wholesale and Customer markets simultaneously. Furthermore, while the Commission acknowledges that information about Customers loans can correspond closely to information about short positions, the Commission believes that concerns about whether the dissemination of Customer loan information will harm short sellers are largely addressed by requiring that an RNSA delay public disclosure of the loan amount by 20 business days. This modification should significantly reduce the novelty of the information disseminated under the final rule regarding short sellers' positions, such that it is less timely than pre-existing sources of short selling transparency, such as FINRA's bimonthly short interest data.

Second, excluding Customer loans from reporting requirements altogether would not only reduce the benefits of the rule from increased transparency as described above, but would also reduce the regulatory benefits of the rule, as the Commission and FINRA would no longer benefit from an improved ability to use Rule 10c–1a data for surveillance and enforcement, market reconstruction, and market research in Customer segment of the lending market. Furthermore, it is unclear whether the costs of the rule for short sellers would be significantly reduced by the exclusion of Customer loans from reporting requirements compared to excluding them from dissemination, as the commenter's concerns related to the inclusion of Customer loans stem primarily from the public dissemination of this information.

4. Require Reporting of Shares Available To Lend and Shares on Loan

The Commission also considered an alternative requiring market participants to report shares available to lend and shares on loan to an RNSA. An RNSA would then produce aggregate statistics by security regarding shares available to loan and shares on loan. This alternative would increase the amount of information available to market participants regarding conditions in the securities lending market, which would increase the benefits of the rule from increased transparency. It could also benefit consumers of Rule 10c–1a data by lowering barriers to entry for firms to start providing securities lending analytics services, as they could incorporate shares available into their analysis without purchasing data from the current commercial data vendors. This could lead to benefits in the form of increased competition for securities lending analytics.

However, commenters opposed the inclusion of shares available on the grounds that any measure of shares available would be inherently noisy. The Commission does not believe that requiring reporting entities to report shares available would result in a measure that is inherently noisier than the current estimates provided by commercial data vendors, which research shows can be informative. As such a measure would be more comprehensive than existing estimates, it would likely be more informative than existing data estimates. However, the Commission acknowledges that any estimate of shares available to lend would be noisy for numerous reasons, including the reasons mentioned by commenters, and that this noise limits the benefits of requiring the reporting of information about shares available to lend.

This alternative would also increase the costs to the commercial data vendors that currently offer utilization rate and shares available information because it would provide market participants with alternate access to such information. This effect could lower demand for the data currently provided by commercial data vendors, leading to a loss of business for these data vendors.

Regarding shares on loan information, the Commission believes that the benefit of requiring covered person to report aggregate information about shares on loan would also be limited, as this information is rendered somewhat redundant by the requirement that an RNSA disseminate aggregated transaction activity information for each security. At the same time, requiring covered persons to report aggregate information about both shares on loan and shares available would increase the compliance costs relative to the final rule by requiring an additional report ( i.e., a report on aggregated securities lending activities) to be reported to an RNSA in addition to the transaction-by-transaction securities lending data.

5. Restrict Covered Persons to Broker-Dealers

The Commission also considered an alternative requiring only broker-dealers, rather than any person that qualifies as a “covered person” under the definition in the rule, to report securities lending transactions to an RNSA. The Commission believes that this alternative would be less costly overall than final Rule 10c–1a. Specifically, covered persons that are not broker-dealers would not incur any of the costs of reporting. As a result, fewer entities would incur costs. Further, most broker-dealers already have connections to FINRA, the only RNSA, so the overall implementation costs associated with connecting to an RNSA for the purposes of reporting would be lower.

In addition, because most broker-dealers currently have relationships with FINRA, the Commission preliminarily believes that this alternative could be implemented sooner, allowing the market and market participants to benefit from an increase securities lending transparency sooner.

However, under this alternative, Rule 10c–1a data would not provide a comprehensive view into the securities lending market. Even though broker-dealer activity makes up a significant percentage of securities lending transactions in the Customer market, the alternative would exclude market participants such as lending programs that are significant players in the securities lending market, particularly in the Wholesale market. Thus, the alternative would obscure a large swath of the Wholesale market, making it more difficult for lending institutions, for example, to benefit from securities lending transparency because the included data would provide a less relevant benchmark.

Requiring only broker-dealers to report data could also create a competitive advantage for non-broker-dealer entities that engage in securities lending. Such entities would not be required to report their transactions and thus would have lower costs. They would also be in a position to attract business from borrowers or beneficial owners seeking to keep their transactions out of the public view, further tilting the competitive landscape in their favor. This could create an uneven playing field for entities engaged in the securities lending market, diminishing the benefits of the rule for competition and transparency.

6. Expand Reporting Agents Beyond Broker-Dealers and Clearing Agencies

The Commission also considered an alternative expanding who can act as a reporting agent to entities other than broker-dealers and clearing agencies. Commenters expressed concern that restricting reporting agents to only broker-dealers would result in conflicts of interest between broker-dealers acting as reporting agents on the one hand, and beneficial owners and lenders who transact with these broker-dealers in the primary market on the other. Since information about an entities' lending activities is likely to correspond closely to that entities' inventory levels, it is possible that such a risk of information leakage may be a significant cost for some lenders and lending agents who are “highly concerned about visibility into inventory levels.”

Furthermore, expanding reporting agents to entities other than broker-dealers and clearing agencies would serve to increase the number of entities that would compete to provide lenders and lending agents with reporting services. Thus, expanding the eligibility of reporting agents would serve to promote more competition in this market, potentially leading to lower fees for lenders and lending agents that would rely on these reporting agents for these services.

However, the Commissions believes that expanding reporting agent eligibility beyond broker-dealers and clearing agencies would increase the costs associated with the rule. The Commission's oversight over reporting agents that are neither broker-dealers nor clearing agencies could be limited, which could shift the costs of monitoring reporting agents for accuracy and compliance onto the lenders and lending agents that would hire them. This would serve to increase covered persons' ongoing costs.

In addition, to the extent that covered persons are concerned about contracting with broker-dealers as reporting agents because of potential conflicts of interest, to the extent that there are clearing agencies that would act as reporting agents, these covered persons could contract with these entities. Furthermore, covered persons are not restricted from contracting privately to use third-party vendors to assist in reporting. Therefore, those market participants that are concerned about revealing sensitive inventory information to broker-dealers but would still benefit from reporting assistance could privately contract with third-party vendors. Since covered persons could not rely on third-party vendors to fulfill their regulatory obligations, contracting with a third-party vendor for assistance in reporting may require covered persons to incur some amount of monitoring costs; however, this option should provide covered persons with additional flexibility and a way to potentially mitigate reporting costs.

7. Publicly Releasing the Information in 10c–1a(e)

The Commission also considered an alternative requiring public disclosure of the information in Rule 10c–1a(e), namely available identifiers for each party to the transaction, whether the security is loaned from a broker's or dealer's securities inventory to a customer of such broker or dealer, and if known whether the loan is being used to close out a fail to deliver.

Information on who the parties to the transaction are and whether a broker or dealer is lending to its own customer could refine the context around the other data elements in Rule 10c–1a that are public. Such refinement would be likely to alter trading strategies, which could have both positive and negative effects on market quality. For example, this information could allow the market to identify the positions of large short sellers. Empirical studies support the idea that short sellers are informed, suggesting that additional information about short selling could help investors better value securities. Professional traders might seek to profit by developing trading strategies based on signals from the identities of those borrowing securities, particularly those borrowing a high volume. In addition, the information could be used to reduce the search costs in the securities lending market.

However, the information on whether the security loan is being used to close out a fail to deliver may be of little use to anyone other than regulators. At this time, the Commission is unaware of potential non-regulatory uses of such information that would be beneficial to the market.

The alternative would result in higher costs to an RNSA, to those who access the data, and to participants in the securities lending market. An RNSA would incur higher costs to release the greater volume of data and those who access the data would incur higher costs to import and process the data. Trading strategies incorporating the identities of borrowers and lenders could negatively impact those borrowers and lenders in ways that could ultimately degrade price efficiency. In particular, identifying large short sellers could facilitate “copycat strategies” that seek to profit by copying the activity of others believed to have better information or by trading ahead of them. If it facilitates such trading strategies, releasing the identities of short sellers could act as a constraint on fundamental short selling, reducing the incentives to conduct fundamental research. Less fundamental research could potentially result in over- or underpricing, because prices would not incorporate information short sellers would have otherwise collected and traded on. Revealing the identities of participants and when they are borrowing to close failures to deliver in the securities lending market could also result in pressure on lenders to recall loans or negative campaigns against short sellers.

8. Additional Information in the Reported or Disseminated Information

The Commission also considered alternatives requiring additional data fields to be reported and requiring an RNSA to compute derived fields for public dissemination.

For example, the Commission could require entities to report in their lending transactions whether a given loan was transacted on their own behalf, or on behalf of a customer, i.e., whether the loan was transacted on a principal or agent basis. This alternative would allow FINRA and the Commission to oversee compliance with various regulations. These data could allow for the review of transactions that occur by an entity on a principal and agent basis to look for systematically different terms between the two different types of transactions by the same broker-dealer. Such differences may flag to regulators that broker-dealers are not fulfilling their obligations and may be in violation of existing rules. Requiring such data would add complexity and additional cost to the rule. However, these costs may be minimal for broker-dealers, who are FINRA members, as the Commission understands that FINRA members already collect much of this information. The Commission is unaware of any regulation or rule requiring non-FINRA members to collect this information, and this alternative may significantly increase costs for non-FINRA members who would be required to build out systems to collect and report such information.

As discussed in the Proposing Release, the Commission considered including in Rule 10c–1a(e) information on whether, if the lender is a broker or dealer, the securities are borrowed from customers who have agreed to participate in fully paid lending programs or from securities owned in its margin customers' accounts. Such information could help protect investors by improving the efficiency of surveillance of, for example, compliance with Rule 15c3–3(b)(3) related to providing the lender collateral to secure the loans of securities when broker-dealers lend shares from fully paid or excess margin securities from customers. However, the Commission has since come to understand that it may not be feasible for broker-dealers to identify which customers' shares are being loaned.

9. Only Allow an RNSA To Charge Fees to Data Reporters

The Commission also considered an alternative that would allow an RNSA to charge fees to reporters of Rule 10c–1a information, but not to subscribers to access the securities lending data. The effect on costs of this alternative would be that any fees levied by an RNSA to help pay for costs to collect and disseminate the data would shift onto covered persons providing data to an RNSA. The Commission believes that many of the market participants providing data to an RNSA under the final rule will also be consumers of the data. If all 361 entities providing data were the only entities to subscribe to the data, for these market participants it is unclear how much difference this shift in fees would make. However, to the extent that more market participants subscribe to the data, prohibiting an RNSA from this additional source of revenue would likely result in higher reporting fees than under the final rule.

This alternative would potentially increase the benefits of the final rule for increased transparency, in that providing access to Rule 10c–1a data for free may increase the number of market participants who access it. However, TRACE has been successful in mitigating inefficiencies in the corporate bond market despite the fact that it is only available for fee. Furthermore, the Commission expects that most of the entities likely in a position to affect the securities lending market or to use information from the securities lending market to affect other markets would subscribe to the data regardless of whether there was a cost to subscribing. Therefore, the Commission does not believe that disallowing an RNSA from charging fees to data subscribers would result in significant additional benefits to the securities lending market relative to the final rule.

10. Longer Holding Period Requirement

The Commission also considered an alternative requiring an RNSA to retain and make publicly available the data for a period longer than the 5 years specified ( e.g., 10 or 20 years). This alternative would ensure that the data are available to regulators and market participants at longer horizons. For instance, if regulators or market participants wanted to evaluate how the lending market reacts to different market events, such as across the business cycle, then five years of data may not be sufficient. The average business cycle is 3–5 years, and so to study the dynamics of the lending market across the business cycle would require at least 10 years, if not more, of data. Additionally, because there is likely persistence in conditions in the securities lending market, a five-year time horizon may not be sufficient for certain statistical analyses. Improved understanding of the dynamics of the securities lending market across various market conditions may benefit both regulators and investors by providing more precise information with which to make regulatory and investment decisions—enhancing many of the benefits described above. For example, longer term data may enable superior statistical analysis by market participants of the dynamics of the securities lending market in various environments, which in turn may lead to better investment decisions and thus improved market performance. Additionally, the Commission could use longer term data to provide more precise estimates of damages in, for example Reg SHO violations or violations of Exchange Act Rule 15c3–3 (Customer Protection Rule), to calculate disgorgement.

The Commission believes that the alternative would impose additional costs on an RNSA not required by the final rule in terms of storing and maintaining historical data. However, since the final rule already requires an RNSA to build systems to collect and disseminate five years of data, these costs would likely be relatively small because the Commission understands that the cost of storing data is relatively small compared to the cost of producing and maintaining the systems needed to collect, process, and disseminate the data.

While the final rule allows FINRA to destroy the data after five years, the Commission believes that it is unlikely that FINRA would do so. This is because the cost of retaining the data is likely relatively small and the data may have commercial value. For instance, an RNSA could levy additional fees for access to historical data. If this is the case, and an RNSA chooses to keep the historical data under the final rule, then the cost difference to an RNSA between the final rule and this alternative would likely be minimal given that this alternative would require an RNSA to comply with a requirement that they may already choose to do on their own.

11. Longer Implementation Period

The Commission considered alternatives that would allow for a longer implementation period. Several commenters requested implementation periods of 18 or 24 months, which may be longer than the final rule's implementation period. Some commenters requested a phased-in approach to implementation, for example, by first only requiring information to be reported for certain types of securities, or initially delaying the public dissemination of data. The Commission recognizes that a longer implementation period would give reporting entities additional time and flexibility while building out the systems necessary to facilitating the reporting required under the final rule, which may allow a more efficiency allocation and resources and thus reduce their implementation costs. Similarly, delaying the public dissemination of data would give an RNSA additional time and flexibility to build out systems related to dissemination, which may reduce their implementation costs.

However, the Commission also recognizes that any delay in the implementation period would also delay the realization of the benefits of final Rule 10c–1a, both those related to increased transparency in the securities lending market, as well as the regulatory benefits. Furthermore, while an initial delay in the public dissemination of the data would maintain the regulatory benefits of the rule, as regulators would still have access to the data, other market participants' access to the information would be delayed, which would delay the benefits of the Rule related to increased transparency. Thus, a choice of implementation period requires balancing the need to allow for sufficient time to build out reporting systems, on the one hand, and ensuring that market participants have timely access to Rule 10c–1a information on the other.

12. Report to the Commission Rather Than to an RNSA

The Commission also considered an alternative requiring that covered persons disclose Rule 10c–1a information directly to the Commission—for example, through EDGAR, rather than to an RNSA. Such an alternative could alter which entities incur costs and would likely increase overall costs relative to the final rule because, for example, many entities who possess reporting capabilities to an RNSA, e.g., members of FINRA, would need to establish comparable reporting relationships with the Commission. In particular, many broker-dealers already have connectivity to FINRA systems that support the kind of submission process required for providing Rule 10c–1a information. Establishing similar connectivity with EDGAR may require additional effort for covered persons compared to the proposal. Additionally, FINRA has expertise creating repositories similar to that called for in the final rule, suggesting that the final rule will likely be more efficient than this alternative.

The Commission is uncertain of how some benefits of this alternative would compare to certain benefits of the final rule. Specifically, while this alternative would not alter the content of the data in the final rule, the accessibility and timeliness of the data under this alternative depend on how the Commission would develop the functionality for distributing the data. In particular, we cannot at this time assess whether the alternative would result in more or less timely or accessible data or if the differences would be meaningful. For example, data obtained from the Commission could be less accessible if the Commission could not develop functionality allowing market participants to access the data with the same ease an RNSA could provide, given an RNSA has more experience collecting and disseminating similar data ( e.g., TRACE). Thus, this alternative could result in significantly higher compliance costs than the final rule.

Additionally, the regulatory benefits of the alternative relative to the final rule will depend on whether the Commission chose to grant an RNSA direct access to the confidential data. If the Commission chose to do so, then the regulatory benefits of this alternative would be the same as the current proposal. If the Commission chose not to grant an RNSA access to the confidential data, then the regulatory benefits would decline significantly as many of the regulatory benefits, such as improved monitoring of broker-dealers for compliance with various legal requirements, require access to the confidential data. Thus, the regulatory benefits of the rule could be severely diminished.

13. Report Through an NMS Plan

Because the nature of securities lending data is similar to the transaction data governed by the NMS data plans, such as the CT Plan, the Commission considered proposing a new NMS Plan to set up a reporting and dissemination process that mirrors the CT Plan. Reporting entities could report the data to Transaction Reporting Facilities operated by competing consolidators, who would consolidate and distribute the data for a fee. The NMS Plan would set the fee paid by (or rebate collected by) the reporting entities to (or from) the competing consolidators for the collection of the data. We did not receive public comments addressing this alternative.

This alternative is more likely than the final rule to improve the competitiveness of the market for securities lending data. Under this alternative, current data vendors, who have experience collecting and disseminating securities lending information, could compete as competing consolidators for securities lending data and have the same access to the supply of consolidated data as any other competing consolidator, including an RNSA or SRO. Conversely, as final Rule 10c–1a gives RNSAs the authority to implement rules regarding the format and manner of collection and dissemination of Rule 10c–1a data, the final rule may be more likely to result in a situation in which a current vendor may need to compete with an RNSA that has exclusive access to the raw Rule 10c–1a data. However, since final Rule 10c–1a does not require the reporting of shares available to lend, to the extent that commercial data vendors' superior access to information about shares available to lend and utilization rates provides them with a competitive advantage under the final rule, it may not necessarily be the case that this alternative would result in significantly better outcomes for current data vendors than the final rule. This alternative would, however, reduce the barriers to entry in the market for securities lending data because all new entrants would have access to the same data for consolidation and distribution.

This alternative could provide for the public dissemination of securities lending transaction information without the reliance on an RNSA alone. It could also leverage the processes of the NMS Plan, but would require compliance costs by one or more SROs that chose to set up and operate a Transaction Reporting Facility. Fees for reporting transactions could offset such compliance costs. While the Commission cannot be sure how these fees under this alternative would compare to the fees under the final rule, competition between competing consolidators could result in reporting facilities that are more efficient than that of an RNSA. However, as most SROs (including all of the national securities exchanges) do not necessarily have expertise that is specific to the dissemination of securities lending data, compliance costs may be higher than those proposed under the CT Plan, for which national securities exchanges do have relevant experience as they disseminate equity market data through their proprietary data feeds. A need to coordinate between multiple SROs in order to establish an NMS Plan may also imply higher coordination costs in this alternative as compared to the final rule.

X. Paperwork Reduction Act

Certain provisions of final Rule 10c–1a contain “collection of information” requirements within the meaning of the Paperwork Reduction Act of 1995 (“PRA”). The hours and costs associated with complying with the final rule's reporting requirements, RNSA rule implementation requirements, publication of data requirements, and data retention requirements, as applicable, constitute PRA burdens.

In accordance with the PRA, the Commission is submitting the final rule amendments to the Office of Management and Budget (“OMB”) for review. The Commission published a notice requesting comment on these collections of information requirements in the Proposing Release and submitted these requirements to the OMB for review in accordance with the PRA. An agency may not conduct or sponsor, and a person is not required to respond to, a collection of information unless it displays a currently valid control number. The title and control number for these collections of information are OMB Control No. 3235–0788.

The Commission received limited comments on the PRA burden estimates included in the Proposing Release. One commenter stated that the Commission has covered most of the impacts of the proposed rule. Another commenter stated that the proposed rule's reporting regime would require considerable time to develop, test, launch, and monitor. Comments on specific parts of the Proposing Release's PRA burden estimates are discussed below, in this part, as they relate to the particular PRA burden estimates that are the subject of those comments. The Commission acknowledges the PRA burdens that will be assumed by covered persons, reporting agents, and RNSAs in complying with the final rule, and its estimates of the collection of information for the amendments, as adopted, have been updated from the estimates included in the Proposing Release, as appropriate, with the updated estimates based on the modifications in the final rule, comments received, and more recently available data.

The information collections are necessary to remediate certain issues present in the securities lending market. As discussed above, in Part I, the Commission is adopting amendments to close securities lending data gaps and increase market efficiency. Final Rule 10c–1a is designed to increase the transparency of information available to brokers, dealers, and investors with respect to the loan or borrowing of securities. Additionally, final Rule 10c–1a is designed to provide an RNSA with data that might be used for in-depth monitoring and surveillance. Further, the data elements are designed to provide regulators with information to understand whether broker-dealers are building up risk; what strategies broker-dealers use to source securities that are lent to their customers; and what loans broker-dealers provide to their customers with fail to deliver positions.

To achieve these objectives, the Commission is adopting the requirements discussed above, in Parts VII.A through M. Under final Rule 10c–1a, certain persons are required to report information about securities loans to an RNSA. The final rule also requires that an RNSA make certain information provided to an RNSA, along with daily information pertaining to the aggregate transaction activity and distribution of loan rates for each reportable security, publicly available. Further, the final rule requires certain confidential information to be reported to an RNSA to enhance RNSA oversight and enforcement functions.

A. Respondents

As a preliminary matter, the opacity of the securities lending market makes estimating the number of respondents difficult. Indeed, as discussed above, an objective of Rule 10c–1a is to close the data gaps in this market. Despite these data gaps, the Commission has estimated the number of respondents who are a covered person, reporting agent, or RNSA.

As described in detail below, in this part, the respondents under final Rule 10c–1a are: (1) covered persons and (2) reporting agents capturing the applicable Rule 10c–1a information and providing it to an RNSA, in the format and manner required by the applicable rule(s) of such RNSA, and within the time periods specified in paragraphs (c) through (e) of the final rule, as applicable to the covered securities loan; and (3) an RNSA complying with the final rule's requirements to implement rules regarding the format and manner of its collection of Rule 10c–1a information, publish the information specified in final Rule 10c–1a(g) in accordance with the time frames required for such publication, and retain and make available to the Commission, or any other persons designated by the Commission, certain specified information. Given the differences in the information collections applicable to these parties, as discussed below, in this part, the discussion of burdens applicable to covered persons and reporting agents are separated from the discussion of those applicable to an RNSA.

1. Covered Persons

As discussed above, in Part VII.A, under final Rule 10c–1a(j)(1), a covered person is any intermediary other than a clearing agency when providing only the functions of a central counterparty pursuant to Rule 17Ad–22(a)(2) of the Exchange Act or a central securities depository pursuant to Rule 17Ad–22(a)(3) of the Exchange Act; any person that agrees to a covered securities loan as a lender when an intermediary is not used unless paragraph (j)(1)(iii) of the final rule applies; or a broker or dealer when borrowing fully paid or excess margin securities pursuant to Rule 15c3–3(b)(3) of the Exchange Act.

In the Proposing Release, the Commission estimated, based on a review of Forms N–CEN filed with the Commission that identify the lending agents used by investment companies, that there would be 37 lending agent intermediaries, 3 of which would provide information directly to an RNSA and 34 of which would provide information to a reporting agent. Using updated figures based on a review of Forms N–CEN filed with the Commission, it is estimated that there are 35 intermediaries. Of the 35 intermediaries identified, four are broker-dealers. Broker-dealers have experience providing information directly to RNSAs. The Commission estimates that those four intermediaries would provide information directly to an RNSA. The other 31 intermediaries are not broker-dealers. Therefore, the Commission estimates that they would provide information to a reporting agent rather than establish connectivity directly to an RNSA.

In the Proposing Release, the Commission estimated, based on the number of funds within investment companies that do not employ a lending agent based on a review of Forms N–CEN filed with the Commission, that there would be 278 lenders that would not employ a lending agent, 139 of which would provide information to an RNSA and 139 of which would provide information to a reporting agent. One commenter stated that this was an underestimate that accounted for only those lenders that are registered with the Commission. The Commission agrees with the commenter and is updating the estimate regarding unregistered entities, which comprise approximately 67 percent of the securities available for lending. The unregistered portion of the securities lending market, therefore, is estimated to be 1,389 entities. However, according to the OFR Pilot Survey, only about 15 percent of securities loans are not intermediated and go directly to end borrowers. Therefore, it is estimated that the number of unregistered lenders that would not use an intermediary is 208 entities. Accordingly, the Commission estimates that there are 434 persons that agree to a covered securities loan as the lender when an intermediary is not used, based on a review of Forms N–CEN filed with the Commission showing the number of investment companies that do not employ an intermediary. Consistent with the estimated ratio included in the Proposing Release, of these 434 persons, the Commission estimates that 217 persons will provide information to an RNSA (including by using a third-party vendor) and that 217 persons will provide information to a reporting agent.

As discussed above, in Part VII.A, paragraph (j)(1)(iii) was added to the proposed rule to specify that, if a broker or dealer is borrowing fully paid or excess margin securities in the covered securities loan, the broker or dealer is required to provide the Rule 10c–1a information to an RNSA. Based on a review of FOCUS Reports Part II, Item 4350, it is estimated that there are 34 brokers or dealers borrowing fully paid or excess margin securities pursuant to Rule 15c3–3(b)(3) of the Exchange Act. Consistent with the estimate for the number of intermediaries who are brokers or dealers and would provide information directly to an RNSA, the Commission estimates that the 34 brokers or dealers borrowing fully paid or excess margin securities would provide information directly to an RNSA.

Consistent with the estimate included in the Proposing Release, the Commission is not estimating that persons who agree to a covered securities loan on behalf of themselves or another person and employ a lending agent assume any PRA burdens because those persons would not be responsible for providing information to an RNSA. Accordingly, the Commission estimates that 503 covered persons will be subject to PRA burdens under final Rule 10c–1a(a). As set forth in paragraph (a)(1) of the final rule, any covered person who agrees to a covered securities loan on behalf of itself or another person is required to provide Rule 10c–1a information directly to an RNSA; however, a covered person may enter into a written agreement with a reporting agent for the reporting agent to provide information to an RNSA. Of the 503 covered persons, the Commission estimates that 255 are providing covered persons and that 248 are non-providing covered persons.

2. Reporting Agents

Under the proposed rule, the term “reporting agent” referred specifically to certain broker-dealers. In the Proposing Release, the Commission estimated, based on the number of broker–dealers that lent securities in 2020, that there would be 94 reporting agents. Under final Rule 10c–1a(j)(4), a reporting agent means a broker, dealer, or registered clearing agency that enters into a written agreement with a covered person under final Rule 10c–1a(a)(2). Based on the number of broker-dealers that lent securities as of December 2022 (97), as well as the number of registered clearing agencies in 2023 (9), it is estimated that there are 106 reporting agents that may be subject to PRA burdens under final Rule 10c–1a(b).

3. RNSA

Under final Rule 10c–1a(j)(5), an RNSA is an association of brokers and dealers that is registered as a national securities association pursuant to section 15A of the Exchange Act. Consistent with the estimate included in the Proposing Release, it is estimated that there is one RNSA that will be subject to PRA burdens under paragraphs (f), (g), and (h) of the final rule.

B. Collection of Information Related to Covered Persons

Final Rule 10c–1a applies to any covered person who agrees to a covered securities loan on behalf of itself or another person. Under final Rule 10c–1a(a)(1), a covered person who agrees to a covered securities loan on behalf of itself or another person shall provide to an RNSA the Rule 10c–1a information, in the format and manner required by the applicable rule(s) of such RNSA, and within the time periods specified in paragraphs (c) through (e) of the final rule. As discussed above, in this part, this type of covered person is referred to as a providing covered person. However, a covered person may rely on a reporting agent to fulfill the covered person's reporting obligations under final Rule 10c–1a(a) if the covered person: (1) enters into a written agreement with the reporting agent that agrees to provide the Rule 10c–1a information to an RNSA on behalf of such covered person, in accordance with the requirements set forth in Rule 10c–1a(b)(1), and (2), provides the reporting agent with timely access to the Rule 10c–1a information. As discussed above, in this part, this type of covered person is referred to as a non-providing covered person.

One commenter stated that the Commission should consider that not all “reporters” have licenses to multiple instrument identifiers required for reporting, meaning that smaller firms would need to purchase additional data licenses and thereby add to their cost structure. As discussed above, in this part, unlike providing covered persons who are estimated to have experience in providing information directly to an RNSA, persons who do not have the necessary licenses and systems are estimated to use a reporting agent as non-providing covered persons for purposes of the PRA estimates. Therefore, as non-providing covered persons, they are not estimated to acquire any additional licenses for purposes of compliance with the final rule.

The below analysis is separated into categories of persons who may provide the Rule 10c–1a information to an RNSA. These categories are providing covered persons and non-providing covered persons. Both providing covered persons and non-providing covered persons assume PRA burdens in complying with final Rule 10c–1a(a)(1).

1. Providing Covered Persons

In the Proposing Release, the Commission estimated that lending agents who would be required to provide Rule 10c–1 information to an RNSA and lenders who run their own securities lending program rather than employ a lending agent and provide Rule 10c–1 information directly to an RNSA would assume certain burdens related to developing and reconfiguring their current systems to capture the required Rule 10c–1 information. The Proposing Release referred to such persons as “providing lending agents” and “self-providing lenders,” respectively. The PRA burden estimates for providing lending agents and self-providing lenders are summarized in PRA Table 1 of the Proposing Release.

a. Initial Burden

Providing covered persons assume PRA burdens related to developing and reconfiguring their current systems to capture the required Rule 10c–1a information. Providing covered persons, in complying with the final rule, will also be required to establish connections that will allow them to provide the information to an RNSA, which involves establishing connections with an RNSA and the persons on whose behalf they are lending securities.

As stated in the Proposing Release, the PRA burden for this requirement is similar to that of establishing the appropriate systems and processes required for collection and transmission of the required information in complying with the CAT because of the general similarity between the systems established under that rule and the systems that would be required to be established under final Rule 10c–1a. One commenter argued that the Proposing Release's operational PRA burden estimates were inappropriate because the CAT would not capture loan modifications. However, the PRA burden estimates included in the Proposing Release accounted for loan modifications. Further, the Commission continues to believe that the CAT-related burden estimates serve as an appropriate basis for the PRA burdens associated with compliance with the final rule, as both the CAT and final Rule 10c–1a require the provision of trade information to a third-party information repository. The systems complying with final Rule 10c–1a will be significantly less complex than those required by the CAT because the systems required for compliance with final Rule 10c–1a will need to capture less information overall.

The PRA burden estimates for systems development and monitoring are based on the burdens applicable to non-Order Audit Trail System (“OATS”) reporters under the CAT. The Commission determined to use this estimate due to the factors it considered, as part of the CAT Approval Order, in categorizing firms and estimating burdens. Non-OATS reporters were estimated to assume the least amount of burdens under the CAT NMS Plan due to the limited scope of their reportable activity. In addition, non-OATS reporters assumed new reporting burdens in complying with the CAT, similar to how providing covered persons will assume new PRA burdens in complying with the final rule's reporting requirements. Based on the overall size of the securities lending market and the number of providing covered persons that may provide information directly to an RNSA, the Commission believes that the volume of securities lending transactions for providing covered persons will be, on average, of a similar scope to the volume of reports estimated by non-OATS reporters under the CAT Approval Order.

In the Proposing Release, the Commission, estimated that each providing lending agent and self-providing lender would assume 3,600 PRA burden hours in developing and reconfiguring their current systems to capture the required data elements, with a total industry-wide burden of 511,200 hours. One commenter stated that most lenders will not be able to meet the proposed 15 minute reporting requirement without substantial technology development and cost. As discussed above, in Part VII.G.1, the change from the proposed rule to impose an end-of-day reporting requirement should help to reduce the number and frequency of reports that would otherwise be required to be provided to an RNSA throughout the day. An intraday reporting system would have required a level of automation and data processing capacity that some providing covered persons would not currently have implemented. Implementing an end-of-day reporting requirement in place of the intraday reporting requirement removes the need for those providing covered persons to acquire this level of automation and processing capacity, thus lowering at least their initial burden associated with system design and configuration. Therefore, the Commission estimates that providing covered persons each will assume 3,000 PRA burden hours in developing and reconfiguring their current systems to capture the required data elements, with a total industry-wide burden of 765,000 hours.

b. Ongoing Annual Burden

Once a providing covered person has established the appropriate systems and processes required for the collection and provision of the Rule 10c–1a information to an RNSA, it is estimated that providing covered persons will assume ongoing annual PRA burdens associated with, among other things, providing the Rule 10c–1a information to an RNSA, monitoring systems, implementing systems changes, and troubleshooting errors. The Commission estimates that the ongoing annual PRA burden will be equivalent to the ongoing burden estimated for non-OATS reporters in the CAT Approval Order, as adjusted for the change from the proposed intraday reporting requirement to the end-of-day reporting requirement, for the same reasons discussed above with respect to automation and processing. In the Proposing Release, the Commission, estimated that each providing lending agent and self-providing lender would assume 1,350 PRA burden hours associated with monitoring systems, implementing systems changes, and troubleshooting errors, with a total industry-wide burden of 191,700 hours. Consistent with the Proposing Release, it is estimated that each providing covered person will assume 1,350 PRA burden hours per year, leading to a total industry-wide ongoing annual burden of 344,250 hours.

2. Non-Providing Covered Persons

In the Proposing Release, the Commission estimated that lending agents who enter into a written agreement with a reporting agent to provide information to an RNSA (“non-providing lending agents”) and lenders that directly employ a reporting agent would assume certain burdens related to developing and reconfiguring their current systems to capture the required Rule 10c–1a information, as well as in entering into an agreement with a reporting agent. The PRA burden estimates for non-providing lending agents and lenders that directly employ a reporting agent are summarized in PRA Table 1 of the Proposing Release. As discussed above, in Part IX.A.1, in light of the rule text changes from the proposed rule, the PRA burdens included in the Proposing Release with regard to such persons are estimated in this release with regard to non-providing covered persons.

Non-providing covered persons will assume distinct PRA burdens from those applicable to providing covered persons. First, non-providing covered persons will assume fewer initial and ongoing PRA burdens related to systems development and monitoring because non-providing covered persons will not, for purposes of compliance with final Rule 10c–1a(a), need to establish connectivity to an RNSA and may have flexibility with regard to the format in which it provides the Rule 10c–1a information to the reporting agent. Second, non-providing covered persons will assume the initial burden of negotiating and executing a written agreement with the reporting agent, as required by final Rule 10c–1a(a)(2) but are not estimated to assume an ongoing annual PRA burden associated with such written agreement.

a. Systems Development and Monitoring

i. Initial Burden

Non-providing covered persons will assume the initial burden of developing and reconfiguring their current systems to capture Rule 10c–1a information. Non-providing covered persons will assume fewer PRA burdens than providing covered persons will because non-providing covered persons may have the flexibility to collaborate with a reporting agent to determine the most efficient means of establishing systems that comply with the final rule's reporting requirements. For example, if agreed to by both the non-providing covered person and the reporting agent, the non-providing covered person could have the flexibility to provide to the reporting agent the applicable Rule 10c–1a information that does not meet the specific format requirements of an RNSA if the reporting agent is able to reformat the information once received. Given these and other potential efficiencies, the Commission estimates that a non-providing covered person will assume half of the initial burden hours that a providing covered person will assume to develop and reconfigure their current systems to capture the Rule 10c–1a information. The Commission, therefore, estimates that each non-providing covered person will assume an initial PRA burden of 1,500 hours, leading to a total industry-wide initial PRA burden for this requirement of 372,000 hours. This method of deriving the burden hours for non-covered persons is consistent with that used in the Proposing Release with regard to non-providing lending agents and lenders that directly employ a lending agent. However, as discussed above, in Part IX.B.1, the estimated number of initial burden hours assumed has been lowered in light of the change from the proposed intraday reporting requirement to an end-of-day reporting requirement.

ii. Ongoing Annual Burden

Once a non-providing covered person has established the necessary systems and processes for the collection and provision of the Rule 10c–1a information to the reporting agent, such person will assume ongoing annual PRA burdens associated with, among other things, providing the data to the reporting agent, monitoring systems, implementing systems changes, and troubleshooting errors. As with the initial PRA burden estimate for the systems development and monitoring requirement, the ongoing annual PRA burden estimate for non-providing covered persons is estimated to be less than that for providing covered persons because non-providing covered persons may have the flexibility to collaborate with a reporting agent to determine the most efficient means of establishing systems for purposes of compliance with the final rule. For example, the reporting agent could design programs that create direct links to a non-providing covered person's systems to facilitate the gathering of information such that ongoing intervention would not be required by the non-providing covered person. In addition, non-providing covered persons and reporting agents could negotiate terms that may allow the non-providing covered person to avoid providing certain Rule 10c–1a information that can be gleaned from another data element, such as an issuer's legal name if a security has a valid CUSIP.

Given these and other potential efficiencies, the Commission estimates that a non-providing covered person will assume half of the ongoing annual PRA burden that a providing covered person will assume with regard to the development and reconfiguration of current systems to capture the Rule 10c–1a information. The Commission, therefore, estimates that each non-providing covered person will assume an ongoing annual PRA burden of 675 hours, leading to a total industry-wide ongoing annual PRA burden for this requirement of 167,400 hours. This method of deriving the ongoing burden hours for non-providing covered persons is consistent with that used in the Proposing Release with regard to non-providing lending agents and lenders that directly employ a reporting agent.

b. Entering Into Written Agreement With Reporting Agent

Final Rule 10c–1a(a)(2)(i) requires a covered person to enter into a written agreement with a reporting agent in order to rely on the reporting agent to fulfill the covered person's reporting obligations under paragraph (a)(1) of the final rule. In meeting this requirement, non-providing covered persons may assume initial PRA burdens associated with drafting, negotiating, and executing the agreements.

These agreements are estimated to be standardized across the industry because the data elements are consistent for all persons. The Commission estimates that the only terms that may require negotiation are price and the format in which the information will be provided. To account for negotiation and any administrative tasks related to processing and executing agreements, the Commission is estimating that non-providing covered persons will spend 30 hours on this task. Accordingly, the Commission estimates that the total industry-wide initial PRA burden attributed to this requirement is 7,440 hours. However, consistent with the Proposing Release, there should be no associated ongoing annual PRA burden once the agreement is signed, as the final rule does not separately impose a requirement to modify the written agreement or take additional action after the agreement is executed.

D. Collection of Information Related to Reporting Agents

Under final Rule 10c–1a(b), any reporting agent that assumes the reporting obligation on behalf of a covered person pursuant to Rule 10c–1a(a)(2) shall do all of the following: (1) provide such Rule 10c–1a information to an RNSA, in the format and manner required by the applicable rule(s) of such RNSA, and within the time periods specified in paragraphs (c) through (e) of the final rule; (2) establish, maintain, and enforce written policies and procedures that are reasonably designed to provide the Rule 10c–1a information to an RNSA on behalf of the covered person in the format and manner required by the applicable rule(s) of an RNSA, and within the time periods specified in paragraphs (c) through (e) of the final rule; (3) enter into a written agreement with an RNSA that permits the reporting agent to provide Rule 10c–1a information to an RNSA on behalf of the covered person; (4) provide an RNSA with a list naming each covered person on whose behalf the reporting agent is providing Rule 10c–1a information to an RNSA and provide an RNSA with any updates to the list of such persons by the end of the day such list changes; and (5) preserve for a period of not less than three years, the first two years in an easily accessible place, the Rule 10c–1a information obtained by the reporting agent from the covered person pursuant to Rule 10c–1a(a)(2), including the time of receipt, and the corresponding Rule 10c–1a information provided by the reporting agent to an RNSA, including the time of transmission to an RNSA, and the written agreements under paragraphs (a)(2) and (b)(3) of the final rule.

Reporting agents will assume PRA burdens associated with their compliance with three requirements in final Rule 10c–1a(b). First, to facilitate the provision of Rule 10c–1a information to an RNSA, reporting agents will assume burdens related to the development and monitoring of systems. This includes establishing, maintaining, and enforcing written policies and procedures that are reasonably designed to provide the Rule 10c–1a information to an RNSA on behalf of the covered person in the format and manner required by the applicable rule(s) of an RNSA, and within the time periods specified in paragraphs (c) through (e) of the final rule. It also includes the provision of an updated list of persons on whose behalf they are providing Rule 10c–1a information. Second, reporting agents will assume burdens in entering into written agreements with non-providing covered persons. Third, reporting agents will assume burdens in entering into agreements with an RNSA to provide Rule 10c–1a information on behalf of a non-providing covered person.

1. Systems Development and Monitoring

Reporting agents will provide Rule 10c–1a information to an RNSA on behalf of non-providing covered persons. In doing so, because reporting agents will need to change their internal systems to collect the Rule 10c–1a information, they will assume initial PRA burdens related to developing and reconfiguring their current systems to capture the required data elements. In addition, reporting agents will need to establish, maintain, and enforce written policies and procedures that are reasonably designed to provide Rule 10c–1a information to an RNSA on behalf of non-providing covered persons, in the format and manner required by the applicable rule(s) of an RNSA, and within certain time periods specified paragraphs (c) through (e) of the final rule. Because reporting agents will provide to an RNSA the same type of Rule 10c–1a information that providing covered persons will provide, the Commission estimates that the initial and ongoing annual PRA burdens related to the development and monitoring of systems that facilitate the provision of Rule 10c–1a information to an RNSA are the same. Therefore,

consistent with the Commission's estimate included above, in Part IX.B.1, the Commission estimates that each reporting agent will assume an initial PRA burden of 3,000 hours in developing and reconfiguring current systems to capture the required data elements. Accordingly, the industry-wide initial PRA burden is estimated to be 318,000 hours.

Once a reporting agent has established the appropriate systems and processes required for the collection and provision of the applicable Rule 10c–1a information, the reporting agent will assume ongoing annual PRA burdens associated with providing such information to an RNSA, in addition to an updated list of persons on whose behalf they are providing information, monitoring systems, implementing changes, and troubleshooting errors. As discussed above, in this part, because reporting agents provide to an RNSA the same information that providing covered persons provide, the Commission estimates that each reporting agent will assume an ongoing annual PRA burden of 1,350 hours related to this requirement. Accordingly, the total industry-wide ongoing annual PRA burden is estimated to be 143,100 hours.

2. Entering Into Written Agreements With Non-Providing Covered Persons

To fulfill a covered person's reporting obligation under final Rule 10c–1a(a), any reporting agent must enter into a written agreement with the covered person on whose behalf they are providing Rule 10c–1a information to an RNSA. In meeting this requirement, reporting agents will assume initial PRA burdens related to drafting, negotiating, and executing such an agreement. Compliance with this requirement, however, should not create any ongoing annual burdens once the agreement is executed. This is because the final rule does not impose any requirement for reporting agents to modify the written agreement or take additional action after the agreement is executed.

These agreements are estimated to be standardized across the industry because the data elements are consistent for all persons. The only terms (of an agreement between a non-providing covered person and a reporting agent) that may require negotiation are the price and the format in which the information would be provided. This process is estimated to be highly automated. Reporting agents are estimated to require the same amount of time to comply with this requirement as non-providing covered persons will require. Accordingly, the Commission estimates that each reporting agent will spend 30 hours on this task. As a result, the total industry-wide initial PRA burden related to this requirement is estimated to be 3,180 hours.

3. Entering Into Written Agreements With an RNSA

Any reporting agent that assumes the reporting obligation on behalf of a covered person pursuant to final Rule 10c–1a(a)(2) must also enter into a written agreement with an RNSA that permits the reporting agent to provide Rule 10c–1a information to an RNSA on behalf of the covered person. Because the Rule 10c–1a information is standardized for all reporting agents, there are no terms of these agreements that will need to be negotiated. Instead, an RNSA may create a form agreement for all reporting agents. Because these agreements are estimated to be standardized, the Commission estimates that reporting agents will assume a one-hour initial PRA burden to account for any administrative tasks related to processing and executing these agreements. Reporting agents that enter into written agreements with an RNSA should not incur any ongoing annual burden to comply with this requirement once the agreement is executed, as the Rule 10c–1a information is standardized. Accordingly, the Commission estimates that the industry-wide initial PRA burden for this requirement is 106 hours.

4. Record Preservation Requirement

Any reporting agent that assumes the reporting obligation on behalf of a covered person pursuant to final Rule 10c–1a(a)(2) must also preserve for a period of not less than three years, the first two years in an easily accessible place, the Rule 10c–1a information it obtained from the covered person pursuant to paragraph (a)(2) of the final rule, including the time of receipt, and the corresponding Rule 10c–1a information provided by the reporting agent to an RNSA, including the time of transmission to an RNSA, and the written agreements that the reporting agent entered into with the covered person and with an RNSA. The initial PRA burden associated with preserving the collected Rule 10c–1a information is related to the reporting agent's burden of developing and reconfiguring its current systems to capture the required data elements. Therefore, an initial burden associated with the preservation of information under final Rule 10c–1a(b)(5) is not separately estimated.

Compliance with this record preservation requirement is estimated to be highly automated. The Commission, therefore, estimates that reporting agents will spend one hour per week on upkeep and testing of records to ensure accuracy in complying with this requirement. Reporting agents each will assume an ongoing annual PRA burden of 52 hours per year. Accordingly, the Commission estimates that the industry-wide ongoing annual PRA burden for this requirement is 5,512 hours.

E. Collection of Information Related to RNSAs

RNSAs will assume new PRA burdens in complying with final Rule 10c–1a(f) through (h). The PRA burdens associated with an RNSA's compliance with paragraphs (f) through (h) are discussed below. All estimates included in this Part X are consistent with the Commission's estimates included in the Proposing Release for an RNSA.

1. Burden Estimates Related to RNSA Rule Implementation

Under final Rule 10c–1a(f), an RNSA is required to implement rules regarding the format and manner of its collection of Rule 10c–1a information and make publicly available such information in accordance with rules promulgated pursuant to section 19(b) and Rule 19b–4 of the Exchange Act. The PRA burden associated with filing any proposed rule changes by an RNSA is already included under the collection of information requirements contained in Rule 19b–4 under the Exchange Act. Therefore, a separate PRA burden estimate is not included for the purposes of the PRA included for this rulemaking.

2. Burden Estimates Related to the Publication of Data

As discussed above, in Part VII.J, an RNSA will be required, following the receipt of information pursuant to paragraph (c) or (d) of the final rule, to assign a unique identifier to the covered securities loan and make certain information publicly available (on its website or similar means of electronic distribution, without use restrictions, for a period of at least five years) within the time frames specified in paragraph (g) of the final rule. In complying with these requirements, an RNSA will need to create, implement, and maintain the infrastructure to enable providing covered persons and reporting agents to provide an RNSA with the Rule 10c–1a information. This includes establishing technical requirements and specifications for such infrastructure, creating a system that can generate unique identifiers, meeting with industry participants to gather feedback on the proposed infrastructure, drafting written policies and procedures to protect the confidentiality of certain information, and entering into written agreements with providing covered persons and reporting agents. In addition, the infrastructure employed will need to comply with paragraphs (h)(1) and (h)(2) of the final rule, which require an RNSA to retain the collected Rule 10c–1a information in a convenient and usable standard electronic data format that is machine readable and text searchable without any manual intervention, for a period of five years, and to make the information collected pursuant to paragraphs (b)(4) and (c) through (e) of the final rule available to the Commission, or other persons as the Commission may designate by order upon a demonstrated regulatory need, respectively.

The initial burden assumed by an RNSA in creating and implementing the infrastructure for providing covered persons and reporting agents to provide the applicable Rule 10c–1a information to an RNSA, and for an RNSA to make such information publicly available, is similar to the initial burden assumed by national securities exchanges and RNSAs in complying with the requirement to establish the appropriate systems and processes for the collection and transmission of the required information under the CAT. However, the systems that are implemented to comply with Rule 10c–1a should be significantly less complex than those that are implemented to comply with the CAT. This is because the Rule 10c–1a systems will need to capture less information overall than what the CAT requires. Further, as discussed above, in Part IX.A.3, there is only one RNSA that will need to create and implement the infrastructure for providing covered persons and reporting agents to provide Rule 10c–1a information, in contrast to the multiple national securities exchanges that create systems to comply with the CAT. In addition, an RNSA will have internal staff that can create and implement the infrastructure for providing covered persons and reporting agents to provide Rule 10c–1a information, unlike certain tasks required under the CAT that may require outsourcing. Accordingly, the PRA burden estimates for this collection of information are substantially reduced as compared to those for the CAT, as discussed below, in this part.

The Commission estimates that it would take an RNSA approximately 10,924 hours of internal legal, compliance, information technology, and business operations time to develop the infrastructure to enable providing covered persons and reporting agents to provide the Rule 10c–1a information to an RNSA, and for an RNSA to assign a unique identifier to the covered securities loan and make the specified information publicly available. The RNSA is not estimated to assume external costs for the implementation of the infrastructure to enable providing covered persons and reporting agents to provide the Rule 10c–1a information, assign a unique identifier to the covered securities loan, and make the final rule's specified information publicly available. This is because the sole RNSA currently existing, FINRA, has experience in implementing systems to collect information from its member broker-dealers. Therefore, the Commission estimates that the average one-time initial PRA burden related to developing the infrastructure to enable providing covered persons and reporting agents to provide the Rule 10c–1a information, assign a unique identifier to the covered securities loan, and make the final rule's specified information publicly available is 10,924 hours.

Once an RNSA has developed the infrastructure to enable providing covered persons and reporting agents to provide the Rule 10c–1a information, assign a unique identifier to the covered securities loan, and make the final rule's specified information publicly available, the Commission estimates that an RNSA will assume ongoing annual PRA burdens of 7,739.5 hours related to ensuring that the infrastructure is up-to-date and remains in compliance with the final rule, for an estimated annual ongoing burden of 7,739.5 hours.

3. Burden Estimates Related to Data Retention and Availability

Under final Rule 10c–1a(h)(1), an RNSA must retain the collected Rule 10c–1a information in a convenient and usable standard electronic data format that is machine readable and text searchable without any manual intervention for a period of five years. The initial burden associated with retaining the collected Rule 10c–1a information is assumed in an RNSA's burdens related to implementing and maintaining the infrastructure for providing covered persons and reporting agents to provide Rule 10c–1a information to an RNSA, as discussed above, in Part X.2. Therefore, the Commission is not separately assessing an initial burden associated with the retention of collected Rule 10c–1a information. The Commission, however, estimates that an RNSA will assume an ongoing annual PRA burden of 52 hours to retain the collected information, for an estimated annual industry-wide burden of 52 hours.