Pay Versus Performance

AGENCY:

Securities and Exchange Commission.

ACTION:

Final rule.

SUMMARY:

The Securities and Exchange Commission (“Commission”) is adopting amendments to implement Section 14(i) (“Section 14(i)”) of the Securities Exchange Act of 1934 (“Exchange Act”), as added by Section 953(a) of the Dodd-Frank Wall Street Reform and Consumer Protection Act (“Dodd-Frank Act”). Section 14(i) directs the Commission to adopt rules requiring registrants to provide disclosure of pay versus performance. The disclosure is required in proxy or information statements in which executive compensation disclosure is required. The disclosure requirements do not apply to emerging growth companies, registered investment companies, or foreign private issuers.

DATES:

Effective date: This final rule is effective on October 11, 2022.

Compliance date: Companies (other than emerging growth companies, registered investment companies, or foreign private issuers) must begin to comply with these disclosure requirements in proxy and information statements that are required to include Item 402 of Regulation S-K (as defined below) disclosure for fiscal years ending on or after December 16, 2022.

FOR FURTHER INFORMATION CONTACT:

John Byrne, Special Counsel, Office of Small Business Policy, at (202) 551-3460, Division of Corporation Finance.

SUPPLEMENTARY INFORMATION:

The Commission is adopting an amendment to add new paragraph (v) to 17 CFR 229.402 (“Item 402 of Regulation S-K”); and amending 17 CFR 232.405 (“Item 405 of Regulation S-T”), 17 CFR 240.14a-101 (“Schedule 14A”), and 17 CFR 240.14c-101 (“Schedule 14C”), each under the Exchange Act.

Table of Contents

I. Introduction

A. Background

B. Overview of Final Amendments

II. Discussion of Final Amendments

A. New Item 402(v) of Regulation S-K

1. Application and Operation of Item 402(v) of Regulation S-K

i. Proposed Amendments

ii. Comments

iii. Final Amendments

2. Format and Location of Disclosure

i. Proposed Amendments

ii. Comments

iii. Final Amendments

B. Executives Covered

1. Proposed Amendments

2. Comments

3. Final Amendments

C. Determination of Executive Compensation Actually Paid

1. Deduction of Change in Actuarial Present Value and Addition of Actuarially Determined Service Cost and Prior Service Cost

i. Proposed Amendments

ii. Comments

iii. Final Amendments

2. Inclusion of Above-Market or Preferential Earnings on Deferred Compensation That Is Not Tax Qualified

i. Proposed Amendments

ii. Comments

iii. Final Amendments

3. Equity Awards

i. Proposed Amendments

ii. Comments

iii. Final Amendments

D. Measures of Performance

1. Requirement To Disclose TSR and Peer Group TSR

i. Proposed Amendments

ii. Comments

iii. Final Amendments

2. Requirement To Disclose Net Income

i. Amendments Considered in the Reopening Release

ii. Comments

iii. Final Amendments

3. Tabular List of the Registrant's “Most Important” Performance Measures

i. Amendments Considered in the Reopening Release

ii. Comments

iii. Final Amendments

4. Requirement To Disclose a Company-Selected Measure

i. Amendments Considered in the Reopening Release

ii. Comments

iii. Final Amendments

E. Time Period Covered

1. Proposed Amendments

2. Comments

3. Final Amendments

F. Permitted Additional Pay-Versus-Performance Disclosure

1. Proposed Amendments

2. Comments

3. Final Amendments

G. Required Disclosure for Smaller Reporting Companies

1. Proposed Amendments

2. Comments

3. Final Amendments

III. Other Matters

IV. Compliance Dates

V. Economic Analysis

A. Background

B. Baseline

1. Affected Parties

2. Existing Disclosures and Analyses

3. Executive Compensation Practices

C. Discussion of Economic Effects

1. Introduction

2. Benefits

3. Costs

4. Implementation Alternatives

i. Registrants and Filings Subject to the Disclosure Requirement

ii. General Disclosure Requirements

iii. Compensation Measures

iv. Performance Measures

VI. Paperwork Reduction Act

A. Background

B. Summary of Comment Letters and Revisions to PRA Estimates

C. Summary of Collection of Information Requirements

D. Incremental and Aggregate Burden and Cost Estimates for the Final Amendments

VII. Final Regulatory Flexibility Analysis

A. Need For, and Objectives of, the Final Rules

B. Significant Issues Raised by Public Comments

C. Small Entities Subject to the Final Amendments

D. Projected Reporting, Recordkeeping, and Other Compliance Requirements

E. Agency Action To Minimize Effect on Small Entities

Statutory Authority and Text of Amendments

I. Introduction

A. Background

Section 953(a) of the Dodd-Frank Act (“Section 953(a)”) added Section 14(i) to the Exchange Act. Section 14(i) mandates that the Commission shall, by rule, require each issuer to disclose in any proxy or consent solicitation material for an annual meeting of the shareholders of the issuer a clear description of any compensation required to be disclosed by the issuer under Item 402 of Regulation S-K (or any successor thereto), including, for any issuer other than an emerging growth company, information that shows the relationship between executive compensation actually paid and the financial performance of the issuer, taking into account any change in the value of the shares of stock and dividends of the issuer and any distributions. Section 14(i) also states that an issuer may include a graphic representation of the information required to be disclosed.

As a part of the Dodd-Frank Act legislative process, in a 2010 report, the Senate Committee on Banking, Housing and Urban Affairs stated that the disclosure required under Section 14(i) “may take many forms.” In addition, the report indicated that the relationship between executive pay and performance has become a “significant concern of shareholders,” and that the required disclosure should “add to corporate responsibility,” as registrants will be required to provide clearer executive pay disclosures.

In 2015, the Commission proposed a new rule to implement Section 953(a) by creating a new requirement in Item 402 of Regulation S-K. The proposed new item would require a registrant to provide a clear description of (1) the relationship between executive compensation actually paid to the registrant's named executive officers (“NEOs”) (including the registrant's principal executive officer (or persons acting in a similar capacity during the last completed fiscal year) (“PEO”)) and the cumulative total shareholder return (“TSR”) of the registrant, and (2) the relationship between the registrant's TSR and the TSR of a peer group chosen by the registrant, over each of the registrant's five most recently completed fiscal years. The comment period for the Proposing Release was reopened in 2022 to permit commenters to further analyze and comment upon the proposed rules in light of developments since the publication of the Proposing Release and our further consideration of the Section 953(a) mandate. In the Reopening Release, we stated that we were considering, and requested public comment on, certain additional disclosure requirements that may better implement the Section 953(a) mandate by providing investors with additional decision-relevant data.

We believe the disclosure mandated by Section 953(a) is intended to provide investors with more transparent, readily comparable, and understandable disclosure of a registrant's executive compensation, so that they may better assess a registrant's executive compensation program when making voting decisions, for example when exercising their rights to cast advisory votes on executive compensation under Exchange Act Section 14A or electing directors. This belief is supported by the fact that Section 953(a) was enacted contemporaneously with other executive compensation-related provisions in the Dodd-Frank Act that are “designed to address shareholder rights and executive compensation practices.” These included Section 951 of the Dodd-Frank Act, which enacted new Exchange Act Section 14A, and Section 953(b) of the Dodd-Frank Act. These provisions required, respectively, that, not less than every three years, a separate resolution be put to a non-binding shareholder vote to approve compensation of executives; and that registrants provide disclosure of the ratio of the median annual total compensation of employees to the annual total compensation of the chief executive officer.

We believe the disclosure mandated by Section 14(i) will allow investors to assess a registrant's executive compensation actually paid relative to its financial performance more readily and at a lower cost than under the existing executive compensation disclosure regime. Under Item 402 of Regulation S-K, which specifies the information that must be included when the applicable form or schedule requires executive compensation disclosure, specific information regarding financial performance is already required, including in the Performance Graph in 17 CFR 229.201(e) (“Item 201(e) of Regulation S-K”), the Supplementary Financial Information in 17 CFR 229.302 (Item 302), and Management's Discussion and Analysis of Financial Condition and Results of Operations in 17 CFR 220.303 (Item 303). In addition, Item 402 of Regulation S-K also requires detailed disclosure of executive compensation and principles-based disclosure requirements regarding the relationship between pay and performance.

There is no single place, however, where issuers must provide investors with direct comparisons of an executive's pay with their company's performance, and specifically financial performance, particularly if investors are interested in that comparison over a timespan longer than the most recent reporting period. Existing disclosures generally provide the necessary components to make these comparisons, including data required for calculations that aid in these comparisons, but doing so may be time-consuming and costly. We believe this information is important to investors in evaluating executive compensation, and that disclosures about executive compensation may be most meaningful to investors when placed in the context of the company's financial performance. Indeed, we are aware that certain third parties ( e.g., proxy advisors or compensation consultants) perform such analyses and charge clients for access to the resulting data. Requiring registrants to compute and report this information will make this information equally accessible to all investors in a consistent manner.

By specifically referencing disclosure of “information that shows the relationship between executive compensation actually paid and . . . financial performance of the issuer,” Section 14(i) calls for information that will supplement management's discussion of material elements of executive compensation in the CD&A. In addition, we believe this disclosure will provide investors with important and decision-useful information for comparison purposes in one place when they evaluate a registrant's executive compensation practices and policies, including for purposes of the shareholder advisory vote on executive compensation, votes on other compensation matters, director elections, or when making investment decisions.

Section 14(i) did not expressly prescribe the manner in which issuers would disclose the required information and we have exercised our discretion to provide for a consistent format that we believe furthers the statutory objectives of making pay-versus-performance data clear and easy for investors to evaluate. Standardizing the format and presentation of data, in particular quantitative metrics, to promote such ease of use requires incremental costs for issuers. We have elected not to pursue a wholly principles-based approach because, among other reasons, such a route would limit comparability across issuers and within issuers' filings over time, as well as increasing the possibility that some issuers would choose to report only the most favorable information. In addition, as we describe more extensively below, the final rules require that issuers calculate the value of certain equity and pension awards in more detail than would have been required in the proposed rule. These changes, in our view, will result in disclosures that more accurately represent the time when the awards change in value, which is important for investors to be able to assess whether such changes correspond to company performance over the appropriate time period.

We received many comment letters in response to the Proposing Release and the Reopening Release. After taking into consideration these public comments, we are adopting the proposed rules, together with certain of the supplemental disclosure requirements considered in the Reopening Release, with some modifications to reflect public comment. As discussed in more detail below, the final rules require registrants to present disclosure that reflects the specific situation of the registrant with respect to pay-versus-performance, and while also providing pay-versus-performance disclosure that can be readily compared across registrants.

B. Overview of Final Amendments

The amendments add new 17 CFR 229.402(v) (“Item 402(v) of Regulation S-K”), which requires registrants to describe the relationship between the executive compensation actually paid by the registrant and the financial performance of the registrant over the time horizon of the disclosure. Item 402(v) of Regulation S-K requires disclosure of the cumulative TSR of the registrant (substantially as defined in Item 201(e) of Regulation S-K), the TSR of the registrant's peer group, the registrant's net income, and a measure chosen by the registrant and specific to the registrant (“Company-Selected Measure”) as the measures of financial performance.

The final rules require the following tabular disclosures, with the asterisked items indicating portions of the final rules from which smaller reporting companies (“SRCs”) are exempt:

In addition, registrants are required to use the information in the above table to provide clear descriptions of the relationships between compensation actually paid and three measures of financial performance, as follows: describe the relationship between (a) the executive compensation actually paid to the registrant's PEO and (b) the average of the executive compensation actually paid to the registrant's remaining NEOs to (i) the cumulative TSR of the registrant, (ii) the net income of the registrant, and (iii) the registrant's Company-Selected Measure, in each case over the registrant's five most recently completed fiscal years. Registrants are also required to provide a clear description of the relationship between the registrant's TSR and the TSR of a peer group chosen by the registrant, also over the registrant's five most recently completed fiscal years. Registrants have flexibility as to the format in which to present the descriptions of these relationships, whether graphical, narrative, or a combination of the two. Registrants will also have the flexibility to decide whether to group any of these relationship disclosures together when presenting their clear description disclosure, but any combined description of multiple relationships must be “clear.” SRCs will only be required to present such clear descriptions with respect to the measures they are required to include in the table and for their three, rather than five, most recently completed fiscal years.

A registrant that is not an SRC also will be required to provide an unranked list of the most important financial performance measures used by the registrant to link executive compensation actually paid to the registrant's NEOs during the last fiscal year to company performance. Although, as discussed below, registrants may include non-financial performance measures in this list, they must select the Company-Selected Measure from the financial performance measures included in this list, and it must be the financial performance measure that in the registrant's assessment represents the most important performance measure (that is not otherwise required to be disclosed in the table) used by the registrant to link compensation actually paid to the registrant's NEOs, for the most recently completed fiscal year, to company performance.

As discussed below, the final rules permit registrants to voluntarily provide supplemental measures of compensation or financial performance (in the table or in other disclosure), and other supplemental disclosures, so long as any such measure or disclosure is clearly identified as supplemental, not misleading, and not presented with greater prominence than the required disclosure.

The final rules apply to all reporting companies except foreign private issuers, registered investment companies, and emerging growth companies (“EGCs”). As proposed, BDCs will be treated in the same manner as issuers other than registered investment companies and, therefore, be subject to the disclosure requirement of new Item 402(v) of Regulation S-K.

II. Discussion of Final Amendments

A. New Item 402(v) of Regulation S-K

1. Application and Operation of Item 402(v) of Regulation S-K

i. Proposed Amendments

We proposed including the pay-versus-performance disclosure in a new Item 402(v) of Regulation S-K, as Section 14(i) explicitly refers to Item 402 of Regulation S-K as the reference point for the executive compensation to be addressed by the new disclosure relating compensation to performance. We proposed requiring registrants to include the Item 402(v) of Regulation S-K disclosure in any proxy or information statement for which disclosure under Item 402 of Regulation S-K is required. By including the requirement in Item 402 of Regulation S-K and requiring this disclosure in proxy statements on Schedule 14A and in information statements on Schedule 14C, shareholders would have available the pay-versus-performance disclosure, along with all other executive compensation disclosures called for by Item 402 of Regulation S-K, in circumstances in which shareholder action is to be taken with regard to executive compensation or an election of directors.

Because the language of Section 14(i) calling for the disclosure to be provided in solicitation material for an annual meeting of the shareholders suggests that the disclosure was intended to be provided in conjunction with a shareholder vote, we proposed limiting the requirement to provide these disclosures to a registrant's proxy or information statement, instead of in all filings where disclosure under Item 402 of Regulation S-K is required (which would also include a registrant's Form 10-K and Securities Act registration statements). In addition, as proposed, the information would not be deemed to be incorporated by reference into any filing under the Securities Act or the Exchange Act, except to the extent that the registrant specifically incorporates it by reference.

ii. Comments

Some commenters generally supported the proposed approach, with one noting that including the disclosure in proxy and information statements would provide “relevant information at a time when (a) it is most useful to shareowners and (b) shareowners are equipped to act on the information if they are so inclined.” One commenter suggested that the Commission limit the requirement to include the pay-versus-performance information to proxy statements only, noting that any other document could just make reference to the proxy statement; while another commenter suggested the pay-versus-performance information “should be included in all materials/filings that discuss compensation.”

iii. Final Amendments

As proposed, we are adopting the requirement to include the new Item 402(v) of Regulation S-K disclosure in any proxy or information statement for which disclosure under Item 402 of Regulation S-K is required. As noted by commenters and in the Proposing Release, placing the pay-versus-performance information in proxy statements and information statements will provide shareholders with the pay-versus-performance disclosure (along with all other executive compensation disclosures called for by Item 402 of Regulation S-K) in circumstances in which shareholder action is to be taken with regard to an election of directors or executive compensation. We are not requiring the pay-versus-performance disclosure in other filings where disclosure under Item 402 of Regulation S-K is required, as we believe that, taken in context, the language of Section 14(i) calling for registrants to provide the disclosure “in any proxy or consent solicitation material for an annual meeting of the shareholders” suggests that the information was intended to be presented in conjunction with a shareholder vote.

2. Format and Location of Disclosure

i. Proposed Amendments

Section 14(i) requires us to adopt rules requiring disclosure of “information” that shows the relationship between executive compensation actually paid and registrant financial performance, but it does not specify the format or location of that disclosure. We proposed allowing registrants to decide where in the proxy or information statement to provide the required disclosure. Although the new disclosure item would show the historical relationship between executive pay and registrant financial performance, and may provide a useful point of comparison for the analysis provided in the CD&A, the Proposing Release indicated that it would be appropriate to provide flexibility for registrants in determining where in the proxy or information statement to provide the disclosure.

We proposed requiring registrants to provide a standardized table containing the values of:

• The total PEO compensation reported in the Summary Compensation Table;
• The value of executive compensation actually paid to the PEO;
• For NEOs (other than the PEO), the average total compensation reported in the Summary Compensation Table;
• The value of the average executive compensation actually paid to the NEOs (other than the PEO);
• The value of a fixed investment scaled by cumulative TSR, for the registrant; and
• The value of a fixed investment scaled by cumulative TSR for the selected peer group.

For the amounts disclosed as executive compensation actually paid, we proposed requiring footnote disclosure of the amounts that were deducted from, and added to, the Summary Compensation Table total compensation amounts to calculate the executive compensation actually paid, and footnote disclosure of vesting date valuation assumptions.

Because the statute specifically references disclosure of the relationship between executive compensation actually paid and registrant's financial performance, we proposed requiring registrants, using the values presented in the table, to describe (1) the relationship between the executive compensation actually paid and registrant TSR, and (2) the relationship between registrant TSR and peer group TSR. The disclosure about the relationship would follow the table and could be described as a narrative, graphically, or a combination of the two.

In the Reopening Release, we requested comment on requiring the tabular disclosure to include disclosure of income or loss before income tax expense, net income, and a Company-Selected Measure. We also requested comment on requiring registrants to provide a clear description of the relationship of each of these additional measures to executive compensation actually paid, but, consistent with the relationship descriptions proposed with respect to TSR and peer group TSR, allowing the registrant to choose the format used to present the relationship, such as a graphical or narrative description (or a combination of the two).

We also proposed that the disclosure be provided in interactive data format using machine-readable eXtensible Business Reporting Language (“XBRL”). Specifically, the proposal would require registrants to tag separately the values disclosed in the required table, and to separately block-text tag the required relationship disclosure and the footnote disclosures. In the Reopening Release, we requested comment on whether we should require registrants also to tag specific data points (such as quantitative amounts) within the footnote disclosures that would be block-text tagged, and to use Inline XBRL rather than XBRL to tag their pay-versus-performance disclosure.

ii. Comments

Commenters were divided over whether we should require registrants to include the pay-versus-performance disclosure in the CD&A, or allow registrants to decide where in the proxy or information statement to provide the required disclosure, as proposed. ³⁷ Commenters in favor of allowing registrants to decide where to provide the disclosure argued that including the disclosure in the CD&A could cause confusion, as registrants do not necessarily consider the information included in the pay-versus-performance disclosure when making decisions about executive compensation. Those in favor of locating the disclosure in the CD&A stated that locating the disclosure alongside other executive compensation disclosure would make the disclosure easier to locate for investors and provide investors the ability to more easily assess the pay-versus-performance disclosure.

Commenters were also divided on the proposal to require the disclosure in a tabular format. Some commenters generally supported the proposed tabular disclosure, while others opposed the tabular format, suggesting it was overly simplistic and would require significant supplemental disclosures.

We received significant comment on the specific performance measures to be included in the table, as discussed in Section II.E below. With respect to the other information proposed to be provided in the tabular format, one commenter suggested dividing the table to separate the TSR disclosure from the compensation actually paid disclosure. In addition, some commenters opposed requiring disclosure of the total compensation from the Summary Compensation Table, with one stating that “including the SCT data would result in redundancy, would add a second figure which is not representative of compensation actually paid, and could result in possible confusion to shareholders.” However, other commenters supported the inclusion of the Summary Compensation Table total compensation figures, with one suggesting that including the Summary Compensation Table figures would help investors understand the pay-versus-performance disclosure alongside the Summary Compensation Table disclosure when evaluating a registrant's annual compensation decisions, and another noting that the Summary Compensation Table figures “will help to clarify potential differences between reported compensation and compensation actually paid.”

A number of commenters suggested that we require or allow graphical disclosures. Some commenters suggested requiring graphical disclosure, while one specifically supported giving registrants the flexibility to choose whether to include graphical disclosure. A few of these commenters suggested requiring inclusion of the performance graph required in Item 201(e) of Regulation S-K, or a modified version of that graph. In addition, a few commenters suggested the Commission mandate formatting requirements for graphical disclosure, if graphical disclosure is permitted. One commenter suggested that we replace the tabular disclosure requirement with a graphical disclosure requirement depicting TSR and compensation actually paid, while another commenter stated that a prescribed graphical format would facilitate comparability.

One commenter generally supported the requirement to provide a clear description of the relationship between the measures disclosed in the table and executive compensation, stating that a “simple-to-understand approach would be particularly valuable to investors.” Another commenter, who supported requiring disclosure only of one (or more) Company-Selected Measure(s), indicated that registrants should be required to provide a clear description of the relationship between the Company-Selected Measure(s) in the table and executive compensation.

Commenters were divided on the proposed XBRL tagging requirement. Of the commenters who opposed the requirement, some made alternative suggestions such as only requiring block-tagging, only requiring tagging of the information in the table, delaying the implementation of the tagging requirement, or permitting but not requiring tagging. One commenter stated the Commission should proceed “cautiously” to ensure that the cost of tagging does not outweigh the benefits, while another suggested the Commission should provide data on how many investors use XBRL disclosures before implementing the requirement. However, a number of commenters supported the XBRL requirement, with one suggesting that tagging should be required for the actual metrics registrants use to determine executive compensation.

In response to the Reopening Release request for comment regarding Inline XBRL, a number of commenters suggested requiring all registrants to use Inline XBRL to tag their pay-versus-performance disclosure, including the tagging of specific data points within the footnote disclosures that would be block-text tagged. One commenter directly opposed requiring the use of the Inline XBRL (as considered in the Reopening Release), while another commenter, who generally opposed an XBRL tagging requirement, stated that, if XBRL tagging is required, Inline XBRL tagging should be permitted. One commenter suggested the Commission give time for registrants to implement any XBRL requirements, due to the “stylized” nature of proxy statements, and that there may be a learning curve because registrant staff preparing the proxy statement may be different from the staff preparing documents that are subject to current tagging requirements.

iii. Final Amendments

The final rules provide registrants flexibility in determining where in the proxy or information statement to provide the disclosure required, as proposed. We believe, as noted in the Proposing Release and by some commenters, that mandating registrants to include the disclosure in the CD&A may cause confusion by suggesting that the registrant considered the pay-versus-performance relationship in its compensation decisions, which may or may not be the case.

We are adopting the tabular disclosure format, as proposed, with the addition of two new financial performance measures—net income and the Company-Selected Measure—as considered in the Reopening Release. Each of these financial performance measures is discussed in more detail below. We are not persuaded by commenters who characterized the tabular disclosure requirement as overly simplistic. The simplicity of the tabular disclosure should allow investors to more easily understand and analyze the relationship between pay and performance. In addition, registrants can supplement the tabular disclosure, so long as any additional disclosure is clearly identified as supplemental, not misleading, and not presented with greater prominence than the required disclosure. We also believe the simplicity of the tabular disclosure matches the requirement in Section 14(i) that registrants provide a “clear description” of their pay-versus-performance, and, consistent with Section 14(i), will better allow investors to compare disclosures within companies over time and across companies, making the disclosure more useful.

We are adopting the requirement to include the Summary Compensation Table total compensation amounts for the PEO and the average ( i.e., mean) of the remaining NEOs, as proposed. Those amounts will appear in columns (c) and (e) of the Pay Versus Performance table, respectively. We believe including these figures as proposed will provide useful information to investors, especially as the “actually paid” figures are directly related to those figures. Requiring disclosure of the Summary Compensation Table measure of total compensation together with executive compensation actually paid will provide shareholders with disclosure of two measures in one single table and, we believe, will facilitate comparisons of the two measures of a registrant's executive compensation to the registrant's performance. For example, to the extent that some shareholders may be interested in considering the relationship of performance with a measure of pay that excludes changes in the value of equity awards, they would be able to refer to the Summary Compensation Table measure of total compensation alongside executive compensation actually paid in the tabular disclosure. As proposed, the final rules will require registrants to provide footnote disclosure of the amounts that are deducted from, and added to, the Summary Compensation Table total compensation amounts reported in columns (c) and (e) to calculate the executive compensation actually paid amounts reported in columns (d) and (f), respectively. We believe any confusion created by the inclusion of the Summary Compensation Table totals in the table will be mitigated by this required footnote disclosure.

As proposed, registrants must also provide a narrative, graphical, or combined narrative and graphical description of the relationships between executive compensation actually paid and the registrant's TSR, and between the registrant's TSR and peer group TSR. We believe the disclosure of the relationship between executive compensation actually paid and TSR will satisfy the language of Section 14(i) that registrants disclose the “relationship” between executive compensation and registrant performance. Further, as noted in the Proposing Release, we believe disclosure about the relationship between registrant TSR and peer group TSR may provide a useful point of comparison to assess the relationship between the registrant's executive compensation actually paid and its financial performance compared to the performance of its peers during the same time period.

In light of the addition of two new performance measures to the table, we are also adopting a requirement that registrants provide a clear description of the relationships between executive compensation actually paid and net income, and between executive compensation actually paid and the Company-Selected Measure. These descriptions may also be provided in narrative, graphical, or combined narrative and graphical format. Since some of these measures and relationships may be more important to some companies or investors than others, we believe including disclosure about each of these relationships will provide investors with a more complete picture of how pay relates to performance.

We believe permitting, but not mandating, graphical disclosure is consistent with an acknowledgement in the Senate Report that there could be many ways to disclose the relationship between executive compensation and financial performance of the registrant, and the specific language of Section 14(i), which provides the pay-versus-performance disclosures “may” include graphic representations. We encourage registrants to present this disclosure in the format that most clearly provides information to investors about the relationships, based on the nature of each measure and how it is associated with executive compensation actually paid. As discussed in the Proposing Release, the required relationship disclosure could include, for example, a graph providing executive compensation actually paid and change in the financial performance measure(s) (TSR, net income, or Company-Selected Measure) on parallel axes and plotting compensation and such measure(s) over the required time period. Alternatively, the required relationship disclosure could include narrative or tabular disclosure showing the percentage change over each year of the required time period in both executive compensation actually paid and the financial performance measure(s) together with a brief discussion of how those changes are related. The required table, along with the required relationship disclosures, should provide investors with clear information from which to determine the relationship between executive compensation actually paid and some basic facets of registrant financial performance. In addition, although the presentation format used by different registrants to demonstrate the relationship between executive compensation actually paid and the financial performance measures included in the table pursuant to Item 402(v) of Regulation S-K may vary, these more variable descriptions may allow investors to understand more easily the registrant's perspective on these required relationship disclosures.

The final rules require registrants to separately tag each value disclosed in the table, block-text tag the footnote and relationship disclosure, and tag specific data points (such as quantitative amounts) within the footnote disclosures, all in Inline XBRL. We recognize that, as noted by commenters, the requirement that registrants use Inline XBRL will increase costs for registrants. However, we believe these costs will be incremental, as registrants are subject to Inline XBRL tagging requirements for other Commission disclosures. In addition, we believe that requiring the data to be structured will lower the cost to investors of collecting this information, permit data to be analyzed more quickly, and facilitate comparisons among public companies, all of which justify the incremental cost to registrants. We also believe that the registrants who will be subject to the pay-versus-performance rule are familiar with Inline XBRL, and for that reason do not believe additional data about the complexity of Inline XBRL, or a phase-in period for the application of the requirement (other than as proposed for SRCs, as discussed below ), are necessary. With respect to comments questioning the utility of a structured data language, we note that investors and market participants have gained experience with XBRL and Inline XBRL filings since the time of the Proposing Release, and that there is increased evidence that data in these formats is useful to investors.

B. Executives Covered

1. Proposed Amendments

Under the approach included in the Proposing Release, registrants other than SRCs would have been required to provide disclosure about “named executive officers,” as defined in 17 CFR 229.402(a)(3); and SRCs would have been required to provide disclosure about “named executive officers,” as defined in 17 CFR 229.402(m). These are the executive officers for whom, under our current rules, compensation disclosure is required under Item 402 of Regulation S-K, including in the Summary Compensation Table and the other executive compensation disclosure requirements. Specifically, we proposed requiring registrants to separately disclose compensation information for the PEO, and as an average for the remaining NEOs. We also proposed that, if more than one person served as the PEO of the registrant in any year, the disclosure for those multiple PEOs would be aggregated for that year, because this reflects the total amount that was paid by the registrant for the services of a PEO.

2. Comments

A number of commenters supported requiring Item 402(v) of Regulation S-K to cover both PEOs and NEOs. These commenters noted that requiring Item 402(v) of Regulation S-K to cover PEOs and NEOs would be consistent with the disclosure in the Summary Compensation Table, and what Congress intended; and would provide investors with useful information about the registrant's compensation practices more broadly. However, a number of other commenters suggested we limit the disclosure to PEOs. Such commenters raised concerns about the inclusion of non-PEO NEOs, including that: NEO groups may vary considerably from year to year; NEOs are more likely to have business-segment-based compensation, the performance of which might not be reflective of the registrant's overall performance; and not all NEOs are in positions to affect overall company performance. Commenters also stated that PEOs are under the most scrutiny from investors and are the only executives comparable across companies; and that requiring disclosure of non-PEO NEOs would create an increased reporting burden. In addition, one commenter expressed belief that Section 14(i) did not require the pay-versus-performance disclosures to include non-PEO NEOs.

Commenters were generally opposed to the proposal's approach of aggregating multiple PEOs for years when a registrant had more than one individual serve as PEO. These commenters proposed a number of alternatives to aggregation, including: allowing separate disclosure for each PEO; only requiring aggregation for external successors; only disclosing the compensation of the PEO serving at the end of the year (either annualized or not ); requiring disclosure of the outgoing PEO only; only aggregating payments for services rendered as PEO; requiring aggregated and disaggregated disclosures; or excluding any disclosures in years where the registrant has multiple PEOs. Additionally, a number of commenters opposed including signing and severance bonuses, either generally, or if the compensation of multiple PEOs were to be aggregated, while some other commenters more specifically stated that these bonuses were reasons not to aggregate PEO compensation.

A few commenters also opposed using the average NEO compensation in the table, while others supported average NEO compensation. A number of other commenters did not expressly oppose the use of average NEO compensation, but stated that this type of disclosure would provide little investor insight, could confuse investors, or would limit comparability. Two commenters suggested requiring separate disclosure for each NEO.

3. Final Amendments

We are adopting requirements for registrants to disclose information pertaining to both NEOs and PEOs in their Item 402(v) of Regulation S-K disclosure, as proposed. As noted in the Proposing Release, Section 14(i) does not specify which executives must be included in the pay-versus-performance disclosure. While we are mindful of concerns raised by commenters that individual NEOs may be in positions less likely to affect overall company performance than the PEO, may have more varied performance measures driving their compensation (including because NEOs within a company have different roles), can vary from year to year, and are less comparable across registrants (with respect to compensation), we believe that Congress intended for the rules to provide disclosure about both PEOs and the remaining NEOs because Section 14(i) specifically refers to “compensation required to be disclosed by the issuer under [Item 402 of Regulation S-K],” and Item 402 requires disclosure of NEO compensation. Further, while we agree that investors are typically most interested in the compensation of the PEO, as indicated by commenters, investors also are interested in how the incentives of NEOs relate to company performance, and our rationale of simplifying and reducing costs for investors who monitor executive performance therefore extends to NEOs.

We are also adopting, as proposed, the requirement that registrants provide separate disclosure of the PEO's compensation. We believe this is appropriate because, as noted by commenters, investors frequently have more interest in PEO compensation, PEOs are generally more comparable across companies, and PEOs are frequently in a position to impact performance more than any other NEO.

Similarly, we are adopting as proposed a requirement to include an average of compensation for the remaining NEOs. We disagree with commenters that suggested that average NEO compensation would provide little investor insight, could confuse investors, or would limit comparability. Rather, we believe disclosure of the relationship of performance to average NEO compensation will be more meaningful to shareholders than individual or aggregate NEO compensation. Because a registrant's individual NEOs may change from year to year, we believe that the disclosure of the average NEO compensation will make it easier for investors to compare the registrant's pay-versus-performance disclosure over time. Further, we believe disclosure of compensation for all NEOs (consisting of the PEO, and the remaining NEOs in the aggregate) aligns with our understanding of the intent of Congress that all NEOs be included in the pay-versus-performance disclosure. In addition, we are adopting a requirement that registrants identify in footnote disclosure the individual NEOs whose compensation amounts are included in the average for each year, so that investors can consider whether changes in the average compensation reported from year to year were due to compositional changes in the included NEOs. We believe this will alleviate concerns raised by commenters that the aggregation of NEOs could confuse investors.

Although some commenters opposed our proposal to require an average of NEO compensation and suggested that we instead require the disclosure of compensation for each of the NEOs as separate columns in the table, we believe that approach could result in a lengthy and potentially confusing table, due to the fact that in any year there are multiple NEOs and, as noted by several commenters, there can be frequent turnover in a registrant's NEOs from year to year. In addition, we are not permitting registrants to remove signing bonuses, severance bonuses, and other one-time payments from the amount of executive compensation actually paid, because, although those figures may not represent the executive's compensation in a `typical' year where no such payment is made, they do reflect amounts that are “actually paid” to the executives. Even if such payments are not ordinarily recurring with respect to a particular executive, shareholders voting on executive compensation or directors may wish to take into account the company resources devoted to such payments in light of the company's performance.

In a change from the proposal, in response to comments, the final rules do not require aggregating the compensation of PEOs in years when a registrant had multiple PEOs. Instead, the final rules require that, in those years, registrants include separate Summary Compensation Table total compensation and executive compensation actually paid columns for each PEO. For example, the below table shows the disclosure that would be required when there were two PEOs in “Year 2”:

We believe including separate disclosure for each PEO, as recommended by some commenters, would address commenters' concerns that aggregating PEO disclosure could lead to confusing or misleading disclosure. In the case of multiple PEOs in a single year, this approach would make the table itself slightly longer, but it would have the added benefit of distinguishing the compensation paid to separate PEOs both visually and in the structured data, instead of presenting a potentially confusing aggregated figure in the table and only having discussion of the separate PEOs in footnote and narrative disclosure.

C. Determination of Executive Compensation Actually Paid

We proposed that “executive compensation actually paid” under Item 402(v) of Regulation S-K would be total compensation as reported in the Summary Compensation Table, modified to adjust the amounts included for pension benefits and equity awards. In both the Proposing and Reopening Releases, we requested comment on the proposed approaches to calculating these amounts, and whether the proposed definition appropriately captures the concept of “executive compensation actually paid,” and in the Proposing Release we offered an economic analysis of an alternative approach to calculating equity awards. We received significant comment, as discussed below, on the proposed approaches to calculating the amounts of pension benefits and equity awards to be included as “actually paid.” In addition, several commenters to the Proposing Release noted that the definition of compensation actually paid as proposed may result in some misalignment between the time period to which pay is attributed and the time period in which the associated performance is reported. After considering the statutory language and the comments received, we are adopting final rules for calculating the amounts reported for pension benefits and equity awards that are modifications of our proposed approach, including, as discussed further below, requiring equity awards to be revalued more frequently than as proposed. We believe that these approaches will more accurately reflect executive compensation actually paid, as required by Section 14(i), and mitigate commenter concerns about timing mismatches by more closely associating compensation with the period of the corresponding performance.

Although Section 14(i) refers to compensation required to be disclosed under Item 402 of Regulation S-K, it also uses the phrase “actually paid,” which differs from disclosure required under Item 402 of “compensation awarded to, earned by or paid to” the NEOs. Because Congress was aware of the language of Item 402 at the time of the Dodd-Frank Act, and adopted text that did not mirror the language of that provision, we believe that Congress intended executive compensation “actually paid” to be an amount distinct from the total compensation as reported under Item 402 because it used a term not otherwise referenced in Item 402. As such, we believe using as a starting point the total compensation that registrants already are required to report in the Summary Compensation Table and making adjustments to some of those figures is appropriate to give effect to the statutory language and reflect executive compensation that is “actually paid.” Commenters generally agreed that adjustments to the Summary Compensation Table total were appropriate to determine “executive compensation actually paid,” noting that there are some items reportable in the Summary Compensation Table total that are not reflective of compensation “actually paid”; or more generally suggesting that the Summary Compensation Table total is not reflective of “executive compensation actually paid.”

1. Deduction of Change in Actuarial Present Value and Addition of Actuarially Determined Service Cost and Prior Service Cost

i. Proposed Amendments

We proposed requiring registrants to deduct the change in actuarial present value of all defined benefit and actuarial pension plans from the Summary Compensation Table total compensation figure, and to add back the actuarially determined service cost for services rendered by the executive during the applicable year, when calculating executive compensation actually paid. We proposed removing the change in actuarial present value of these plans in order to avoid potential volatility associated with revaluing previously accumulated benefits with changes in actuarial inputs and assumptions. However, as discussed in the Proposing Release, we believed that including the service cost from the applicable year was appropriate because it more closely reflected compensation “actually paid” during that year, in that it could be seen as an estimate of the value that would be set aside by the registrant to fund the benefits payable in retirement for the service provided during the applicable year. We also stated that we believed that using the actuarially determined service cost, instead of the Summary Compensation Table pension measure, may increase comparability across registrants of the amounts “actually paid” under both defined benefit and defined contribution plans. For defined contribution plans, the Summary Compensation Table requires disclosure of registrant contributions or other allocations to vested and unvested defined contribution plans for the applicable fiscal year, which will also be included in computing compensation actually paid for purposes of the new disclosure.

In the Reopening Release, we stated that some commenters had noticed challenges with using the pension service cost approach to determining the value of pension benefits “actually paid,” and requested comment on whether there is an alternative measure of the change in pension value attributable to the applicable fiscal year that is better representative of the amount of pension benefits “actually paid.”

ii. Comments

Some commenters generally supported limiting the pension benefits included in executive compensation actually paid to service cost. In addition, some commenters supported the proposed deduction of the change in actuarial present value of defined benefit and pension plans not attributable to the applicable year of service, or generally supported the Commission's choice to exclude the value associated with actuarial assumptions.

There were also a number of commenters who opposed the inclusion of pension service cost in executive compensation actually paid, noting it may remain subject to vesting conditions and may not ever actually be paid; has assumptions built in that would prevent comparability across registrants or distort the figure; is not presently calculated on a per participant basis, so would add cost; or generally that it does not equal compensation “actually paid.” However, a number of commenters who opposed the inclusion of service cost noted their view that it would be a better representation of compensation “actually paid” than the current Summary Compensation Table figure. A few commenters suggested excluding changes in pension values entirely, while some others suggested that the registrant should have the option to exclude service cost, if the executive is not vested in the pension benefits.

A number of commenters suggested other ways to include pension amounts in executive compensation actually paid. Some commenters recommended an approach requiring registrants to calculate the change in pension value to equal the actuarial present value of the benefit earned during the year, noting that it tracks the actual pattern of benefit increases resulting from pay increases and plan amendments, and links directly to the existing approach and assumptions used for the Summary Compensation Table. Another suggested multiplying the value of the pension increase during the year, net of any inflationary increase and contribution by the employee, by twenty.

Some commenters requested clarification regarding the calculation of the service cost amount. Two commenters suggested alternatives to the application of FASB ASC Topic 715, with one suggesting that the Commission instead clarify that the intended measurement is the change in pension values attributable to an additional year of service, and the other suggesting the Commission use the accumulated benefit obligation service cost or the change in present value of accrued benefits, using the same assumptions at the beginning and end of each year. Two commenters suggested the Commission eliminate the reference to the required use of future salary increases to estimate service cost, because it would require significant new data and reveal new information to investors, with one also suggesting the Commission clarify that the intended measurement is the change in pension values attributable to an additional year of service.

Three commenters responded to our request for comment in the Reopening Release asking if there is an alternative measure of the change in pension value attributable to the applicable fiscal year that is better representative of the amount of pension benefits “actually paid.” One suggested that the “value of dollars set aside to provide a pension benefit to an executive” be disclosed. Another suggested that registrants should be required to disclose the “change in (increase) the actuarial present value of pension benefits over the applicable fiscal year using the same economic assumptions as used in the calculation at the start of the applicable fiscal year.” The third stated that pension benefits should be fully excluded from the “actually paid” amount, but also stated that service cost was “far more representative of the compensation received” than the change in actual present value amount included in the Summary Compensation Table total.

iii. Final Amendments

With respect to pension compensation, we are adopting final rules largely as proposed with a modification in response to commenters' suggestion to also include the value of plan amendments in the calculation of compensation actually paid. The final rules will require registrants to deduct from the Summary Compensation Table total the aggregate change in the actuarial present value of all defined benefit and actuarial pension plans, and add back the aggregate of two components: (1) actuarially determined service cost for services rendered by the executive during the applicable year, as proposed (the “service cost”); and (2) the entire cost of benefits granted in a plan amendment (or initiation) during the covered fiscal year that are attributed by the benefit formula to services rendered in periods prior to the plan amendment or initiation (the “prior service cost”), in each case, calculated in accordance with U.S. Generally Accepted Accounting Principles (“U.S. GAAP”).

As noted above, the change in actuarial present value, generally, reflects the difference between the actuarial present value of accumulated benefits at the end of the fiscal year and at the end of the prior fiscal year. The change in actuarial present value would be deducted only if the value is positive, and therefore included in the sum reported in column (h) of the Summary Compensation Table. Where such amount is negative (and therefore not reflected in the Summary Compensation Table and reported only in a footnote to column (h)), no amounts should be deducted for purposes of Item 402(v) of Regulation S-K.

The below table shows the changes from the proposed rules to the final rules with respect to pension compensation (specific changes are bolded and italicized):

We believe that it is appropriate to include pension compensation in the calculation of compensation “actually paid.” The adopted approach in particular provides an appropriate measure for purposes of determining compensation “actually paid” during the applicable year because it reflects the benefits an executive may expect to receive based on additional service the executive provided during the year (or service cost), and it incorporates additional benefits attributable to changes in the pension contract between the executive and the company (or prior service cost). In many cases, this measure will approximate the value that would be set aside currently by the registrant to fund the pension benefits payable upon retirement for the service provided, and any plan amendments made, during the applicable year. In addition, the inclusion of pension compensation is consistent with other compensation disclosure requirements, such as Item 402(c) of Regulation S-K. These same rationales apply whether or not the pension amounts are vested. Consistent with the equity compensation adjustment, the pension adjustment will be included even when unvested until an officer leaves the company.

Another advantage to the approach we are adopting is that it is more closely associated with underlying information from the GAAP financial statements. In particular, the pension's service cost and prior service cost, while not required to be reported separately and for a subset of employees, is computed in the process of calculating the aggregate service cost and prior service cost at the plan level. As a result, a registrant would not be required to collect significant new data or prepare a new calculation of the actuarial present value of the benefit earned during the year, but would rather calculate service cost and prior service cost for a subset of employees for which the underlying information is already available and subject to internal control over financial reporting. The direct relationship of this information to the amounts recognized in the audited financial statements may also provide an additional level of comfort to investors as to its accuracy and reliability. In addition, because this approach excludes changes that derive only from differences in the actuarial assumptions used to estimate the value of benefits already earned in prior periods, it will provide for a more meaningful comparison across registrants of the amounts “actually paid” under both defined benefit and defined contribution plans. Further, as noted above, commenters were generally more supportive of a service cost approach rather than an approach that would include the amount required to be disclosed in the Summary Compensation Table.

One weakness in the proposed approach, identified by commenters, was that the service cost approach would not fully account for changes in the value of an executive's expected benefit arising from plan amendments or initiations. Our modified approach as adopted addresses this concern by requiring that the registrant include, as a component of this item of compensation actually paid, the entire cost of benefits granted in a plan amendment (or initiation) that are attributed by the benefit formula to services rendered in periods prior to the plan amendment or initiation. Such prior service cost information is part of the underlying information required to account for a defined-benefit plan under U.S. GAAP.

For purposes of the final rules, “prior service cost” also refers to any credit arising from a reduction in benefits related to services rendered in prior periods as a result of a negative plan amendment. We acknowledge that including the prior service credit associated with such a negative plan amendment would result in a reduction of compensation actually paid. We believe that such an outcome would be consistent with the statutory objective of capturing compensation actually paid, because the reduction in the accrued benefit reflects a reduction in compensation in the same manner that an increase in the accrued benefit reflects an increase in compensation.

Although one commenter also noted that service cost would exclude the costs related to unexpected compensation changes, we are not adopting a modification in this regard. Under U.S. GAAP, the effects on the projected benefit obligation of unexpected compensation changes ( i.e., changes from the estimated future compensation levels used in measuring service cost) are recorded in actuarial gain or loss. In considering whether to add another component to the tabular pension measure related to actuarial gain or loss due to unexpected compensation changes, we determined that the benefits of isolating these items from other actuarial gains and losses did not merit the costs and complexities associated with calculating the additional adjustment. However, we note that information about compensation changes should still generally be discernible by investors, as such compensation amounts would be included as other components of the compensation disclosed in the Item 402(v) of Regulation S-K table.

We are not persuaded that the other alternative approaches recommended by commenters would more accurately reflect compensation “actually paid.” Although some of the suggested alternatives could more fully account for changes in compensation levels by reflecting unexpected increases in pay as well as plan amendments, we believe that the benefits discussed above with respect to the adopted approach, including its direct relationship to the values already calculated for the purpose of financial statement reporting, outweigh the potential benefits of the alternatives. Further, while we acknowledge there may be an additional cost to obtain the service cost and prior service cost information on a per participant basis, the other calculations suggested by commenters also would include additional costs since registrants are not currently performing those calculations in the manner suggested. In the case of commenters who suggested that we omit all pension cost amounts, we disagree that their suggested approach would be a reasonable interpretation of compensation “actually paid.” Although the approach we are adopting may not always perfectly reflect all potential changes in pension value, the resulting measure is considerably more accurate than a measure that treats the value of promised pension awards as zero when they may ultimately cost the registrant millions of dollars.

We are also requiring that the calculation of “service cost” and “prior service cost” be consistent with the definitions provided under U.S. GAAP. As discussed above, we acknowledge that some commenters suggested alternatives to the U.S. GAAP definition; however, we believe that this definition is appropriate because it reflects the service cost amount included in the financial statements, and therefore is familiar to registrants. The final rules require the entire amount of prior service cost related to a plan amendment to be included in the pension measure rather than the amortized portion of prior service cost recognized as part of periodic pension cost under U.S. GAAP for the year.

2. Inclusion of Above-Market or Preferential Earnings on Deferred Compensation That Is Not Tax Qualified

i. Proposed Amendments

Consistent with Summary Compensation Table disclosure requirements, we proposed that the executive compensation actually paid would include above-market or preferential earnings on deferred compensation that is not tax qualified.

ii. Comments

Two commenters generally agreed with the proposed rules on disclosure of deferred compensation that is not tax qualified. Two other commenters recommended permitting registrants to exclude unvested amounts of deferred compensation that is not tax qualified.

iii. Final Amendments

We are adopting, as proposed, the requirement that executive compensation actually paid include above-market or preferential earnings on deferred compensation that is not tax qualified. We believe, as discussed in the Proposing Release, that excluding those amounts until their eventual payout would make the amount “actually paid” contingent on an NEO's choice to withdraw or take a distribution from their account, rather than the registrant's compensatory decision to pay the above-market return, which we do not believe would be an accurate representation of compensation “actually paid.” As with pension awards, these amounts may be viewed to approximate the value that would be set aside currently by the registrant to satisfy its obligations in the future. In addition, excluding those amounts would be inconsistent with the approach in the Summary Compensation Table, which requires disclosure of the underlying deferred amounts when earned. We believe that, to the extent the Summary Compensation Table approach aligns with the statutory “actually paid” language and purpose of the disclosure, we should minimize adjustments to the Summary Compensation Table figures, in order to make disclosures easier to understand for investors and easier to produce for registrants. To that end, we are also not permitting registrants to voluntarily exclude unvested amounts of deferred compensation that is not tax qualified, as we believe that could complicate investors' understanding of the disclosure, and would limit the comparability of the “actually paid” amounts across different registrants.

3. Equity Awards

i. Proposed Amendments

We proposed that equity awards be considered “actually paid” on the date of vesting, and valued at fair value on that date, rather than fair value on the date of grant as required in the Summary Compensation Table. In proposing this approach, we noted that an executive does not have an unconditional right to an equity award before vesting, and therefore unvested options or other equity awards may not be “actually paid” prior to the vesting conditions being satisfied, which can be viewed as representing payment by the registrant. In addition, we noted that using the vesting date fair value would incorporate changes in the value of the equity awards from the grant date to the vesting date, with that change being one of the key ways that pay is linked to registrant performance.

With respect to the calculation of the vesting date fair value, we noted that the vesting date fair value of stock awards is already disclosed (by registrants other than SRCs) in the Option Exercises and Stock Vested Table, and that the vesting date fair value of option awards can be calculated using existing models and methodologies. Specifically, the proposed approach would require (i) the amounts reported pursuant to 17 CFR 229.402(c)(2)(v) and (vi) to be deducted from Summary Compensation Table total, and (ii) the vesting date fair value of stock awards and options (with or without stock appreciation rights), each computed in accordance with the fair value guidance under U.S. GAAP, to be added. As proposed, a registrant would be required to disclose vesting date valuation assumptions if they are materially different from those disclosed in its financial statements as of the grant date.

In response to comments received on the Proposing Release (discussed below), we included a request for comment in the Reopening Release, noting commenters' concerns that there was a potential misalignment between the time period to which pay is attributed and the time period in which the associated performance is reported, and asking if there were other approaches that would alleviate this misalignment, or if the inclusion of the additional measures considered in the Reopening Release would affect this misalignment.

ii. Comments

We received a number of comments on both the proposal to use fair value methodology to value equity awards in the calculation of executive compensation actually paid, and on the proposal to value such awards as of the vesting date.

Some commenters supported the proposed fair value methodology. However, a number of commenters opposed the approach, noting that the calculation of fair value is time consuming and expensive, particularly when many separate fair value calculations would be required, as in the case of awards that are on a pro-rata vesting schedule or with multiple tranches in a given year; few companies have familiarity with valuing options that have been outstanding for several years; the assumptions that are included in fair value calculations are company-specific and therefore would reduce comparability; and that the fact that assumptions and projections are included in fair value calculations is inconsistent with the concept of “actually paid.” As an alternative to fair value, a number of commenters suggested the Commission require options to be valued at their intrinsic value, or permit registrants to choose between disclosure of fair value and intrinsic value (with the non-chosen value being provided in footnote disclosure). These commenters argued that intrinsic value is easier and cheaper to calculate; aligns with the value that the executives would receive upon immediate exercise; and does not include the valuation assumptions that accompany the fair value methodology. Some commenters suggested that if the final rules did not use intrinsic value, they should instead use fair value with certain safe harbors or simplified assumptions that would reduce the effort required to compute the valuation.

Some commenters supported valuing equity at the vesting date, stating that valuing equity at the vesting date will incorporate the grant date fair value and changes until vesting (which “represent a direct channel, and one of the primary means, through which pay is linked to registrant performance”), but will not include post-vesting changes (which “generally reflect investment decisions made by the executive rather than compensation decisions made by the registrant”); will avoid “underestimating the actual compensation received by executives,” which could occur if grant date reporting was required; and “better reflect[s] the value ultimately delivered to executives.” Some commenters specifically opposed exercise date valuation, while others supported requiring the vesting date valuation of stock awards, but the exercise date valuation of options or requiring the vesting date valuation of performance-based awards, but the grant date valuation of time-based awards. Some commenters opposed vesting date valuation, with one arguing that valuing options at vesting date would be misleading because executives do not generally include the option value in their income at the time of vesting. As alternatives, commenters suggested: valuing awards at the end of a multi-year period, such as a three-year period; valuing equity at grant date but reversing the value at the vesting date for awards that fail to vest; revaluing outstanding equity awards annually; or revaluing all equity granted during a period at the end of the most recent completed fiscal year.

A number of commenters opposed the reporting of equity as of the vesting date. Some of these commenters noted that vesting date reporting of equity would lead to a timing misalignment between actual performance and executive compensation actually paid, as the performance that “earned” the equity would have occurred between the grant date and the vesting date, but only the total amounts of equity would be reported on the vesting date. However, two commenters, who acknowledged the misalignment, indicated that there was no other approach that would eliminate all misalignment.

Several commenters requested clarifications about the proposed approach. A few commenters expressed that reporting equity on the vesting date creates uncertainty in application, and either sought clarification regarding the vesting date or the meaning of when “all applicable vesting conditions were satisfied.” One commenter suggested that an award should be considered vested on the date the executive is able to monetize the award, while another suggested that awards should only be considered “actually paid” when restrictions on equity lapse, even if already vested. Two commenters also made suggestions that awards should be considered vested when the associated performance period is completed, even if the vesting of the award is still subject to board certification.

Commenters suggested a number of alternatives to vesting date reporting of equity, including: grant date reporting; exercise date reporting; exercise date reporting of the equity's intrinsic value; principles-based reporting ( i.e., allowing companies to make their own modifications to the reporting date); reporting “in the fiscal year for which the compensation was considered as paid”; and annual reporting, starting in the grant year, of the year-end fair value of the award, with annual reporting of any change in the fair value until, and including, the year of vesting. Two commenters also suggested the Commission adopt the “2 1/2 month rule,” under which equity vesting in the first two and one half months of the calendar year would be attributed to the prior year. One commenter stated that, because the proposed rules would move away from grant date fair value calculations for equity awards, it would be important that the disclosure include dividends paid on unvested equity or equivalents for a given year.

A few commenters supported the proposed requirement that changes in the underlying assumptions for valuation that are materially different from those made in the financial statements as of the grant date must be disclosed, with one specifically supporting the proposed requirement, one supporting requiring any changes from the assumptions in the current financial statements to be disclosed, and two opposing the disclosure of changes in valuation assumptions.

In response to a request for comment in the Reopening Release, one commenter indicated that the additional performance measures considered in the Reopening Release would not exacerbate the timing misalignment, while another stated the additional measures would not improve the misalignment.

iii. Final Amendments

After consideration of the comments received, we are modifying our approach to the treatment of equity awards in relation to the total compensation reported in the Summary Compensation Table. While the final amendments continue to use “fair value” as the measure of the amount of an equity award, which is consistent with accounting in the financial statements, we are adjusting the date on which the award is valued in response to comments, so that the first fair value disclosure is made in the year of grant, and changes in value of the award are reported from year to year until the award is vested. We believe this approach will better align the timing of the disclosure and valuation with when the award is actually “earned” by the executive, resulting in disclosure that more clearly shows the relationship between executive compensation and the registrant's performance.

In particular, the proposed rules would have required the deduction of the equity award amounts reported in the Summary Compensation Table total and the addition of:

• The vesting date fair value of stock awards and options (with or without stock appreciation rights), each computed in accordance with the fair value guidance under U.S. GAAP.

The final rules also require the deduction of the equity award amounts reported in the Summary Compensation Table total; however, instead of the addition of the vesting date fair value of stock awards and options, the final rules require the addition (or subtraction, as applicable) of the following:

• The year-end fair value of any equity awards granted in the covered fiscal year that are outstanding and unvested as of the end of the covered fiscal year;
• The amount of change as of the end of the covered fiscal year (from the end of the prior fiscal year) in fair value of any awards granted in prior years that are outstanding and unvested as of the end of the covered fiscal year;

• For awards that are granted and vest in the same covered fiscal year, the fair value as of the vesting date;

• For awards granted in prior years that vest in the covered fiscal year, the amount equal to the change as of the vesting date (from the end of the prior fiscal year) in fair value;

• For awards granted in prior years that are determined to fail to meet the applicable vesting conditions during the covered fiscal year, a deduction for the amount equal to the fair value at the end of the prior fiscal year; and

• The dollar value of any dividends or other earnings paid on stock or option awards in the covered fiscal year prior to the vesting date that are not otherwise reflected in the fair value of such award or included in any other component of total compensation for the covered fiscal year.

We believe fair value is an appropriate measure for compensation “actually paid.” Although fair value calculations, like all accounting estimates, do involve some subjective assumptions, we do not agree with commenters that stated that the assumptions and projections included in fair value calculations render such amounts inconsistent with the concept of “actually paid.” Fair value is an estimate of the amount by which an executive is compensated as a result of an award, and therefore represents a reasonable measure of that executive's “actual pa[y].” Specifically, the fair value of an option is a widely-used measure to estimate the total value of the asset, including both its value if exercised immediately (“intrinsic value”) and the additional value created by the holder's contractual right to exercise at some time in the future (“time value” of the option). In our view it also represents a more accurate measure of actual pay than alternatives recommended by some commenters.

We are not adopting the approach suggested by some commenters that we use other measures such as intrinsic value. Intrinsic value would ignore the option value inherent in exercisable awards prior to exercise, including the option value inherent in an option award that is at-the-money or out-of-the-money ( i.e., the stock price is equal to or less than the strike price of the options), and therefore has zero intrinsic value. Intrinsic value (or any similar measure used to calculate compensation “actually paid”) would also be a departure from the primary disclosures related to equity compensation, and the recognition and measurement of such compensation in the financial statements under U.S. GAAP, and we believe would not allow investors to as easily link and analyze “compensation actually paid” with the other information they are receiving about executive compensation. Further, in 2004, the accounting for stock-based compensation in U.S. GAAP was revised to require fair value accounting. In the revised accounting standard, it was noted that other equity instruments and the consideration the issuing entity receives in exchange for them are recognized in the financial statements based on the fair value of the instrument at the date issued. The fact that the equity instruments would be issued for goods or services rendered or to be performed did not seem to be a reason to measure the cost of the goods or services performed on a different basis. The standard further noted that most advocates of intrinsic value favored its use only at a grant date measurement, and noted that there are weaknesses in its use even in that case, such as treating most fixed share options as though they were a “free good.” However, even at the grant date, employee services received in exchange for share options are not free and there is value in the employee services performed and the related stock and stock options received.

Registrants and investors are already familiar with fair value calculations and the determination of the assumptions for such calculations through their use in existing Commission disclosure requirements as well as U.S. GAAP. For example, the Grants of Plan-Based Awards Table requires grant date fair value disclosure of each individual equity award granted during the last completed fiscal year. U.S. GAAP requires information about grant date fair value for equity awards, including the weighted-average grant-date fair value of awards that were granted, vested and forfeited during the year and a description of the significant assumptions used during the year to determine the fair value of share-based compensation awards.

We do not agree with the suggestion from commenters that we consider an option or other award requiring exercise to be “actually paid” only upon its exercise, as we believe doing so would commingle the registrant's compensatory decision with the executive's investment decision about when to exercise and would allow executives to influence pay-versus-performance disclosure by controlling the fiscal year in which they receive the compensation. We additionally determined that year-over-year change in fair value better meets the statutory purposes than grant-date fair value, because valuing awards only at grant date fails to reflect increases in value to the executive after the grant date, during the period over which the compensation actually paid is earned. Even if year-over-year change in fair value is only a reasonable estimate, we believe it is far more accurate to include this estimate than to omit such increases in value entirely.

We have changed the reporting and valuation date requirements from the Proposing Release to first require the year-end reporting and valuation of awards granted during the fiscal year and then the year-over-year change in fair value of such awards until the vesting date (or the date the registrant determines the award will not vest).

We have made these changes to the reporting and valuation requirements to address commenters' concerns about potential misalignment between the time period to which pay is attributed and the time period in which the associated performance is reported, and the degree to which this would affect the usefulness of the disclosure. We believe that, compared to the vesting date valuation approach included in the Proposing Release, the adopted approach will more effectively allow registrants to describe the relationship between compensation and registrant performance, as the reported amounts of compensation will annually adjust based on the registrant's performance, among other things, in that year. In addition, we acknowledge commenters' observation that comparability may be somewhat reduced by the assumptions that are included in fair value calculations, which, as noted by a commenter, may differ from issuer to issuer. Because investors are already familiar with fair value as the measurement approach for equity awards under U.S. GAAP, they are aware of the reduced comparability that may occur due to the use of different assumptions from issuer to issuer. However, we believe that the use of a consistent measurement approach to equity compensation in the Summary Compensation Table, the financial statements, and the calculation of compensation “actually paid,” along with the required disclosures about significant assumptions under U.S. GAAP in the final rules, allows for comparability with respect to an individual issuer's disclosures from year to year. Further, as discussed in the Proposing Release, we believe that, overall, comparability regarding the awards included by registrants in the disclosure will be greater under the adopted approach than it would have been under the proposed approach, as volatility in executive compensation actually paid across the disclosure periods that is due simply to vesting patterns should decrease (as the amount of executive compensation actually paid will be adjusted each year as it is “earned” over the course of the vesting period).

Investors will also be able to more easily understand the impact of performance on awards-based compensation over time, because under the final rules as adopted investors will be able to observe the amount by which the value of an executive's compensation changes each year, rather than only observing the value of that compensation in the year an award vests. Furthermore, we believe that the adopted approach in the final rules is similar to the concept of realizable pay, recommended by some commenters, as it reflects an attempt to measure the change in value of an executive's pay package after the grant date, as performance outcomes are experienced.

This approach to unvested equity compensation is consistent with the treatment of other unvested elements of compensation under the final rules, such as unvested pension benefits and contributions to unvested defined contribution plans. In each case, the adopted approach reflects this compensation as it is earned rather than at vesting. We believe the consistent use of this approach should reduce misalignment between the timing of when compensation is earned and when it is reported, and allow the disclosure to more clearly represent the relationship of pay with performance over time.

We also believe this revised approach for equity awards comports with the statutory term “executive compensation actually paid.” While non-vested amounts of compensation could be considered unpaid due to their contingent nature, over time the values reported in connection with a particular award will aggregate to its ultimate value upon vesting. Aligning the compensation reporting more closely with when the compensation changes in value also provides investors with a clearer picture of “the relationship between executive compensation actually paid and the financial performance of the issuer.” For example, where an award vests over a three-year period and the registrant's financial performance is positive in the first of those two years and negative in the third, reporting the full value of the award only in the vesting year may give investors the misleading impression that the executive was not rewarded for positive performance in years one and two and was rewarded despite negative performance in year three. In addition, the required reporting of the year-over-year change in fair value of such awards until the vesting date (or a deduction for prior reported amounts as of the date the registrant determines the award will not vest) will account for any amounts that fail to vest; will address concerns, noted by commenters, that grant date reporting undervalues compensation “actually paid”; and will not include those post-vesting changes that generally reflect the executives' investment decisions, not compensation.

We recognize that requiring fair value calculations for each equity award at a date other than the grant date may be burdensome for some issuers, as noted by some commenters, particularly those that have compensation programs with numerous and complex equity grants. However, in the final rules we are not adopting a safe harbor or simplified assumptions other than those generally accepted under U.S. GAAP, as suggested by some commenters. Since accounting for share-based compensation in U.S. GAAP was revised in 2004 to require fair value accounting, registrants have been accounting for equity compensation based on a fair value approach and must determine valuation assumptions every time a new award is granted. While commenters correctly noted that companies are not as familiar with the fair valuation of options after the grant date, U.S. GAAP requires the re-valuation of an award when modified, so the concept of valuing a stock award before vesting is also not novel to registrants. As such, registrants are required to have internal controls and processes over the valuation of stock awards, including the assumptions used in determining fair value. We believe that registrants will likely rely upon the existing fair value processes and internal controls for stock-based compensation, which should mitigate the concerns raised by commenters about assumptions. In addition, the option and contingent-equity valuation models are well-developed and related software solutions are widely available, which will further mitigate those additional burdens and concerns related to valuation approach and related inputs.

The final rules also require footnote disclosure of any valuation assumptions that materially differ from those disclosed at the time of grant, as in the proposal. The proposal did not specify how to disclose the valuation assumptions. Similar to U.S. GAAP, when multiple awards are being valued in a given year, a registrant may disclose a range of the assumptions used or a weighted-average amount for each assumption. In addition, the fact that certain institutional investors and third parties (often proxy advisors or compensation consultants) are already incorporating similar computations in their own pay for performance analyses, suggests that the adopted approach is already considered useful and operational by some investors.

Further, we are also requiring the dollar value of any dividends or other earnings paid on stock or option awards in the covered fiscal year prior to the vesting date to be included in the amount of executive compensation actually paid, if such amounts are not reflected in the fair value of such award or included in any other component of total compensation for the covered fiscal year. As noted by a commenter, the pay-for-performance disclosure should include dividends paid on unvested equity or equivalents “as a result of the move away from grant date fair value calculations for equity awards.” Under the Summary Compensation Table total, any such amounts would be typically included in the grant date fair value, as no such dividends or earnings would have been paid on that date. However, if any dividends or other earnings are paid on stock or option awards over time, these amounts would decrease future fair value amounts. This decrease would not be reflective of a decrease in the amount “actually paid” to the executive, to the contrary, the amount of the decrease would reflect actual dividends or earnings paid to the executive prior to the valuation. We believe these amounts are compensation “actually paid” and should be reflected in the disclosure.

D. Measures of Performance

1. Requirement To Disclose TSR and Peer Group TSR

i. Proposed Amendments

We proposed requiring all registrants subject to the proposed rule to use TSR as the measure of financial performance of the registrant for purposes of the required disclosure. In addition, we proposed requiring registrants that are not SRCs to disclose peer group TSR, using either the same peer group used for purposes of Item 201(e) of Regulation S-K or a peer group used in the CD&A for purposes of disclosing registrants' compensation benchmarking practices.

ii. Comments

Commenters were divided on the use of TSR as a required financial performance measure, with some commenters generally supportive, and some generally opposed. Additionally, some commenters opposed TSR being used as the sole measure of financial performance.

Commenters in favor of including TSR as a required financial performance measure noted that TSR is well-understood by investors; is widely used by companies in setting compensation; is generally a fair representation of company performance; will assist companies “in articulating and providing justification for their compensation practices”; will increase comparability; and reflects stock price fluctuations that regularly occur in response to publicly known information and company leadership. Commenters in favor of TSR also observed that requiring its disclosure is consistent with the language in Section 953(a) that the pay-versus-performance disclosure should “tak[e] into account any change in the value of the shares of stock and dividends of the issuer and any distributions.”

Commenters opposed to the use of TSR, generally or as the sole measure of performance, as well as a few commenters in favor of the use of TSR, noted that TSR has specific limitations, including: not necessarily being used by the subject company to determine compensation; being an unreliable performance measure for thinly-traded stocks; incentivizing short-term performance at the expense of investors' long-term best interests (which some commenters indicated could incentivize companies to incorporate strategies to inflate stock prices over the short term, or to engage in buybacks ); requiring lengthy explanatory disclosures to explain any misalignments between compensation and TSR; causing companies to adjust their compensation programs to more heavily rely on TSR; being subject to fluctuations based on circumstances outside of the control of companies, industries, and executives; and being affected by the granting and vesting of stock options. In response to these concerns, some commenters (including commenters in favor of using TSR ), suggested permitting disclosure of other metrics alongside TSR. Other commenters generally stated that there was no single performance measure that would align with the compensation plan of every registrant, and therefore suggested adopting a principles-based approach, allowing companies to choose their own performance measures. Alternatively, a number of commenters suggested requiring registrants to disclose the actual metrics used in determining their executive compensation, or revising Item 402 of Regulation S-K to require disclosure of “all” metrics actually used to determine NEO incentive compensation.

A number of commenters raised questions or made comments regarding the calculation of TSR. A few commenters suggested that TSR should be presented as a percentage change instead of an indexed dollar value. Others generally raised questions about the method used for calculating TSR, with some suggesting TSR should be calculated and disclosed as a one-year measure, others suggesting that TSR should be calculated as a rolling average, and a third group suggesting TSR be calculated as a cumulative average over the time period of the disclosure. Other commenters suggested that we permit registrants to decide the time period used to calculate their TSR.

Commenters were also divided on our proposal to require registrants, other than SRCs, to disclose peer group TSR. Some commenters supported requiring the inclusion of peer group TSR, while others suggested peer group disclosure should be optional. A number of other commenters opposed the requirement to disclose peer group TSR, arguing peer group disclosure: is already disclosed in the performance graph required by Item 201(e) of Regulation S-K; is beyond the mandate of the Dodd-Frank Act; will confuse or mislead investors; will be expensive and/or time-consuming for registrants to calculate; is difficult for registrants to explain and would require lengthy disclosures; is difficult to understand given that frequent changes in peer groups and different market conditions or performance cycles affect different “peer” companies differently; and creates issues relating to the difficulty for companies to find adequate peers, limiting the ability to make direct comparisons between registrants. A number of commenters also opposed requiring weighted peer group TSR (weighted by market capitalization), as used in Item 201(e) of Regulation S-K. In addition, one commenter suggested we permit multiple peer groups to be disclosed, if peer group TSR disclosure is required.

Commenters generally supported allowing registrants to have flexibility in setting their peer groups for the pay-versus-performance disclosure. Commenters had various suggestions as to how to achieve this flexibility, including allowing registrants to choose any peer group referenced in the CD&A; allowing the use of the peer group from either Item 201(e) of Regulation S-K or the CD&A; or allowing registrants to choose a peer group other than the Item 201(e) of Regulation S-K or CD&A peer groups. These commenters generally supported requiring registrants to provide disclosure explaining the make-up of their peer group. One commenter, however, opposed giving flexibility to registrants in setting their peer groups, and instead suggested requiring that the peer group should be the same as the peer group used in benchmarking executive compensation.

Commenters raised questions about the impact of a registrant changing its peer group. Some commenters advocated for requiring additional disclosure in the event that a registrant changes its peer group, including requiring the disclosure of comparative results of TSR for all peer groups used in the disclosed time period. Others questioned what impact the change of a peer group would have on cumulative TSR, with some commenters suggesting we only require disclosure of the current peer group. One commenter suggested that, if annual TSR is used, the peer group in place in the respective year of disclosure should be the peer group used to calculate the peer group TSR for that year of disclosure.

iii. Final Amendments

We are adopting the requirement, as proposed, that all registrants subject to the final rules use TSR, and that registrants (other than SRCs) use peer group TSR, as measures of performance. As noted in the Proposing Release, Section 14(i) does not mandate we require specific measures in the pay-versus-performance disclosure. However, the statute does provide that the disclosures should “tak[e] into account any change in the value of the shares of stock and dividends of the issuer and any distributions.” While we recognize commenters' concerns that TSR is not an equally useful measure for all registrants (as it is not necessarily used by all registrants to set compensation and is seen by some commenters to be an unreliable performance measure for thinly-traded stocks), is subject to fluctuations based on circumstances outside of the control of the registrant, and may be affected by the granting and vesting of stock options, we believe that TSR is consistent with that statutory language. In addition, we believe mandating a consistently calculated measure for all registrants will further the comparability of the pay-versus-performance disclosures across registrants, as noted by some commenters. We acknowledge, as noted by some commenters, that some registrants may need to provide somewhat lengthy explanatory disclosures to explain any misalignments between compensation and TSR; however, we believe those disclosures are the types of disclosures intended by the language of Section 14(i), and will help investors understand the relationship between executive compensation actually paid and the registrant's performance.

We are not requiring registrants to disclose all measures they use to set executive compensation, as recommended by some commenters, because we believe such a requirement would be a significant change from the current executive compensation disclosure requirements, and would be more appropriately considered by the Commission in a broader context not related to the Section 953(a) mandate. In addition, as noted below, as with other mandated disclosures, registrants would be permitted to disclose additional measures of performance, so long as any additional disclosure is clearly identified as supplemental, not misleading and not presented with greater prominence than the required disclosure. While this does not provide registrants with the full flexibility of a principles based approach suggested by some commenters, we believe this ability to supplement the required disclosures will provide registrants with adequate discretion to provide sufficiently fulsome disclosure of the relationship between their performance and the compensation actually paid to their executives.

We also believe that absolute company performance alone, as reflected in TSR, may not be a sufficient basis for comparison between companies, and that peer group TSR will provide investors with more comprehensive information for assessing whether the registrant's performance was driven by factors common to its peers or instead by the registrant's own strategy and other choices. The final rules require a registrant to disclose weighted peer group TSR (weighted according to the respective issuers' stock market capitalization at the beginning of each period for which a return is indicated), using either the same peer group used for purposes of Item 201(e) of Regulation S-K or a peer group used in the CD&A for purposes of disclosing registrants' compensation benchmarking practices. If the peer group is not a published industry or line-of-business index, the identity of the issuers composing the group must be disclosed in a footnote. A registrant that has previously disclosed the composition of issuers in its peer group in prior filings with the Commission would be permitted to comply with this requirement by incorporation by reference to those filings. We believe this would avoid the potential for duplicative disclosure. Consistent with the approach taken in Item 201(e) of Regulation S-K, as proposed, if a registrant changes the peer group used in its pay-versus-performance disclosure from the one used in the previous fiscal year, it will only be required to include tabular disclosure of peer group TSR for that new peer group (for all years in the table), but must explain, in a footnote, the reason for the change, and compare the registrant's TSR to that of both the old and the new group. Some commenters advocated for more disclosure when a peer group changes (including requiring the disclosure of comparative results of TSR for all peer groups used in the disclosed time period), while other commenters suggested we only require disclosure of the current peer group. We believe the adopted approach strikes the appropriate balance of providing investors information when a peer group changes, while also not requiring overcomplicated disclosure. In addition, as proposed, we are requiring weighted peer group TSR, as calculated under Item 201(e) of Regulation S-K, as registrants are already familiar with this calculation.

In response to commenters' questions about the calculation of TSR, we are clarifying the definition of “measurement period” in the final text of the rule. TSR will continue to be calculated on the same cumulative basis as is used in Item 201(e) of Regulation S-K, measured from the market close on the last trading day before the registrant's earliest fiscal year in the table through and including the end of the fiscal year for which TSR is being calculated ( i.e., the TSR for the first year in the table will represent the TSR over that first year, the TSR for the second year will represent the cumulative TSR over the first and the second years, etc. ). We are also clarifying that both TSR and peer group TSR should be calculated based on a fixed investment of one hundred dollars at the measurement point. As noted by a commenter, the TSR presented in the stock performance graph includes a starting investment amount on the y-axis, from which the subsequent TSR amounts are calculated. As the final rules mandate a tabular not graphical disclosure of TSR, we are clarifying that the TSR amounts should be calculated based on an initial fixed investment of one hundred dollars, to clarify for investors what amount is used to calculate the TSR figures, and to standardize the disclosure across registrants. We are not requiring, as suggested by some commenters, that TSR be calculated as a percentage change instead of a dollar value; be disclosed as a one-year measure; be calculated as a rolling average; or be calculated based on a time period chosen by the registrant as we believe all of those approaches would depart from the existing approach used in Item 201(e) of Regulation S-K, and therefore could be burdensome to registrants and confusing to investors. Similarly, we believe that permitting registrants to choose their own criteria for calculating their TSR and peer group TSR for the pay-versus-performance disclosure could also lead to investor confusion.

We disagree with commenters who raised concerns that peer group TSR would be confusing to investors, expensive to calculate, and hard to understand. Peer group TSR is already included in other disclosures, meaning both investors and registrants are generally familiar with it. While peer group TSR is not specifically included in Section 14(i), we believe it is a useful measure for evaluating a registrant's performance, as noted by other commenters, and we are therefore using our discretionary authority to require this additional information to enhance the Dodd-Frank Act mandated disclosures. As we described above, peer group comparisons are often used by registrants' compensation committees, and may help in determining whether a registrant's performance was driven by factors common to its peers, which may have been outside of the control of its executives.

As discussed below, to address commenters' concerns with respect to the proposal to use TSR and peer group TSR as the sole measures of performance (such as causing companies to adjust their compensation programs to more heavily rely on TSR), we are also requiring registrants to include net income and a Company-Selected Measure as performance measures in the tabular disclosure, and also permitting companies to voluntarily include additional measures of their choosing in the table, as suggested by some commenters. The inclusion of the Company-Selected Measure and the ability of registrants to voluntarily include additional measures may also address commenters' concerns with respect to incentivizing short-term performance at the expense of shareholders' long-term best interests. We believe these additional measures should help alleviate concerns expressed by some commenters that disclosing only TSR (for a registrant and its peer group) would put too much emphasis on that one measure.

2. Requirement To Disclose Net Income

i. Amendments Considered in the Reopening Release

In the Reopening Release, we requested comment on requiring registrants to disclose both income or loss before income tax expense and net income in their pay-versus-performance disclosure. We stated we were considering these two measures because in reflecting a registrant's overall profits (net of costs and expenses), they could be additional important measures of company financial performance that may be relevant to investors in evaluating executive compensation, and could complement the market-based performance measures required in the Proposing Release (TSR and peer group TSR) by also providing accounting-based measures of financial performance. In addition, both net income and income or loss before income tax expense are measures that are familiar to registrants and investors, as both are generally required to be presented on the face of the Statement of Comprehensive Income by Regulation S-X. Net income is also a line item required by U.S. GAAP and International Financial Reporting Standards (“IFRS”) as issued by the International Accounting Standards Board.

ii. Comments

Commenters were divided over the potential inclusion of income or loss before income tax expense and net income. A number of commenters generally supported the inclusion of the measures as additional measures in the table; noting that they will be useful to investors in assessing executive compensation; will cause minimal compliance challenges, as they are already calculated by registrants; and will increase comparability. However, other commenters opposed requiring registrants to disclose the measures, noting they are not relevant for or comparable across all companies (particularly early stage companies and real estate investment trusts (“REITs”) ); are not used by many companies in setting executive compensation; would be incomplete or misleading without appropriate context; and can vary period over period due to one-time adjustments and events such that the relationship with pay would be distorted. Other commenters opposed the measures more generally, as non-company-specific measures, indicating their inclusion would “substantially lengthen” the pay-versus-performance disclosure, without providing specific insight into the registrant, would not address the shortcomings of TSR because they have similar weaknesses (such as encouraging short-termism or “overemphasiz[ing] financial performance”), or would stifle innovation by encouraging more uniform compensation structures given the standardized disclosure across registrants.

iii. Final Amendments

We are adopting the final rules to require registrants to include net income in their tabular disclosure. As discussed above, registrants would also be required to provide a clear description of the relationship of net income to executive compensation actually paid, in narrative or graphical form, or a combination of the two.

Although, as noted by some commenters, net income itself may not be frequently used by registrants directly in setting compensation, we believe that net income is closely related to other profitability measures that we believe, based on Commission staff experience, may be used by registrants in setting compensation, while also being widely understood and standardized, as a required disclosure item under Regulation S-X, U.S. GAAP, and IFRS. The inclusion of net income as an additional financial performance measure could complement the market-based performance measure of TSR, and, to the extent that TSR does not (in the view of management) fully reflect a company's performance, could help to provide investors more ready access to an additional key measure of the company's recent financial performance. As noted in the Reopening Release, to the extent that net income would otherwise be considered by investors when evaluating the alignment of pay with performance, its inclusion in the table may lower the burden of analysis for those investors.

We also believe that the standardized disclosure of net income could assist investors in generally understanding and analyzing the relationship between pay and performance. While, as noted by some commenters, net income may not be relevant for all registrants at all times, including it may allow investors to have a standard baseline from which to analyze a registrant's pay-versus-performance disclosure. Moreover, by requiring a Company-Selected Measure and giving registrants the ability to disclose additional registrant-specific measures, we believe registrants can avoid concerns raised by commenters that financial performance would be overemphasized or disclosure overly standardized by the required disclosure of net income.

The final rules do not require disclosure of income or loss before income tax expense, as considered in the Reopening Release. Net income and income or loss before income tax expense are highly correlated, so we believe requiring both could lead to unnecessarily duplicative disclosure, which could have raised questions for investors trying to understand what, if any, meaningful differences there were between the measures. This potentially duplicative disclosure also would have required registrants to prepare additional relationship disclosure (about the relationship between income or loss before income tax and executive compensation actually paid), which would have created an additional burden on registrants, and may have been less clear for investors. By requiring only one of the two net income measures, we also partially address the concern that adding both net income and income or loss before income tax expense could “substantially lengthen” the pay-versus-performance disclosure. In addition, we believe net income may, based on statistics provided by a commenter, be used by significantly more companies in linking pay to performance than income or loss before income tax expense.

3. Tabular List of the Registrant's “Most Important” Performance Measures

i. Amendments Considered in the Reopening Release

In the Reopening Release, we requested comment on requiring registrants to provide a ranked tabular list of the five most important measures that they use to link executive compensation actually paid during the fiscal year to company performance, over the time horizon of the disclosure. We requested comment on the inclusion of such a ranked list, in part, in response to commenters who stated that the proposal should be revised to require disclosure of the quantitative metrics or key performance targets companies actually use to set executive pay. We noted that this disclosure, if required, would be supplemental to the existing CD&A disclosure, which requires registrants to disclose “all material elements of the compensation paid,” including, for example, which “specific items of corporate performance are taken into account in setting compensation policies and making compensation decisions,” but does not specifically mandate disclosure of the performance measures that determined the level of recent NEO compensation actually paid. We noted that, under the considered approach, registrants would be able to cross-reference to existing disclosures elsewhere in the applicable disclosure document that describe the various processes and calculations that go into determining NEO compensation as it relates to these performance measures, if they elected to do so.

ii. Comments

A number of commenters supported the inclusion of a ranked list. Some of the commenters who supported the ranked list also suggested additional disclosures to supplement the list itself, including requiring “clear description of the relationship between the measures and executive compensation,” the metrics and methodology used to calculate the measures, and the “percentage of total compensation paid at the vesting date” with respect to each of the measures included in the list. In addition, some commenters supported requiring or permitting environmental, social and governance (“ESG”) metrics to be included in the ranked list. One commenter also specifically supported using a tabular format for the list, stating that it would help make company-to-company comparisons.

A number of other commenters opposed the ranked list, with some indicating that its ranking requirement would be difficult to satisfy, as registrants do not rank their measures in the compensation setting process and measures can interact in determining pay in complex ways. Some commenters objected that the list oversimplifies the compensation setting process, particularly because there could be difficulty ranking multiple measures, which might be related or hold equal importance at any given time. Others indicated the list and associated clarifications and explanations would increase the length and complexity of disclosure and associated burdens with little or no corresponding benefit. In contrast, one commenter indicated that it was not aware of any additional costs to disclose the five most important performance measures, and that the disclosure of sensitive or competitive information should not be necessary to provide the list. One commenter suggested that the ranked list was beyond the scope of the Dodd-Frank Act mandate, and others noted that similar disclosure is already available in the CD&A.

There were also a number of commenters who commented on the “most important” concept, which we considered applying both to the ranked list and the Company-Selected Measure (discussed below). Two commenters suggested that defining the “most important” measures would be burdensome for companies, particularly given that many companies overlap and interrelate the measures they use to set compensation. One commenter, who opposed the requirement to include a Company-Selected Measure, stated that, if a “most important” concept is included in the final rules, the Commission should not define “most important” on behalf of registrants. However, another commenter suggested that the Commission make explicit that the “most important” measures are those that drove the outcome of compensation payments, not those that were the most important in compensation decision-making. Some commenters suggested that the Commission clarify whether certain market-linked measures could be considered the “most important” measures, with one suggesting that companies should be able to select the measure they believe to be most important “regardless of whether that measure is one that it uses in a performance or market condition in the context of an incentive plan.” One commenter suggested that the “most important” concept would be improved, “if the definition includes the registrant's assessment that the measure will assist investors in better understanding how the registrant's pay programs contribute to the company's long-term shareholder return,” while another suggested that the standard to evaluate “most important” should be “most useful” for the company.

A number of commenters supported allowing the companies' “most important” measures to be non-financial measures.

Two commenters specifically commented on the time period over which the “most important” measures should be measured: one supported using the measure that was the “most important” over the time horizon of the disclosure, while the other suggested that the “most important” evaluation should be made annually.

A few commenters were concerned that requiring companies to disclose a specific “most important” measure may lead companies to provide disclosure that highlights the measure that makes the company look the best.

iii. Final Amendments

The final rules require registrants provide a list of their most important financial performance measures used by the registrant to link executive compensation actually paid during the fiscal year to company performance (“Tabular List”), and permit registrants to include non-financial performance measures in the Tabular List if such measures are among their most important performance measures. However, in response to comments received on the Reopening Release, certain of the requirements for this list differ from the approach discussed in the Reopening Release.

First, in response to comments, we are not requiring the Tabular List to be ranked. As noted by a number of commenters, numerically ranking measures may be difficult for companies, given the frequent interplay between different measures within a company's compensation program. We believe an unranked list will provide investors with insights into companies' executive compensation programs by still presenting them with the “most important” measures, while avoiding potentially burdensome calculations and analysis that could be involved in specifically designating a first, second, third, etc. “most important” measure. We are not requiring registrants to provide the methodology used to calculate the measures included in the Tabular List. We believe such a requirement would be burdensome on registrants, particularly when the measures are already well understood by investors or otherwise disclosed. However, registrants should consider if such disclosure would be helpful to investors to understand the measures included in the Tabular List, or necessary to prevent the Tabular List disclosure from being confusing or misleading.

Second, under the final rules, the “most important” determination is made on the basis of looking only to the most recently completed fiscal year, as opposed to “the time horizon of the disclosure,” as described in the Reopening Release. We believe this approach will alleviate commenters' concerns that identifying the “most important” measures would be difficult, particularly when companies have overlapping or interrelating measures, by narrowing the universe of measures to be considered when selecting the “most important” to those used in the prior year (instead of the prior five years). In addition, we believe focusing the disclosure on the registrant's “most recently completed fiscal year” will accommodate changes in compensation programs and in the compensation related to specific measures over time, and avoid situations where a registrant is disclosing measures that are no longer used in, or important to, its executive compensation program, but would still be “most important” based on the measure's usage in prior years disclosed in the table.

Finally, although the Reopening Release considered a list that would include the five most important measures, the final rules we are adopting require disclosure of at least three, and up to seven financial performance measures, and also permit registrants to include non-financial performance measures in that list. We believe that providing registrants with flexibility in the number of measures they can include in the list may also lessen the difficulty, noted by commenters, of identifying a registrant's “most important” measures. For example, a registrant with three, four, five, or six equally “most important” measures would not need to increase or decrease their “most important” measure disclosure to specifically disclose five measures. We acknowledge that, for certain issuers, this concern may still remain due to the minimum of three and limit of seven measures imposed by the final rules; however we are of the view that providing an upper bound for the list will reduce the risk of lengthy, overly complicated lists, which would fail to advance the statutory objective of providing clear and simple comparisons of pay with performance. In addition, we believe allowing an unlimited number of measures could in some cases result in misleading or confusing disclosures by obscuring which performance measures are principally driving compensation actually paid.

As discussed in the Reopening Release, the final rules specify that measures required to be included in the Tabular List are financial performance measures that, in the registrant's assessment, represent the most important financial performance measures used by the registrant to link compensation actually paid during the fiscal year to company performance. As discussed in the Reopening Release, we believe that a list of the measures that the registrant assesses to be the “most important” may enable investors to more easily assess which performance measures actually have the most impact on compensation actually paid and make their own judgments as to whether that compensation appropriately incentivizes management. In addition, we believe this list will provide investors with helpful context for interpreting the pay-versus-performance disclosure, more generally, particularly when analyzing the other measures included in the table, by showing which (if any) of those measures are considered “important” by the registrant, in determining pay. While we recognize that some commenters supported permitting non-financial performance measures to be included in the list, the final rules specify that the only required disclosures in the Tabular List are “financial performance measures” given the “financial performance” language in Section 14(i). However, in response to commenters, the final rules provide that registrants have the option of including non-financial performance measures in the Tabular List. Registrants may do so only if such measures are included in their three to seven most important performance measures, and they have disclosed at least three (or fewer, if the registrant only uses fewer) most important financial performance measures. Regardless of whether registrants elect to disclose non-financial performance measures in their Tabular List, they still may only disclose a maximum of seven measures in the list.

Under the final rules, registrants may disclose the Tabular List in three different ways. First, registrants may present one list with at least three, and up to seven, performance measures, which in the registrant's assessment represent the most important performance measures used by the registrant to link compensation actually paid to the registrant's NEOs, for the most recently completed fiscal year, to company performance, similar to the ranked list contemplated in the Reopening Release.

Second, registrants may break up the Tabular List disclosure into two separate lists: one for the PEO and one for the remaining NEOs. Third, registrants may break up the Tabular List disclosure into separate lists for the PEO and each NEO. If the registrant elects to provide the Tabular List disclosure in multiple lists (the second or third options, described above), each list must include at least three, and up to seven, financial performance measures. As in situations where a registrant elects to provide one Tabular List, registrants electing to provide the Tabular List disclosure in multiple lists may include non-financial performance measures in such lists if such measures are among their most important performance measures. Requiring the Tabular List to include measures related to both PEO and NEO compensation is consistent with the approach taken throughout Item 402(v) of Regulation S-K and we believe this consistency in disclosure will make the disclosure more readily understandable to investors.

As noted above, commenters suggested that such a list was beyond the scope of the Dodd-Frank Act mandate, and that similar disclosure is already available in the CD&A. We believe the Tabular List would further the objectives of the Section 14(i) mandated disclosure, as it provides another avenue for investors to understand the relationship between executive compensation actually paid and the registrant's financial performance. It is within our authority to specify the form and content of this disclosure as well as to require additional information to enhance the Dodd-Frank Act mandated disclosures. While it is possible that some registrants provide similar disclosure in the CD&A, we note that the CD&A requires disclosure of performance measures that are “material elements of the registrant's compensation of named executive officers,” not the “most important” measures used by the registrant to link executive compensation actually paid to company performance. There would be an overlap between those two disclosure requirements when the “most important” measures are also “material elements of the registrant's compensation of named executive officers”; however, they are not necessarily the same. Even in situations where the performance measures included in the Tabular List are already included in CD&A disclosure, we believe that the presentation of the measures in the Tabular List should allow investors to more readily understand what measures in the registrant's view are the “most important” to its compensation program, and thus better understand the relationship between registrant performance and executive compensation, as the statute provides.

Finally, as considered in the Proposing Release, under the final rules, registrants may cross-reference to other disclosures elsewhere in the applicable disclosure document that describe the registrant's processes and calculations that go into determining NEO compensation as it relates to these performance measures, if they elect to do so.

4. Requirement To Disclose a Company-Selected Measure

i. Amendments Considered in the Reopening Release

The Reopening Release requested comment on requiring registrants to disclose a Company-Selected Measure—a measure that in the registrant's assessment represents the most important performance measure (that is not already included in the table) used by the registrant to link executive compensation actually paid during the fiscal year to company performance, over the time horizon of the disclosure. We considered adding this requirement in order to both provide additional useful disclosure to investors regarding the measures the registrant actually used to set compensation, and to lessen the likelihood that the mandated measures in the tabular disclosure would misrepresent or provide an incomplete picture of how pay relates to performance. We believed that requiring disclosure of a Company-Selected Measure would not be overly burdensome on registrants, as, by definition, the Company-Selected Measure would be a measure already considered by registrants when making executive compensation determinations, and may already be discussed, in a different form, in the CD&A.

ii. Comments

A number of commenters provided feedback on potential disclosure of a Company-Selected Measure, as discussed in the Reopening Release. Some commenters supported mandatory disclosure of a Company-Selected Measure, with one suggesting that the Company-Selected Measure (or Measures) should be the only mandated performance measure(s). One commenter, who generally favored requiring registrants to disclose “all” measures used by registrants in linking executive compensation paid to performance, suggested that the Company-Selected Measure should be limited to financial measures, to provide an “alternative” to TSR, and suggested that companies should be permitted to omit the Company-Selected Measure if they do not have a single measure used to assess financial performance for compensation purposes. Another commenter suggested requiring the disclosure of multiple Company-Selected Measures, such as three such measures, with corresponding peer group disclosure to prevent registrants from “cherry-pick[ing] measures.” Other commenters suggested that the Company-Selected Measure should be based on the compensation paid to the PEO, not all of the NEOs.

Some commenters suggested that the Company-Selected Measures be disclosed alongside the methodology used to calculate it, with two commenters specifically suggesting the Company-Selected Measure must be “auditable/assurable” or accompanied by “an explanation of its calculation and a complete GAAP reconciliation, if possible.” Two commenters specifically said that, if ESG metrics are used as Company-Selected Measures, additional information about the metrics used should be disclosed.

A number of commenters opposed the mandatory inclusion of a Company-Selected Measure, stating that the idea that there is one “most important” measure “oversimplifies” the compensation setting process, and that different measures cannot be considered in “isolation.”

As discussed above, a number of commenters supported allowing the companies' “most important” measures to be non-financial measures, with some supportive of allowing non-financial measures to be a registrant's Company-Selected Measure. Other commenters opposed either allowing non-financial measures to be included as Company-Selected Measures, indicating that doing so “would be at odds with both the language and intent of Section 953(a),” or requiring or encouraging companies to incorporate ESG metrics in setting executive pay.

Commenters were divided on whether Company-Selected Measures should be permitted to be changed from year to year, and if so, what disclosure should be required. One commenter was directly opposed to regular changes in the Company-Selected Measure, stating the measure should be required to remain the same for at least five years, in order to avoid companies rationalizing the “best” measure each year. Other commenters supported allowing annual changes to the Company-Selected Measure, so long as accompanying disclosure about the reason for the change or a period of disclosure of the `old' and `new' measures was provided. One commenter alternatively suggested that the Company-Selected Measure should be the “most important” measure over a given period, and not the “most important measure” for all five years in the table.

One commenter suggested that, if the “most important” measure is already included in the tabular disclosure, the next-most important measure should be included as the Company-Selected Measure, while another commenter (who generally opposed the inclusion of the Company-Selected Measure) stated that, if it is a measure otherwise required to be disclosed in the table, the Company-Selected Measure should be able to be an already-included measure.

iii. Final Amendments

The final rules require registrants to disclose a Company-Selected Measure in the table required under new 17 CFR 229.402(v)(1). The Company-Selected Measure must be a financial performance measure included in the Tabular List, which in the registrant's assessment represents the most important performance measure (that is not otherwise required to be disclosed in the pay-versus-performance table required under new Item 402(v) of Regulation S-K) used by the registrant to link compensation actually paid to the registrant's NEOs, for the most recently completed fiscal year, to company performance. If the registrant's “most important” measure is already included in the tabular disclosure, the registrant would select its next-most important measure as its Company-Selected Measure. As discussed above, registrants would also be required to provide a clear description of the relationship of the Company-Selected Measure to executive compensation actually paid, in narrative or graphical form, or a combination of the two.

We believe that providing a quantified Company-Selected Measure, along with the Tabular List, will provide investors with useful context for understanding the measures actually used by registrants in their compensation programs. In order to allow investors to understand the measure that is most important, we are only requiring registrants to provide one Company-Selected Measure. However, we recognize some registrants may have additional performance measures (including non-financial measures) that they believe are “important” measures and that could warrant quantified disclosure. We note that, under the Plain English principles (discussed below ), registrants may provide additional performance measures as new columns in the table. However, such additional disclosures may not be misleading or obscure the required information, and the additional performance measures may not be presented with greater prominence than the required disclosure. If a registrant elects to provide any additional performance measures in the table, each additional measure must also be accompanied by a clear description of the relationship between executive compensation actually paid to the registrant's PEO, and, on average, to the other NEOs, and that measure. We believe clarifying that registrants have the flexibility to include additional measures will, to some degree, alleviate concerns raised by some commenters in response to the Reopening Release that selecting one Company-Selected Measure was overly simplistic and did not reflect how companies actually approach their compensation programs, while also providing registrants the opportunity to provide context to the other mandatory measures disclosed in the table.

As the Company-Selected Measure must be a measure included in the Tabular List, the determination of “most important” that registrants must use for selecting Company-Selected Measures is the same as the determination they must use for selecting required measures for the Tabular List ( i.e., the “most important” determination is made based on the most recently completed fiscal year and the measures required to be disclosed are financial measures of performance). We are limiting the measures required to be included in the Tabular List (and to be included as the Company-Selected Measure) to financial performance measures given the statutory language referencing “the relationship between executive compensation actually paid and the financial performance of the issuer.” We recognize that some registrants may consider one or more non-financial performance measures to be their most important measures for executive compensation purposes. In addition to the option under the final rules to include such measures in the Tabular list, under the Plain English principles, those registrants can supplement their mandatory pay-versus-performance disclosure with disclosure about those non-financial performance measures, as discussed below.

The table will include the numerically quantifiable performance of the issuer under the Company-Selected Measure for each covered fiscal year. For example, if the Company-Selected Measure for the most recent fiscal year was total revenue, the company would disclose its quantified total revenue performance in each covered fiscal year. The Company-Selected Measure could change from one filing to the next, and we acknowledge that some commenters were concerned that registrants may change their Company-Selected Measure in order to present the relationship of pay to performance in a positive light. However, we believe limiting the Company-Selected Measure to compensation linked to performance for the most recently completed fiscal year will provide investors with visibility into the registrant's current executive compensation program, and avoid situations in which the Company-Selected Measure is not a measure that is currently used by the registrant ( i.e., when a measure is only the “most important” measure based on historical usage). In addition, as is the case for the Tabular List, we believe limiting the Company-Selected Measure to the most recent fiscal year will allow registrants to more easily calculate and assess which measure is the “most important.”

Similarly to the Tabular List, we do not believe it would be appropriate to limit the Company-Selected Measure to a measure relating only to the PEO's compensation, because our understanding is that Congress intended for the rules to provide disclosure about both PEOs and the remaining NEOs.

We are not mandating that the methodology used to calculate the Company-Selected Measure be included in the registrant's disclosure. However, as discussed above, registrants will be required to provide a narrative, graphical, or combined narrative and graphical description of the relationships between executive compensation actually paid to the PEO, and, on average, the other NEOs, and the Company-Selected Measure, and may cross-reference to other disclosures elsewhere in the applicable disclosure document that describe the processes and calculations that go into determining NEO compensation as it relates to the Company-Selected Measure, if they elected to do so. In addition, registrants are permitted to supplement their Company-Selected Measure disclosure, so long as any additional disclosure is clearly identified as supplemental, not misleading and not presented with greater prominence than the required disclosure.

Further, we recognize that a registrant's Company-Selected Measure, or additional measures included in the table, may be non-GAAP financial measures. Under existing CD&A requirements, if a company discloses a target level that applies a non-GAAP financial measure in its CD&A, the disclosure will not be subject to the general rules regarding disclosure of non-GAAP financial measures, but the company must disclose how the number is calculated from its audited financial statements. Because the disclosure required by the final rules is intended, among other things, to supplement the CD&A, we believe it is appropriate to treat non-GAAP financial measures provided under Item 402(v) of Regulation S-K consistently with the existing CD&A provisions. As a result, the final rules specify that disclosure of a measure that is not a financial measure under generally accepted accounting principles will not be subject to Regulation G and Item 10(e) of Regulation S-K; however, disclosure must be provided as to how the number is calculated from the registrant's audited financial statements.

E. Time Period Covered

1. Proposed Amendments

We proposed requiring all registrants, other than SRCs, to provide the pay-versus-performance disclosure for the five most recently completed fiscal years, and requiring SRCs to provide disclosure for the three most recently completed fiscal years. We also proposed providing transition periods for registrants: SRCs would only be required to provide the Item 402(v) of Regulation S-K disclosure for the last two fiscal years in the first applicable filing after the rules became effective; and all other registrants would be required to provide the disclosure for three fiscal years, in the first applicable filing after the rules became effective, and to provide disclosure for an additional year in each of the two subsequent annual proxy filings where disclosure is required.

The Proposing Release also provided that the pay-versus-performance disclosure would only need to be provided for years in which a registrant was a reporting company pursuant to Section 13(a) of the Exchange Act or Section 15(d) of the Exchange Act (“Section 15(d)”), consistent with the phase-in period for new reporting companies in their Summary Compensation Table disclosure.

2. Comments

Several commenters supported the proposed disclosure periods, while several others generally opposed them. Some commenters who opposed the proposed disclosure periods stated that the periods were too short to measure management's performance; while others argued the periods were too long, creating burdensome costs for registrants, and were inconsistent with other approaches taken in the proxy statement.

Commenters suggested a number of different alternative time periods. Some commenters suggested permitting registrants to voluntarily disclose additional years in the tabular disclosure, while others opposed permitting additional years of disclosure. Some other commenters recommended the Commission use a three-year period, with some of those commenters noting that three-year periods will have less NEO turnover, meaning registrants will need to make less explanatory disclosure. One commenter suggested we only require the disclosure for one year. Another commenter suggested allowing registrants to set the time period covered, with a minimum requirement, such as three years. Finally, one commenter did not propose a specific time period, but rather suggested the longer the period the better.

Commenters were also divided on the suggested transition period. Some commenters supported the transition period, while one commenter opposed it. Others questioned whether there would be significant enough costs to justify applying a transition period. One commenter specifically supported a transition period for newly public companies.

Commenters offered a few alternatives to the proposed transition period, including a one-year transition period, not requiring reporting until the anniversary of the effective date of the rule, and a longer transition period.

3. Final Amendments

We are adopting the time periods as proposed. We believe that requiring registrants, other than SRCs, to provide pay-versus-performance disclosure for a five year period will provide a meaningful period over which a relationship between annual measures of pay and performance over time can be evaluated. Further, we are requiring that the disclosure be in order beginning with the most recent fiscal year. We believe that requiring a shorter time period, for all registrants, may not provide investors with enough data to evaluate the pay-versus-performance relationship, while requiring a longer period may be overly burdensome to registrants. We also believe that the scaled disclosure requirement under which SRCs may elect to provide three years of pay-versus-performance disclosure will provide investors with an appropriate time horizon over which to observe a relationship between pay and performance, while also remaining consistent with the scaled-disclosure approach generally applied to SRCs under our executive compensation rules. While SRCs generally are only required to provide two years of executive compensation disclosure in filings with the Commission, because the final rules include a transition period that permits an existing SRC to provide two years of disclosure, instead of three, in the first applicable filing after the rules become effective, and three years of disclosure in subsequent filings, we do not believe requiring three years of pay-versus-performance data will be unduly burdensome on SRCs.

We are also adopting the transition periods and the requirement that a registrant provide pay-versus-performance disclosure only for years that it was a reporting company pursuant to Section 13(a) or Section 15(d) of the Exchange Act, as proposed. We believe both of these provisions will mitigate concerns expressed by some commenters regarding the costs of the potential disclosure, while also, over time, providing investors with a meaningful way to evaluate a registrant's period pay-versus-performance disclosure. In order to give companies adequate time to implement the new disclosures, we are providing that companies are required to comply with Item 402(v) of Regulation S-K in proxy and information statements that are required to include the Item 402 of Regulation S-K disclosure for fiscal years ending on or after December 16, 2022.

With respect to some commenters' suggestions that we should permit registrants to voluntarily provide additional years of disclosure, as noted below, under the Plain English principles, the final rules will permit registrants to provide additional years of disclosure, so long as doing so would not be misleading and would not obscure the required information.

F. Permitted Additional Pay-Versus-Performance Disclosure

1. Proposed Amendments

We proposed applying the Plain English principles in 17 CFR 240.13a-20 and 17 CFR 240.15d-20 to the pay-versus-performance disclosures. We noted that, under those principles, registrants would be permitted to provide additional information beyond what is specifically required by the rules so long as the information is not misleading and would not obscure the required information. As discussed in the Proposing Release, we note that the Plain English principles applicable to compensation disclosure would permit registrants to “include tables or other design elements, so long as the design is not misleading and the required information is clear, understandable, consistent with applicable disclosure requirements, consistent with any other included information, and not misleading.”

2. Comments

Some commenters supported applying the Plain English principles to the pay-versus-performance disclosure, noting that their application would be beneficial for both investors and the financial community.

3. Final Amendments

The final amendments allow registrants to provide additional pay-versus-performance information beyond what is specifically required by Item 402(v) of Regulation S-K, so long as doing so would not be misleading and would not obscure the required information. For example, registrants that are already providing voluntary pay-versus-performance disclosures may generally continue to provide such disclosures in their present format, or could include disclosure of long-term performance metrics measured over periods longer than a single fiscal year. Subject to these same principles, registrants will be permitted to include additional compensation and performance measures, or additional years of data, in the newly required table. Any supplemental measures of compensation or financial performance and other supplemental disclosures provided by registrants must be clearly identified as supplemental, not misleading, and not presented with greater prominence than the required disclosure. For example, depending on the facts and circumstances, a registrant could use a heading in the table indicating that the disclosure is supplemental, or include language in the text of its filing stating that the disclosure is supplemental. As noted above, to the extent additional performance measures are included in the table, these must also be accompanied by a clear description of their relationship to executive compensation actually paid to the PEO and to the average such compensation of the other NEOs. As discussed in the Proposing Release, and noted by commenters, we believe applying the Plain English principles to the pay-versus-performance disclosure will facilitate investors' understanding and decision-making with respect to the pay-versus-performance disclosure.

G. Required Disclosure for Smaller Reporting Companies

1. Proposed Amendments

The Proposing Release would have required SRCs to provide disclosure under Item 402(v) of Regulation S-K, but the disclosure would be scaled for those companies, consistent with SRCs' existing scaled executive compensation disclosure requirements. Specifically, as proposed, SRCs would:

• Only be required to present three, instead of five, fiscal years of disclosure under new Item 402(v) of Regulation S-K;
• Not be required to disclose amounts related to pensions for purposes of disclosing executive compensation actually paid;
• Not be required to present peer group TSR;
• Be permitted to provide two years of data, instead of three, in the first applicable filing after the rules became effective; and
• Be required to provide disclosure in the prescribed table in XBRL format beginning in the third filing in which it provides-pay-versus performance disclosure.

In the Reopening Release, the Commission indicated that it was considering requiring SRCs to disclose the income or loss before income tax expense and net income measures, but not the Company-Selected Measure or the list of their five most important measures.

2. Comments

Some commenters supported fully exempting SRCs from the pay-versus-performance disclosure requirements, stating that the disclosure requirements would be disproportionally burdensome to SRCs; executive disclosure issues are less acute at SRCs; and TSR is a more problematic measure for SRCs due to the relative illiquidity and volatility of SRCs' shares. One commenter suggested that the Commission exempt SRCs from the disclosure requirements for five years, so that the Commission could first analyze the impact of the disclosure requirements on larger registrants. Another commenter suggested that the pay-versus-performance disclosure be voluntary for SRCs.

Other commenters stated that we should not exempt SRCs from the disclosure requirements. One commenter opposed to exempting SRCs indicated that a lack of transparency could have negative market effects for SRCs. In addition, one commenter specifically supported requiring SRCs to disclose income or loss before income tax expense, net income, the Company-Selected Measure, and the list of the five most important measures.

With respect to the timing of the disclosure, one commenter, who supported SRCs being subject to the full pay-versus-performance disclosure requirement, suggested a one year “grace period.” Another commenter suggested that SRCs provide five years of data, but that we provide SRCs with a three year transition period requiring two years of data in the first applicable filing after the rules became effective, and increasing until the fourth applicable filing after the rules become effective, when all five years of data would be required.

As discussed above, a number of commenters supported requiring all registrants to use Inline XBRL to tag their pay-versus-performance disclosure, with one specifically stating that all filers are now familiar with Inline XBRL. On the other hand, one commenter specifically suggested, in response to the Proposing Release, that we exempt SRCs from any XBRL tagging requirement.

3. Final Amendments

We are adopting the scaled disclosure requirements for SRCs as described in the Proposing Release (and with respect to the net income measure, the Reopening Release). For the reasons noted above, we believe requiring SRCs to provide three instead of five years is appropriate, and is aligned with SRCs' existing scaled executive compensation disclosure requirements. While the three-year period applicable for the disclosure is longer than what SRCs currently are required to disclose in the Summary Compensation Table, we believe the pay-versus-performance calculations, or the information required to make the calculations, for the additional year would generally be available in SRCs' disclosures from prior years.

We also believe that requiring SRCs to provide peer group TSR, a Company-Selected Measure, a Tabular List, or disclose amounts related to pensions would be unduly burdensome for SRCs, which, unlike larger registrants, are not otherwise required to present the TSR of a peer group or disclosure of how executive compensation relates to performance in a CD&A, and are subject to scaled compensation disclosure that does not include pension plans. Finally, we believe a transition period that would permit SRCs to provide two years of data, instead of three, in the first applicable filing after the rules become effective is appropriate, as is a phase-in period to allow SRCs to provide the required Inline XBRL data beginning in the third filing in which it provides pay-versus-performance disclosure, instead of the first.

III. Other Matters

If any of the provisions of these rules, or the application thereof to any person or circumstance, is held to be invalid, such invalidity shall not affect other provisions or application of such provisions to other persons or circumstances that can be given effect without the invalid provision or application.

Pursuant to the Congressional Review Act, the Office of Information and Regulatory Affairs has designated these rules as not a “major rule,” as defined by 5 U.S.C. 804(2).

IV. Compliance Dates

In order to give companies adequate time to implement these disclosures, we are requiring registrants to begin complying with Item 402(v) of Regulation S-K in proxy and information statements that are required to include Item 402 disclosure for fiscal years ending on or after December 16, 2022.

V. Economic Analysis

A. Background

We are adopting these final rules to satisfy the statutory mandate of Section 14(i). The Senate Report that accompanied the statute references shareholder interest in the relationship between executive pay and performance as well as the general benefits of transparency of executive pay practices. As discussed above, we believe that the statute is intended to provide further disclosures concerning a registrant's executive compensation program for shareholders to consider when making related voting decisions, such as decisions with respect to the shareholder advisory vote on executive compensation, votes on other compensation matters, and director elections.

The final rules require the disclosure of information that is largely already reported under current disclosure rules, but that is currently not computed or presented in the way the final rules will require. This repackaging of some of the information from existing disclosures into the required pay-versus-performance disclosure is intended to allow investors to more quickly or easily process the information accurately.

The final rules require registrants to present the values of prescribed measures of executive compensation and financial performance for each of their five most recently completed fiscal years (three years for SRCs) in a standardized table in proxy or information statements in which executive compensation disclosure is required. Registrants will also be required to provide “clear descriptions” of the relationships between the compensation and performance measures in the table (and between TSR and peer group TSR), but will be allowed to choose the format used to present the relationships, such as graphical or narrative descriptions (or a combination of the two). The final rules will also allow registrants to supplement the required elements of the disclosure with additional measures or additional years of data, subject to certain restrictions. Registrants will be required to provide the disclosure in a structured data language using Inline XBRL.

The final rules reflect several modifications relative to the proposed rules in response to comments received. For example, one area of significant comment on the Proposing Release was the proposal's reliance on TSR (and, for registrants other than SRCs, peer group TSR) as the exclusive measure of financial performance used to present the relation of pay with performance. The Reopening Release discussed, solicited comment on, and analyzed the economic effects of some possible additional measures of financial performance that the Commission was considering requiring. The final rules introduce two of these additional measures to the table: net income and, for registrants other than SRCs, a Company-Selected Measure. In addition, the final rules require registrants other than SRCs to provide a Tabular List of the most important financial performance measures used to link executive compensation actually paid, for the most recent fiscal year, to company performance. The additions will broaden the picture of registrant performance presented in the disclosure, providing additional detail and context that could enhance the usefulness of the disclosure by certain registrants or for certain investors. The additions will also entail some additional compliance costs and could make it more difficult for investors to quickly review the disclosure.

Many commenters to the Proposing Release also raised concerns that, under the proposed approach, the year to which company performance would be attributed and the year in which associated pay would be recognized would frequently be mismatched, which could significantly limit the usefulness of the proposed disclosure. To address these comments, the final rules require equity awards to be revalued more frequently than had been proposed in order to better align pay and any related performance, at the expense of somewhat greater costs to registrants of computing the prescribed measure of pay.

We are mindful of the costs and benefits of the final rules. The discussion below addresses the economic effects of the final rules, including their anticipated costs and benefits, as well as the likely effects of the final rules on efficiency, competition, and capital formation. The final rules reflect the statutory mandate in Section 14(i) as well as the discretion we exercise in implementing that mandate. For purposes of this economic analysis, we address the costs and benefits resulting from the statutory mandate and from our exercise of discretion together, recognizing that it is difficult to separate the costs and benefits arising from these two sources. We also analyze the potential costs and benefits of significant alternatives to the final rules.

B. Baseline

To assess the economic impact of the final rules, we are using as our baseline the current state of the market without a requirement for registrants to disclose the relationship between executive compensation actually paid and the financial performance of the registrant.

1. Affected Parties

We consider the impact of the final rules on investors and registrants (and their NEOs). The final rules will apply to all companies that are registered under Section 12 of the Exchange Act (“Section 12”) and are therefore subject to the Federal proxy rules, except EGCs. The final rules will also not apply to foreign private issuers or companies with reporting obligations only under Section 15(d) of the Exchange Act, which are not subject to the proxy rules. In addition, for some Section 12(g) of the Exchange Act (“Section 12(g)”) registrants, such as limited partnerships, the disclosure requirement might not apply in some or all years because these registrants might not file either proxy or information statements every year.

We estimate that approximately 4,530 registrants will be subject to the final rules, including approximately 1,860 SRCs. The proportion of SRCs among the affected registrants is expected to be similar to that which was reported at the time of the Proposing Release. Among all registrants subject to the Federal proxy rules, we estimate that there are approximately 1,275 EGCs, of which approximately 1,065 are also SRCs, none of which will be subject to the final rules.

2. Existing Disclosures and Analyses

The registrants that will be subject to the final rules must currently comply with Item 402 of Regulation S-K, which requires the disclosure of extensive information about the compensation of NEOs, and, except in the case of SRCs, with Item 201(e) of Regulation S-K, which requires graphical disclosure of registrant TSR and peer group TSR. They are also subject to financial statement and disclosure requirements under Regulation S-X. The underlying information necessary to provide the required pay-versus-performance disclosure is, with limited exceptions discussed below, already encompassed by these existing disclosure requirements. However, the existing disclosures might not present the underlying information in a format that allows investors to readily assess the alignment of pay and performance.

Under the final rules, the definition of executive compensation actually paid for a fiscal year is, generally, total compensation as reported in the Summary Compensation Table for that year (i) less the change in the actuarial present value of pension benefits, (ii) less the grant-date fair value of any stock and option awards granted during that year, (iii) plus the pension service cost for the year and, in the case of any plan amendments (or initiations), the associated prior service cost (or less any associated credit), and (iv) plus the change in fair value of outstanding and unvested stock and option awards during that year (or as of the vesting date or the date the registrant determines the award will not vest, if within the year) as well as the fair value of new stock and option awards granted during that year as of the end of the year (or as of the vesting date or the date the registrant determines the award will not vest, if within the year). Adjustments (i) and (iii) with respect to pension plans will not apply to SRCs because they are not otherwise required to disclose executive compensation related to pension plans.

Under the baseline, investors generally should already have the required data to compute a reasonable estimate of executive compensation actually paid as defined in the final rules, even though registrants are not required to compute or disclose this measure. Specifically, under existing requirements of Item 402 of Regulation S-K, registrants must report, in the Summary Compensation Table, the value of total compensation and each of its components, including the aggregate grant-date fair value of equity awards and, for registrants other than SRCs, the total change (if positive) in actuarial present value of pension benefits, for each NEO. The total compensation and amounts required to be subtracted from this total in the computation of executive compensation actually paid for each NEO, or adjustments (i) and (ii) referenced above, are thus already available in the Summary Compensation Table.

The amounts that must be added back in this computation, or adjustments (iii) and (iv) referenced above, are not required to be directly reported under existing disclosure requirements, but can be estimated based on existing disclosures. In particular, Item 402 of Regulation S-K requires further disclosure about equity awards and pension plans, such as, for non-SRCs, the Grant of Plan-Based Awards Table and the Pension Benefits Table and the associated narrative and footnotes, which include the detailed terms of these components of compensation and certain valuation assumptions. Using these existing disclosures and other public data, it is possible for investors to make reasonable (though perhaps not identical) estimates of the annual and vesting-date fair values of outstanding stock and option grants. In fact, various third parties, such as proxy advisory service providers and compensation consultants, currently make similar computations using existing disclosures in order to construct alternative pay measures as part of the services they provide to certain investors and/or registrants. Market participants other than those providing actuarial services may have less experience with the computations required with respect to pension plans. However, it is still possible to compute an estimate of pension service cost for the year (plus the prior service cost, or credit, associated with any plan amendments or initiations) by using existing disclosures and public data to construct the required actuarial assumptions and computations.

That said, these computations can be complex and investors would bear costs to make such computations or obtain them from third parties. Further, if investors or third parties were to estimate executive compensation actually paid based on existing disclosures, these estimates may differ from each other and from similar estimates made by registrants themselves. For example, because registrants are not currently required to disclose the equity valuation assumptions that they would apply at any time after the grant date (which may differ from the grant-date assumptions), investors may not know how the registrant would apply its discretion in choosing from a range of reasonable assumptions to compute fair values at these other dates. Estimates constructed by or on behalf of investors may also differ from registrant estimates if simplifications are made in order to more easily produce estimates for a large number of registrants.

Information about registrant financial performance is readily available to investors under the baseline. The final rules require the disclosure of historical TSR, peer group TSR, and net income for up to five years. Disclosure of historical TSR and TSR of a particular peer group is already required under Item 201(e) of Regulation S-K: specifically, this item requires the disclosure of the TSR for the registrant as well as a peer group (a published industry or line-of-business index, peer issuers selected by the registrant, or issuers with similar market capitalizations), for the past five years, in annual reports. The final rules allow registrants to choose to use either the peer group required under Item 201(e) of Regulation S-K or, if the registrant uses a peer group in benchmarking its compensation, the peer group disclosed in its CD&A in its pay versus performance disclosure. In the latter case, however, the components of such a peer group would be disclosed in the CD&A and the shareholder returns of these companies would be publicly available from many sources, if not already reported in the CD&A. Similarly, while SRCs are not required to comply with Item 201(e) of Regulation S-K or CD&A disclosure requirements and yet would still have to report their own TSR under the final rules, data about their returns is publicly available. The final rules do not require SRCs to present the TSR of a peer group. Finally, all of the affected registrants are currently required to disclose net income as part of their financial reports filed in Form 10-K, including three years of data for registrants other than SRCs, and two years of data for SRCs, with additional history generally available in previous filings.

We expect that the quantitative disclosure of Company-Selected Measures called for in the new disclosures is also generally encompassed by existing financial statement disclosure requirements or voluntarily disclosed in existing proxy statements. However, if registrants do not already disclose historical quantitative data for these measures over the past five years, the required disclosure may provide new information relative to the baseline to the extent that any computations required to derive the value of these measures from reported financial data may not always be straightforward for investors to replicate. The disclosure of a Company-Selected Measure may also provide investors with new information in the form of any insight gained based on the registrant's choice of which of the measures reported in the CD&A in this or previous years was deemed to be the most important with respect to the most recent fiscal year.

While the bulk of the information about compensation and registrant performance to be included in the new disclosure is currently available to investors elsewhere, not all of this information is accessible for large-scale analysis under the baseline. Currently, every affected registrant is, or will soon be, subject to Inline XBRL tagging requirements for a subset of its other Commission disclosures, including the financial statements and financial statement footnotes. Thus, information that is already available from these sources—such as net income, some Company-Selected Measures or statistics used to compute these measures, and information in footnotes regarding inputs and assumptions used to compute pension liabilities and stock-based compensation expense—is already tagged and thus readily machine-readable. However, other information that will be reflected in the required pay-versus-performance disclosure, such as the compensation measures, as well as most of the information required to compute these measures, is not currently tagged, and could therefore become more readily available for analysis as a result of the final rules.

For the affected registrants other than SRCs, Item 402 of Regulation S-K requires a description in the CD&A of how the registrant's compensation policy relates pay to performance, if material to the registrant's compensation policies and decisions. This description must include information about any performance targets that are a material element of a company's executive compensation policies or decisions. While the final rules will newly require registrants other than SRCs to name the top three to seven most important performance measures used by the registrant to link NEO pay to performance in the most recent fiscal year, these registrants likely already disclose these measures in the CD&A under existing requirements. However, as in the case of the Company-Selected Measure, the Tabular List may provide new information relative to the baseline in the form of any insight gained based on the registrant's choice of which of the measures reported in the CD&A were deemed to be the most important with respect to the last completed fiscal year.

Registrants are not currently required to disclose, in a side-by-side fashion, or report the actual historical relationship between, any measures of executive compensation and registrant financial performance. As discussed in the Proposing Release, some registrants voluntarily provide such disclosures, which are generally limited to analyses of the compensation of the PEO and which vary with regard to the compensation and performance measures used. Such voluntary disclosures remain a minority practice, with the rate of such disclosures declining somewhat since the time of the Proposing Release, and they remain highly varied. Whether or not they directly disclose the relationship of pay with performance, some registrants disclose alternative measures of pay to demonstrate the variation in the value of pay after it is granted, but, again, this is a minority practice and the measures used vary. Thus, even when voluntary disclosures are provided, their comparability is limited, which can make them difficult for investors to use. Commenters and other observers have also raised concerns that registrants choose to present measures that make the alignment of pay and performance appear more favorable.

Certain investors also have access to analyses of historical pay-versus-performance data produced by third parties, such as proxy advisory firms and compensation consultants. These analyses are based on compensation and performance information disclosed by registrants. Compared to voluntary disclosures by registrants, these third-party analyses are available for a larger number of registrants, and apply more consistent methodologies across registrants. However, this consistency has led to criticism that the analyses are not appropriately tailored to the circumstances of different kinds of registrants. Further, these analyses are only available to investors who pay for these services, and the computations and analytical approaches used vary across the third-party information providers. Some other investors generate their own pay-versus-performance analyses for the registrants in their portfolios, using a variety of approaches. Given the resources required, smaller investors, particularly retail investors, are the least likely, under the baseline, to subscribe to third party services or to do their own detailed pay-versus-performance computations for each of their holdings.

As was the case at the time of the Proposing Release, there continues to be no consensus around the best approach to analyzing the alignment of pay and performance, and we do not have complete information about the approaches used by all investors. However, the varied statistics and analyses that we can observe investors using may still shed some light on the type of information that they find to be useful for this purpose, particularly as many of the third-party analyses have evolved over time based on shareholder demand. For example, while many third party and shareholder analyses use a measure of pay based on grant date valuations of stock and options, potentially because this has historically been the most readily available measure, most of the recent analyses that we have observed also include a “realizable pay” measure. While there are various approaches to defining and computing “realizable pay,” it is generally intended to capture both pay that has been realized by an executive in the period as well as an updated value, to reflect actual company performance, of outstanding equity awards that could potentially be realized in the future. A recent survey by one proxy advisory firm found that 84 percent of investors support the use of an outcomes-based pay measure such as realizable pay in a quantitative pay-for-performance evaluation, further demonstrating investor demand for such computations.

With respect to performance measures, the analyses by or on behalf of investors that we observe all use TSR as a primary measure of performance. However, most also supplement TSR with other measures of financial performance. For example, some of the performance measures presented by third parties as part of pay-for-performance analyses in recent years include operating cash flow growth; earnings per share growth; growth in earnings before interest, taxes, depreciation and amortization (“EBITDA”); return on equity; return on invested capital; return on assets; and various ratios and growth rates using “economic value added.” The inclusion of these measures may demonstrate investors' interest in additional measures of performance, particularly with respect to profitability, when considering compensation. Shareholder demand for such information is further supported by a recent survey by one proxy advisory firm, in which 84 percent of investors surveyed supported the continued reporting of some of the profitability measures listed above as part of the proxy advisory firm's proxy research in the area of pay-for-performance.

Overall, we have observed, and commenters have identified, an increasing sophistication in how investors are evaluating executive compensation disclosures as well as an increasing refinement in how registrants are crafting these disclosures, particularly after about a decade of experience with “say-on-pay” votes. However, despite the significant amount of information about executive compensation disclosed by registrants under the baseline, investors have expressed some discontent with current disclosures. For example, commenters have indicated that existing disclosures can be challenging to review, in that investors find it difficult to collect or interpret the information in which they are interested. Commenters also highlighted shareholder concerns about the length and complexity of existing compensation disclosures. These disclosures have generally increased in length since the time of the Proposing Release.

3. Executive Compensation Practices

The structure of executive compensation, and how it varies across the affected registrants, will influence the effects of the final rules and how those effects will vary across registrants. For example, because the final rules require that equity awards and compensation related to pension plans be reflected differently than in the Summary Compensation Table, the prevalence and variation in usage and design of these items in executive compensation packages may affect the benefits of the disclosures as well as the burden involved in making the required calculations to provide the disclosures. Similarly, variation in the number and nature of performance metrics in executive compensation plans may also affect the variation in costs and benefits of the final rules across registrants.

The final rules require that executive compensation actually paid include the annual change in fair value, through year-end or the vesting date, if earlier, of any outstanding stock and option awards. A majority of CEO direct compensation is delivered in the form of such equity awards, and their contribution to the total value of such compensation at the grant date has grown in recent years. The use of stock grants, and the frequency of such grants to the CEO, by some of the potentially affected registrants is reported in the table below.

Per the first row of each panel of Table 1, roughly 80 to 90 percent of registrants, both large and small, make use of stock grants to CEOs and other NEOs in a given year. The last row of the first panel of Table 1 indicates that about two-thirds of registrants, and slightly more among the largest registrants, make such grants to the CEO every year. The prevalence and frequency of stock grants have not changed markedly since the time of the Proposing Release.

The use of option grants, and the frequency of such grants to the CEO, by some of the potentially affected registrants is reported in the table below.

Per the first row of the first panel of Table 2, roughly 30 percent of the largest registrants, and about 20 percent of smaller registrants, grant options to their CEOs in a given year. This represents a significant drop, of greater than half, in the use of options to incentivize CEOs across all categories since the time of the Proposing Release. The decline in option grants to CEOs has largely been offset by an increase in the number and size of performance-contingent stock grants, marking the continuation of a trend also discussed in the Proposing Release. Per the first row of the second panel of Table 2, the granting of options to any other NEO is a bit more prevalent, with roughly 40 percent of the largest and about 30 percent of smaller registrants using such grants in a given year, but these rates have also dropped significantly since the time of the Proposing Release. In contrast to stock grants, option grants are also less frequent; per the last row of the first panel of Table 2, about 10 to 15 percent of registrants grant options to the CEO every year.

Because the final rules require the valuation of equity awards annually until the time of vesting, we have also considered the variation in vesting schedules. Equity awards may be subject to time-based or performance-based vesting, or a combination of the two. Awards with time-based vesting may vest in full at the end of their vesting period (“cliff vesting”) or in increments over the period of vesting (“graded vesting”).

Market practices regarding vesting schedules have remained relatively consistent since the time of the Proposing Release. We estimate that about 45 percent of stock grants are subject to time-based vesting, though this has declined slightly (by about three percentage points) since the time of the Proposing Release with the growth in reliance on performance-contingent stock. Of the time-vesting stock awards, roughly one-third have cliff-vesting schedules while the vast majority of the remaining have graded vesting in annual increments. For the stock awards that vest based on achieving performance conditions (approximately 55 percent of stock awards), the vast majority have cliff-vesting schedules. Approximately ten percent of awards with performance-based vesting also have an additional time-based vesting period at the end of the performance period. For option awards, the vast majority have time-based, graded vesting in annual increments. Given the decline in option awards (which tend to have graded vesting schedules) and the increasing prevalence of performance-contingent stock (which tends to cliff-vest) discussed above, there has been a corresponding increase in cliff-vesting overall.

For affected registrants other than SRCs, compensation related to pension plans is also measured differently in executive compensation actually paid, as reported under the final rules, than it is in the Summary Compensation Table. The use of pension plans and the years of credited service at some of the potentially affected registrants are reported in the table below.

There has been a decrease of about ten percentage points in the prevalence of pension plans for CEOs or other NEOs since the time of the Proposing Release. Per Table 3, such pension plans, and, for those with pension plans, a higher number of years of creditable service, remain more common among larger registrants. For the affected registrants other than SRCs, the final rules require that executive compensation actually paid include only the service cost for the year (and any prior service cost, or credit, associated with plan amendments or initiations), a value which is not currently required to be reported at this disaggregated level and which will usually differ from the total change in actuarial value of pension benefits included in total compensation reported in the Summary Compensation Table. In particular, the value currently included in total compensation reflects the change in actuarial pension value related to changes in the value of benefits accrued in prior years as well as the value of benefits earned during the applicable fiscal year. As such, the value currently included with respect to pensions in total compensation reported in the Summary Compensation Table will generally be more volatile (because of changes in interest rates and other actuarial assumptions) than the value to be included with respect to pensions in the executive compensation actually paid measure. The degree of difference between these two computations will generally increase with an executive's total number of years of credited service (and thus the extent of benefits already accumulated) under the pension plan.

Besides the decreased prevalence of option awards and pension plans, and the increased reliance on performance contingent-stock awards, there have also been changes since the time of the Proposing Release in the performance metrics used by registrants in their incentive plans. For example, as noted in the Reopening Release, there appears to have been a decline in the use of TSR as the sole metric used in long-term incentive plans, in those cases where the awards' vesting or quantities are contingent on one or more performance metrics. Among large companies, most use one to three financial metrics in their CEO's long-term incentive plan, with two metrics being the most common number. The most commonly used metric among these companies is still TSR, followed by profitability measures (particularly measures of operating income), and then scaled profitability measures (such as return on equity or return on invested capital). Commenters pointed out that the metrics used are often non-GAAP financial measures.

Some commenters indicated that another recent change in compensation practices has been an increased linkage of pay to ESG performance. Our research confirms that this appears to be a growing practice, but that consideration of ESG metrics does not often seem to be tied to specific quantitative goals and that ESG metrics are generally used in short-term incentive plans. These plans, such as annual bonus programs, generally make up a significantly smaller portion of total executive pay as compared to long-term incentive plans. As in the case of metrics for long-term incentive plans, among large companies, most use one to three financial metrics in their CEO's short-term incentive plan, with two financial metrics being the most common. The most commonly used metric among these companies is profitability (particularly measures of operating income), followed by revenues, and then measures of cash flow. It is also common to include business unit performance goals and non-financial metrics, such as measures of individual performance, strategic goals, or ESG metrics. There may be overlap in the measures used in executive's short-term incentive plans and those used in their long-term incentive plans, but more often than not these metrics are different.

There is no consensus in the market on the number of metrics that should be used in designing executive compensation, with some advocating for the use of more metrics and others advocating for fewer.

Overall, it is clear that the structure of executive compensation continues to evolve, as noted by commenters, and further changes may be on the horizon. For example, recent tax law changes and concerns about the complexity and effectiveness of performance-contingent stock awards could encourage registrants to reduce their reliance on such awards. Uncertainty in the wake of the COVID-19 pandemic and lower say-on-pay approval at large companies in recent years, as compared to previous years, could also drive changes in compensation structure, though it remains difficult to predict whether these factors will have lasting effects and what such effects are likely to be.

C. Discussion of Economic Effects

The final rules require registrants to present, in one location, information that for the most part is disclosed in various other locations (and using different computations) under existing rules, and to tag the new disclosure using a machine-readable data language (Inline XBRL). The anticipated benefits and costs of the final rules are therefore driven by the impact that this additional format for presenting information may have on investors and registrants, rather than by the disclosure of new underlying informational content that investors could not already access or that would require registrants to collect significant new data. The economic benefits and costs of the final rules, including impacts on efficiency, competition and capital formation, are discussed below. We also discuss the relative benefits and costs of significant, reasonable alternatives to the implementation choices reflected in the final rules.

1. Introduction

As discussed in the Proposing Release, compensating executive officers with pay that varies with registrant performance may encourage executive officers, through financial incentives, to exert effort and make decisions that create shareholder value. However, there are also potential negative consequences of such compensation plans. For example, some such plans may cause executives to focus overly on short-term performance to the detriment of long-term performance, or may make some executives less likely to take on risky but (from a typical shareholder's perspective) valuable projects if they are unwilling to take the chance that the project could fail and result in lower compensation than would result from less risky projects.

An optimal compensation policy is generally considered to be one that maximizes shareholder value in the long term by balancing the need to provide executives with the incentive to perform well against the monetary costs and potential detrimental effects of the compensation policy. What constitutes an optimal compensation policy, including which performance metrics should be considered and how much compensation should vary with these metrics, is difficult to ascertain and will vary with a registrant's individual circumstances. Academic research remains mixed as to whether prevailing compensation structures are optimal, are too closely linked to company performance, or should be more sensitive to company performance. Thus, it is unclear whether changes that would more closely link executive pay with registrant performance than current compensation structures would have a positive, a negative, or no impact on shareholder value creation.

In addition to uncertainties about the optimality of pay-versus-performance alignment, there are challenges in measuring such alignment. For example, the available performance statistics may not adequately measure a given executive's contribution to a registrant's performance, such as when registrant performance is strongly related to market moves, sector opportunities, commodity prices, or other factors unrelated to managerial effort or skill. Even if the performance measure were not subject to such concerns, it could be difficult to match registrant performance with the associated executive actions and, perhaps, related compensation because of timing differences. For example, an executive may be rewarded with extra compensation for an accomplishment in the year it is made, even though a registrant's expected profits related to this executive performance (such as an investment or restructuring decision) might not follow until several years later. Similarly, a registrant's stock price may rise at the announcement of a new PEO who is expected to add significant value to the registrant, even though he or she may not commence employment and begin receiving compensation until the following year. The alignment of an executive's financial incentives with registrant performance can also be difficult to evaluate without also considering holdings of vested equity which link an executive's wealth accumulation to the performance of the company whether or not they were obtained as compensation. Such issues may lead to concerns with any standardized approach to presenting the relationship between pay and performance.

Despite the uncertainty and challenges involved in evaluating the relation of pay with performance, pay-versus-performance alignment is likely important to investors. In fact, academic research concludes that the incentives created for executives through the linkage of their pay with registrant performance outcomes may be the most value-relevant feature of current executive compensation plans, beyond even the level of executive pay. Accordingly, investors may consider the optimality of pay-versus-performance alignment as part of their evaluation of executive compensation packages when making voting decisions relating to the compensation of the NEOs and the election of directors, as well as when making investment decisions.

2. Benefits

For the most part, the final rules require a different presentation of certain existing information rather than the disclosure of new underlying informational content. The primary benefits of the final rules relative to the baseline will therefore depend on the extent to which the computations provided or the format used for the required disclosure makes it easier or less costly for investors to evaluate how executive compensation relates to registrant performance.

As discussed above, investors currently have access to detailed information disclosed by registrants with respect to executive compensation and registrant financial performance, but some investors have expressed dissatisfaction with existing disclosures. Data from the currently required, standardized tables and accompanying information may require further computation and analysis before investors can evaluate actual historical pay-versus-performance alignment under the baseline. Also, voluntary disclosures that provide more direct measures of the historical pay-versus-performance relationship are provided by a minority of registrants and lack standardization and comparability, as discussed in the Baseline section above. The more standardized quantitative analyses of pay-versus-performance alignment provided by the major proxy advisory firms to their clients, as well as the analyses undertaken by certain large institutional investors on their own, demonstrate shareholder demand for additional computations regarding this relationship, beyond existing disclosures.

Investors may therefore benefit from the final rules to the extent that the new presentation of data required by these final rules lowers their burden of analysis in evaluating the executive compensation policies of the affected registrants. If the repackaging of some of the information from existing disclosures into the required pay-versus-performance disclosure, and the Inline XBRL tagging of this disclosure, allows investors to more quickly or easily process the information accurately, the final rules may generate productive efficiencies by preventing duplicative analytical effort by investors. If the disclosure helps investors process and understand compensation data faster, this information may also be more quickly incorporated in market prices, marginally increasing the informational efficiency of markets.

The final rules should make it much easier for an investor reviewing a proxy statement to relate registrant performance with concurrent changes in the value of compensation, because the amount disclosed as executive compensation actually paid will more closely track these changes than currently required compensation disclosure. Further, for a number of reasons, the disclosure required under the final rules is expected to be significantly more comparable across registrants and across time than existing required disclosures in the CD&A regarding how pay relates to performance as well as current voluntary pay-versus-performance disclosures. This enhanced comparability will likely enable more efficient processing of the information. For example, the consistent tabular format will likely make the information easier to find, and standardization of the measures of pay, TSR, and net income will allow investors to understand what these measures represent without having to examine varying definitions used by different registrants. In addition, prescribing particular measures of pay and performance reduces the ability of registrants to only include measures that lead to more favorable pay-versus-performance disclosures, which, in turn, would reduce their utility and comparability. The specific definition of executive compensation actually paid under the final rules also enhances the comparability of the disclosures, as discussed in more detail below, as it treats similar economic situations relatively consistently, allowing investors to more easily evaluate the disclosure in the context of the disclosure of other registrants.

Some commenters agreed that such disclosures may reduce the time, effort, and/or cost required to review proxy statements, with several noting that the proposed disclosure could be used by investors to more easily review disclosures to identify which registrants' compensation arrangements they should investigate in greater detail. Also, many commenters supported the importance of the consistency and comparability of the disclosures.

On the other hand, a number of commenters indicated that meaningful comparability of pay-versus-performance disclosure is not feasible or not desirable given, for example, the degree of variation in the circumstances of registrants and the vast, differing array of considerations that go into their compensation programs. We acknowledge that perfect comparability may be impossible to achieve, and that some registrants may choose to supplement the required disclosures to better communicate their specific situation. However, compensation and performance, and their alignment, also cannot be properly evaluated in a vacuum. Broader economic conditions and the labor market for executive talent have significant effects on the appropriate level and performance-sensitivity of pay. Pay-versus-performance disclosures that can be compared across registrants should facilitate investors' consideration of these factors. Registrants already have substantial flexibility to provide tailored disclosures in proxy statements with respect to the relation of pay with performance. However, as discussed above, many investors are obtaining standardized third-party analyses of pay-versus-performance across different registrants, or constructing their own, which demonstrates demand for more consistent, comparable disclosure.

Some commenters indicated that, whether or not comparability is desirable, the proposed amendments would not actually provide disclosures that could be compared across registrants. These commenters stated that the proposed disclosure would not be comparable because, for example, equity granting and vesting practices vary across registrants, valuation assumptions may vary across registrants, and there is no single way to uniformly measure performance across different registrants. We expect that the revised definition of executive compensation actually paid will increase the comparability of this measure across registrants with different compensation structures. In particular, for outstanding equity awards between their grant and vesting date, the change in value reported as part of this measure for a particular year is equal to the change in fair value during that particular year, and therefore may be associated with performance during the same year. This is true regardless of the grant and vesting patterns, such that similar economic exposure for executives across different registrants should be reflected more similarly than under the proposed amendments, even when the formal structure differs.

With respect to the concern about the lack of comparability of performance measures, several commenters agreed with our view that, despite certain concerns discussed below, TSR is the most comparable financial performance measure available. Given that TSR is nonetheless an imperfect measure, the inclusion of peer group TSR, net income, and at least one Company-Selected Measure may provide useful context for investors when comparing the disclosed performance across registrants. Finally, with respect to the concern about varying valuation assumptions, the disclosure of equity award valuation assumptions when they differ materially from the disclosures of assumptions as of the grant date may help investors to identify if a particular registrant's approach to these assumptions appears to be an outlier. Overall, as noted above, perfect comparability is difficult to achieve. However, the final rules are intended to provide some basic standardized elements that can be more easily reviewed and compared across registrants. At the same time, they also include more tailored elements that may better reflect registrants' individual circumstances, such as additional registrant-specific context, significant latitude in how registrants describe the relationships between the measures in the prescribed table, and the option of supplemental disclosures in case, in the registrant's view, additional detail or clarifications would be helpful.

The overall size of the potential benefit to investors depends on the extent to which the required disclosure approximates or contributes to any of the calculations and analyses that investors would choose to perform in order to process the existing disclosures. That is, the benefits of consistency and comparability will apply only to the extent that investors find the prescribed measures to be useful. While the specific extent of benefits is difficult to ascertain, commenters as well as our observations of current analyses by or on behalf of investors provide support that the disclosures are likely to be useful to investors.

For example, the new measure of executive compensation actually paid will reflect new required computations (based on information in existing disclosures) that may be particularly relevant in the context of evaluating the relationship of pay with performance. These computations may make information of interest to investors more readily available than it is under the baseline. Commenters indicating that investors would find the proposed measure of executive compensation actually paid to be useful generally cited potential benefits discussed in the Proposing Release, such as the fact that this measure would reflect the change in value of equity awards based on performance outcomes after they are granted, that it would focus on economic exposure due to compensation committee intent and not executives' personal investment decisions, that it would reflect all elements of compensation for completeness and comparability, and that it would eliminate noise caused by the revaluation of pension benefits earned in prior periods. The revised definition of executive compensation actually paid preserves all of these features, while also mitigating concerns raised by a large number of commenters about a likely timing mismatch between the proposed measure of pay and the associated performance. By requiring the revaluation of equity awards every year, the revised measure significantly improves the degree of matching between the period to which a change in pay is ascribed and the period of the associated performance, which should make the measure substantially more useful for investors.

The revised measure is also very similar to the concept of realizable pay, discussed above. A number of commenters indicated that a realizable pay measure would be particularly appropriate for evaluating the alignment of pay and performance. While definitions of realizable pay vary, they reflect, like executive compensation actually paid, an attempt to measure the change in value of an executive's pay package—including outstanding awards that have not yet been realized—after the grant date, as performance outcomes are experienced. We believe that the increasing consideration of realizable pay (as computed by third parties) by investors when evaluating pay and performance alignment is evidence that a measure with similar features, such as the adopted measure of executive compensation actually paid, is likely to be useful to investors in this context.

Although investors could estimate executive compensation actually paid using existing disclosures, and may already be making similar estimates on their own or relying on third party estimates of related measures, they may benefit from these computations becoming readily available in the prescribed compensation measure. The newly disclosed computations could reduce duplicative analytical effort by replacing or validating related investor or third party estimates. In addition, some investors or third parties hired by investors may be interested in leveraging the disclosures to more easily compute slightly different pay measures, whether these are the measures they currently use under the baseline or refined versions of these measures that are more feasible to construct due to the availability of the new disclosures, or in using parts of the required computations for other purposes. In such cases they are likely to benefit from the required footnote disclosure of the adjustments made to compute executive compensation actually paid and the disclosure of equity valuation assumptions, if materially different from the grant date assumptions. Also, requiring that the disclosure be provided in a structured data language may benefit investors interested in extracting and analyzing some or all of the data in the disclosure across a large number of filings.

With respect to the performance information required in the new disclosures, as discussed above, there are challenges associated with measuring an executive's contribution to registrant performance that may lead to concerns with any performance measure. Commenters expressed a number of concerns with the use of TSR in particular in evaluating executive performance, such as its sensitivity to external factors outside of the control of executives, a possible emphasis on short-term performance, and the possibility of strategies that could artificially inflate TSR. However, we are not aware of, and commenters did not identify, any standard, singular measure that would be a uniformly better alternative, and some commenters noted that TSR would be a useful measure. In particular, commenters that indicated that investors would find TSR to be useful noted that it is the ultimate measure of corporate success and shareholder value creation and it is widely comparable across registrants. We agree with these commenters that, despite its limitations, TSR is likely to be a useful measure in this context, particularly because it incorporates information about a variety of facets of registrant performance, including market expectations of the future impact of current executive actions, and it is responsible for a significant amount of the variation in compensation outcomes experienced by executives. Specifically, academic studies indicate that changes in the value of equity awards after the grant date, with the movement of stock prices, are the primary channel though which pay is linked to registrant performance. TSR is mechanically a significant determinant of executive pay outcomes, as it is the most commonly used metric in long-term incentive plans, and, more importantly, a majority of CEO compensation is awarded in the form of equity awards, whose value is closely tied to stock prices even when TSR is not explicitly used as a performance metric. Current market practices provide further evidence that TSR is likely to be useful to investors in this context: every investor and third-party analysis of pay-for-performance that we have observed incorporates TSR as a primary performance measure.

However, even if TSR, despite the limitations noted above, is a particularly useful measure for the purpose of evaluating the relation of pay with registrant performance, it may not provide a complete picture of registrant performance. Further, relying solely on TSR to evaluate registrant and executive performance may even be misleading in certain situations, such as when expected outperformance is already reflected in the starting stock price, when a stock is thinly traded, or when market dynamics cause stock returns to become particularly disconnected from fundamental performance. The required disclosure of additional financial performance measures may help to address these concerns by broadening the picture of registrant performance presented in the disclosure, providing additional detail and context that could enhance the usefulness of the disclosure by certain registrants or for certain investors.

For example, several investors commented that the inclusion of TSR of a peer group would enhance the comparability of TSR, perhaps by providing a benchmark for some of the market- or industry-wide factors that may affect performance at each registrant. Some commenters indicated that the required inclusion of a Company-Selected Measure and net income would provide a more complete picture of registrant performance. More specifically, commenters stated that a Company-Selected Measure would provide insight into the registrant's perspective and a facet of performance that is directly relevant for understanding compensation, and that net income would provide a more objective accounting benchmark that is not affected by items like non-GAAP adjustments and stock buybacks. Similarly, some commenters indicated that including a list of the most important performance measures used by the registrant to link compensation actually paid to company performance would provide useful context or a more complete view of pay-for-performance programs, and may therefore help address concerns that the pay-versus-performance disclosure could otherwise “mislead” investors. Finally, to the extent registrants include additional supplemental measures of performance, commenters indicated they generally expect investors to benefit from an even more complete picture of performance.

As discussed in the Baseline section above, all of the required performance information is generally already available in existing disclosures in annual reports or the CD&A of proxy statements. However, including this performance information in the pay-versus-performance disclosure may be useful to investors to the extent it limits the time they need to spend referring to other disclosures in order to interpret the pay-versus-performance disclosure, or prevents some investors from overlooking important context about the broader performance or pay-for-performance programs of a registrant. The required description, in graphical or narrative form, of the relationship between pay and the performance measures in the prescribed table is not anticipated to provide significant additional information beyond the contents of the table, but if it presents this information effectively, it may help investors to more easily interpret the disclosure.

If the required disclosure is useful to investors, the benefits are likely to vary across investors of different types. For example, it may be particularly beneficial to those investors who do not have access to third-party analyses, have fewer analytical resources, or are less adept at interpreting current disclosures on their own. That said, some such investors may limit their proxy statement review to items like a voluntarily-provided proxy summary section regardless of the existence of the new disclosure, in which case they are unlikely to benefit. Among investors with more resources or sophistication, some may benefit by being able to more quickly review proxy statements to determine which to investigate in more detail, and some may reduce their analytical burdens by relying on information from the new disclosure to replace, to validate, or to more easily construct the inputs for their existing analyses. To the extent third parties are able to similarly leverage information provided in the new disclosures in constructing their own quantitative analyses, they may pass on some of these benefits in the form of a lower cost or a more useful analysis to subscribing investors. On the other hand, some investors or the third parties they subscribe to may continue to independently construct their own analyses without using any elements of the new disclosure; these investors are unlikely to benefit from the disclosure. For all of the investors that would benefit from the disclosures, they are likely to benefit the most in the case of (i) registrants with particularly complex compensation plans, and where the alignment of pay and performance may therefore be difficult to assess, and (ii) registrants that do not already provide useful pay-versus-performance disclosure on a voluntary basis.

Overall, the direct benefits of the final rules hinge on the new disclosures being relatively easy to review and including the information investors are most interested in when evaluating the relation of pay with performance. Therefore, if the included measures are significantly different from those investors would collect or construct on their own in order to evaluate executive compensation, or if the disclosure is too long or complicated to review quickly, benefits to investors could be limited. Some commenters expressed such concerns, indicating that the proposed disclosures would be of minimal or no benefit to investors. However, as discussed above, there is evidence that the revised measure of executive compensation actually paid and TSR are similar to measures currently used by many investors in quantitative analyses of pay and performance alignment, which suggests that these elements of the new disclosure are likely to be at least somewhat useful to investors. It is less clear to what extent the overall effect of the additional required performance measures will be to enhance the utility of the new disclosures to investors, recognizing that the usefulness of these components may be reduced by their contribution to the overall length and complexity of the disclosures, which may make it difficult to quickly interpret the basic elements of the disclosures. Any supplemental explanations registrants include may further increase the length and complexity of the new disclosures. That said, the tabular disclosure of the underlying data will provide a degree of consistency and comparability, which can aid investors in quickly processing the information.

The final rules could also have indirect benefits if the required disclosures lead to more optimal compensation policies, perhaps as a result of increased attention on the level or structure of NEO compensation and/or registrant performance. Specifically, if, by virtue of the disclosure, NEOs become less likely to demand, or boards become less likely to approve, a compensation level or structure that is not optimal (in that, as discussed above, it does not maximize long-term shareholder value), then benefits will arise to investors and registrants. The resulting pay packages may represent either a benefit or a cost to the NEOs depending on whether or not the more optimal compensation structure, including the level of compensation as well as the risk exposure, is preferred by the executives. The final rules could also indirectly benefit investors and registrants in the form of more optimal board composition, if, by virtue of the disclosure, shareholders make more informed voting decisions.

The likelihood of such indirect effects is difficult to estimate because the ideal pay-versus-performance analysis, as well as the optimal pay structure, is uncertain and may vary by company, and because reactions to the repackaging of information are difficult to predict. As discussed above, the disclosure is intended to facilitate investors' consideration of the alignment between pay and performance when making related voting decisions. Several commenters indicated that they anticipated that the proposed amendments would therefore result in improvements in compensation and/or corporate governance. However, because the final rules do not require the disclosure of significant new underlying informational content, and given the high level of existing attention to pay practices—including increased engagement on these matters with institutional investors, and the sophisticated methods and processes that many investors and third parties have developed for evaluating pay—we believe that it is unlikely that the final rules will play a significant role in encouraging more optimal pay packages or corporate governance. We therefore believe that the final rules are likely to have no material beneficial effects on competition or capital formation.

Lastly, we note that the required pay-versus-performance disclosure will provide some incremental information relative to the underlying informational content already available to the public in other formats, but that the extent of this information is limited. For example, the valuation of equity awards such as options and performance-contingent stock involve certain assumptions and expectations, and registrants are not currently required to disclose valuation assumptions for most such awards on dates other than the grant date. Vesting-date values currently are provided for stock awards in the Stock Vested and Options Exercised Table, but the applicable fair values at times before these dates, other than the grant date, and for options at all dates other than the grant date, are not separately presented by registrants. That said, for some awards, additional assumptions are not required to compute their fair values at these other dates. Specifically, for stock awards, such as restricted stock, that only have service-based conditions, the fair value would generally simply equal the stock price at the time. For stock with performance-based conditions other than market conditions, determining the fair value would involve a reassessment of the probable outcome with respect to the performance metrics involved, but registrants are also required to reassess these probable outcomes each period for the purpose of financial statement reporting, and associated footnotes should provide insight into the registrant's evaluation to the extent the changes in estimates are material.

Computing the fair value of other awards, such as options and stock with market-based conditions, after the grant date would likely require new assumptions. Using existing disclosures, investors can themselves make estimates of the fair values of options and stock with market-based conditions at dates beyond the grant date based on the disclosed terms of these awards, and by using publicly available data to make reasonable valuation assumptions. In contrast, a fair value estimate provided directly by the registrant would reflect its discretion in choosing a valuation methodology and estimating the inputs required, such as the expected option life and the expected volatility of the stock. The grant-date valuations provided by registrants already demonstrate, to some extent, how the registrants choose to apply their discretion in the valuation process. It is unclear to what extent investors would find information about what valuation assumptions registrants would apply at later dates, which would similarly reflect registrant discretion, to represent meaningful new information beyond what is available in existing disclosures (though investors may find the computations useful regardless of whether they reflect meaningful new information).

With respect to pensions, while aggregate service costs are reported in financial statement disclosures, and pension plan terms and assumptions are disclosed in detail, registrants are not currently required to separately report the service cost, or prior service cost due to any plan amendments or initiations, that is associated with each individual NEO, so the disclosure of these costs may reveal marginal new information about actuarial assumptions specific to the estimation of service costs for these individuals, such as any embedded assumptions about future compensation levels.

Additional potential sources of new information for investors include the Company-Selected Measure and the Tabular List. As discussed above, if registrants do not already disclose the historical outcomes for their Company-Selected Measure over the past five years, the disclosure may provide new information to the extent that any required adjustments or computations required to derive the value of these measures from reported financial data may not always be straightforward for investors to replicate. Finally, both the Company-Selected Measure and the Tabular List may provide new information in the form of any insight gained based on the registrant's choice of which of the measures reported in the CD&A were deemed to be the most important with respect to the last completed fiscal year.

Overall, the extent of new underlying informational content that could be made available in the disclosures is limited, and, while some investors may find the incremental information to be useful, it is unclear to what extent it would be meaningful to investors more broadly. We therefore believe that the potential benefits of the final rules derive primarily from the manner in which the information is presented rather than the disclosure of any significant new underlying informational content. The benefits of some specific implementation choices are discussed in more detail in the Implementation Alternatives section below.

3. Costs

The primary costs of complying with the final rules reside largely with registrants and include the time and expense to make the required computations; to select the tailored components of the required disclosure; to design a format for the required descriptions and create these elements of the disclosure; to draft the footnotes and any supplementary disclosures that are deemed necessary; to apply Inline XBRL data tagging; and to ensure appropriate review, such as by management, in-house counsel, outside counsel and members of the board of directors. The costs will be mitigated by phasing in the time periods for the disclosure for both new and existing registrants, thereby limiting the computations required when first producing the disclosure, and providing scaled requirements and a phased-in tagging requirement for SRCs.

In the Proposing Release, we indicated that we believed that the costs to registrants of complying with the proposed amendments likely would be relatively low, given that the required disclosures would not require the collection of any significant new information relative to the baseline and the required additional computations would be straightforward. Some commenters agreed that the compliance costs would be relatively low and/or that the required computations would not be difficult. However, some other commenters indicated that the Proposing Release may not have fully accounted for the costs of the proposed disclosures, particularly with respect to the expense of producing new option valuations and supplemental disclosures that would be required to prevent confusion. Also, we acknowledge that the compliance costs associated with the final rules will generally be higher than those that would have been associated with the approach set forth in the Proposing Release, given the revised definition of executive compensation actually paid and the disclosures with respect to additional performance measures that were not included in the proposal. We have, accordingly, revised our burden estimates for purposes of the Paperwork Reduction Act of 1995 (“PRA”), as discussed below and in Section VI of this release. However, we believe that, given that the disclosures require the collection of minimal new information, the overall compliance costs of the final rules should be modest.

In particular, while some of the computations involved are more complex than simple arithmetic, existing models and established methodologies should aid in making the required calculations. For example, commenters indicated that the determination of pension service cost, disaggregated by executive, would require minimal effort by the actuaries who are already making the required computations to produce aggregate pension service cost for the financial statements. While there may be an incremental charge to obtain these estimates, or to make the required additional computations in the case of any plan amendments, we expect it to be low. The annual revaluation of restricted stock and performance-contingent stock should only require consideration of the prevailing stock price and any updates with respect to the probable outcome of performance conditions, which are already reassessed as of the end of each fiscal year for financial reporting purposes. Finally, the annual revaluation of options (as well as any stock with market-based conditions) can generally be accomplished by reevaluating the appropriate inputs and entering these into the existing valuation models used to calculate currently disclosed values. Several commenters indicated that this process would be tedious and generate administrative burdens, and that the appropriate models as well as inputs may need to be reconsidered when revaluing option awards beyond the grant date.

We acknowledge that the revaluation of options, which will be required more frequently under the final rules than under the proposal, will likely be the most computationally-intensive requirement of the final rules. However, a minority of registrants utilizes option awards in compensating NEOs, and we agree with several commenters who indicated that annual computations of fair value of outstanding equity awards would not be overly burdensome. Option valuation is a well-established discipline, and existing models and software, as well as reliance on third-party experts when necessary, should aid the registrants that grant options to their NEOs in making the required calculations. Further, on an ongoing basis, the value of executive compensation actually paid will only need to be computed for a single fiscal year at a time (and, given the phase-in of requirements, for three fiscal years at inception, or two fiscal years in the case of SRCs), limiting the total computations required in order to update the disclosure each year. Also, as discussed above, some investors, or third parties on behalf of investors, are currently making similar computations. While the required computations may represent a burden for registrants, they may reduce such duplicative efforts and place responsibility for the calculations in the hands of registrants, who are best positioned to produce them.

Several commenters raised concerns about the extent of supplemental disclosure that would be required to clear up “misconceptions” that could result from the required elements of the proposed disclosure. While we expect that some registrants may choose to provide supplemental disclosure, such as to clarify the required disclosure, and that producing such disclosure will be associated with further compliance costs, we believe that the revised definition of executive compensation actually paid should reduce the need for clarifying disclosures because, relative to the proposed measure of pay, it is less likely to require the reporting of pay in a different period than the associated performance.

Commenters to the Reopening Release also raised concerns about the cost to include the additional information with respect to performance measures contemplated in that release. The final rules include modifications that should limit these costs. For example, some commenters indicated that the inclusion of net income and income or loss before income tax expense would increase the length and/or cost of disclosure. The final rules require the inclusion of net income, but not income or loss before income tax expense, which should limit the size and costs of the associated disclosure. Similarly, some commenters indicated that the selection of a single Company-Selected Measure would be difficult and result in substantial additional cost to registrants, in part because of the prominence of this single measure and the resulting scrutiny required from board members and senior management, with input from outside advisors. The final rules require the inclusion of a Company-Selected Measure, but registrants will be permitted to include additional supplemental measures in the table, which may mitigate burdens in cases where it is difficult to isolate a single most important measure.

Finally, some commenters indicated that the list of the top five most important performance measures contemplated in the Reopening Release would be difficult to produce, particularly because of the difficulty in ranking such measures, and that it would increase the length and complexity of disclosure due to the additional explanations registrants might consider necessary for clarification. The final rules do not include a ranking requirement and allow a variable number (from three to seven) of the most important measures, which may make it easier for registrants to find a more natural break-point in isolating a group of the measures they consider to be most important. This additional flexibility may thereby also limit the amount of additional explanatory disclosure that registrants choose to provide.

We also note that the number of relationships that the final rules will require registrants to describe in narrative or graphical form has increased to seven, for registrants other than SRCs, from the three that would have been required per the Proposing Release. For SRCs the number has increased from two to four. In particular, a registrant must describe the relationship of each required performance measure (TSR, net income, and, for non-SRCs, the Company-Selected Measure) with the PEO's compensation actually paid as well as with the average such pay of the other NEOs, and (for non-SRCs) they must also describe the relationship of TSR to peer group TSR. We acknowledge that these additional requirements will increase compliance costs, but we expect that the descriptions can be scaled depending on their relevance to a particular registrant. For example, if TSR or net income have little correlation, or only a spurious correlation, with pay at a particular registrant, and is not a metric used in their compensation plans, a simple statement to this effect may suffice.

Overall, the expansion of the disclosures with respect to performance measures will increase the compliance costs of the final rules relative to the requirements reflected in the Proposing Release, but, as discussed above, these disclosures may provide helpful context to investors.

As discussed above, registrants will be required to file the pay-versus-performance disclosure in certain proxy or information statements. While much of the disclosure will be based on information that is otherwise disclosed, the new computations and new presentation of this underlying information, as well as the inclusion of existing measures—TSR and peer group TSR—that are otherwise “furnished” but not “filed,” may create an incremental risk of litigation under Section 18 of the Exchange Act (“Section 18”). Several commenters indicated that this may increase the cost to registrants of the disclosures, because of the need for additional assurance and because of litigation risks. However, we note that Section 18 does not provide for strict liability with respect to “filed” information.

Compliance costs associated with the final rules are likely to vary among registrants depending on the complexity of their compensation structures. For example, the computation of executive compensation actually paid from total compensation reported in the Summary Compensation Table involves adjustments to the treatment of equity awards and pension benefits. Registrants that include these elements in their executive compensation plans are therefore expected to require more computations to produce the disclosure. This is particularly the case for registrants that use options, both because the required computations are more involved, as discussed above, and also because options tend to vest ratably over time, so registrants may need to track and value many different tranches of options in a given year. As shown in Tables 2 and 3 in the Baseline section above, the use of both options and pensions has declined since the time of the Proposing Release, but each still has a prevalence of roughly 20 percent among S&P 1500 CEOs (and 30 percent among their other NEOs). Overall, though, the registrants for whom the computations will be more burdensome—those with more complex compensation packages—are also generally those for which investors are expected to benefit most from the disclosure: in the absence of the disclosure, it is more difficult for investors to assess the alignment of pay and performance when compensation is more complex.

Large companies are more likely than smaller ones to have pension plans and grant stock and option awards to executives. However, a significant fraction of mid-sized and smaller companies feature these components in their compensation plans as well. Thus, while the compliance costs are likely to be relatively low, these costs may be slightly more burdensome for those affected registrants that have complex compensation packages and yet are small enough that the costs of the disclosure are relatively more consequential in comparison to their size. That said, SRCs will be subject to scaled requirements consistent with their existing disclosure requirements, including fewer years of disclosure; no requirement to report peer group performance, a Company-Selected Measure, or a list of the most important performance measures; and the exclusion of items related to pension plans in computing executive compensation actually paid. SRCs are not currently required to comply with Item 201(e) of Regulation S-K, so they may face a small incremental burden of computing their own TSR for the purpose of this disclosure as compared to other affected registrants.

Based on analysis for purposes of the PRA, as discussed in Section VI of this release, we estimate that the total incremental burden on all registrants of the final rules will be, annually, approximately 95,800 hours for internal company time, and about $12.8 million for the services of outside professionals. These estimates represent an increase in estimated burden hours per affected registrant of about 87 percent (from 15 to 28 hours) for non-SRCs, and about 13 percent (from 15 to 17 hours) for SRCs, relative to the estimates in the Proposing Release. As discussed above, these costs are expected to vary across registrants depending on the complexity of their compensation structures. Also, certain registrants—such as those whose executive compensation is not tied closely to TSR or net income—may be more likely to voluntarily supplement the disclosure with additional measures, explanations, or analyses in order to explain the patterns in the required disclosure, and may thus face higher overall costs. However, we do not believe that any of the variation in the compliance burden will be large enough to have a material detrimental effect on competition or capital formation.

While the new disclosure requirements are intended to make it easier for investors to assess the alignment of pay and performance, investors may instead bear increased information processing costs as a consequence of the final rules if they increase the length and complexity of existing disclosures without significantly adding to the ease of interpretation. Some commenters raised concerns that the proposed disclosures would result in such information overload. The likelihood and extent of such costs resulting from the final rules may be a function of the degree of supplementary disclosures registrants choose to provide, as well as the complexity of and variation in presentation formats. The risk of information overload may also be exacerbated by the required disclosures with respect to additional performance measures, which could provide helpful context for investors, or could end up complicating or obscuring the elements of the disclosure that would be most useful to investors. If the required disclosures complicate rather than facilitate the task of understanding executive pay policies, they may marginally decrease the informational efficiency of markets.

The final rules could confuse investors about the optimality of pay practices if they bring attention to a particular relationship that might not be relevant, given the facts and circumstances of a particular registrant, in evaluating the alignment of pay and performance at that particular registrant. As discussed above, there are challenges in measuring pay-versus-performance alignment which are likely to impact any standardized approach to presenting this relationship. However, the required inclusion of additional context in the disclosure may help to mitigate potential confusion. For example, the inclusion of net income, a Company-Selected Measure, and a Tabular List could be helpful in limiting confusion stemming from differences in the timing of an executive's accomplishments and when they may be reflected in TSR, to the extent that other performance measures may better align with executive performance in such cases. Further, including peer group TSR in the disclosure may help investors to identify when registrant TSR could be driven by market moves, sector opportunities, commodity prices, or other factors unrelated to managerial effort or skill. That said, the required disclosure may be less meaningful at a particular registrant if TSR, even relative to peers, is very different from the contribution of the given NEO to performance, or if the disclosed relationship between compensation and TSR does not ( e.g., because of vested equity holdings that are not reflected in executive compensation actually paid) fully capture the economic relationship between the company's performance and the financial rewards to the NEO. Similarly, the required net income disclosure may be less meaningful at registrants at which net income is not particularly relevant to understanding executive performance.

As discussed in the Proposing Release, the potential for confusion is especially concerning given that the new disclosure may be of particular interest to less sophisticated investors, who may be less likely to have access to third-party pay-versus-performance analyses or may be less adept at conducting their own such analyses. The possibility of confusion is mitigated by allowing registrants to provide supplemental measures of pay and performance, as well as the ability of registrants to provide further explanatory disclosures. Some commenters agreed that this flexibility to supplement the disclosure would improve investors' understanding or mitigate potential confusion. However, such clarifying disclosures may be more likely to be provided when the disclosure is perceived by the registrant to incorrectly indicate the misalignment of pay and performance than when the disclosure is perceived to incorrectly indicate strong alignment. Further, as noted by other commenters, less sophisticated investors may be unlikely to consider these supplemental disclosures. While some commenters were not convinced that a Company-Selected Measure or list of most important performance measures would help in such cases, it is possible that these additional required elements of the disclosure may help mitigate confusion by providing a mandatory, prominent indicator of the broader performance landscape in the specific context of a given registrant.

The final rules could also lead to indirect costs if the required disclosures lead to changes in compensation packages that are not beneficial. Registrants may make changes to avoid disclosure that they perceive indicates the misalignment of pay and performance, whether that indication is valid or merely due to limitations of the standardized approach. For example, by virtue of the disclosure, boards may become more likely to approve compensation structures that more strongly link pay to stock price performance, even in situations in which this would not be optimal. The inclusion of net income in the disclosure could mitigate this risk, or could instead encourage the use of net income as a performance metric in incentive programs, even when this is not beneficial. Commenters raised concerns that such pressures on compensation design could lead to compensation that incentivizes short-termism and/or the inappropriate homogenization of compensation plans. If such changes are indirectly encouraged by the final rules, they may entail costs to registrants and their shareholders. As in the case of any shifts towards more optimal compensation structures, discussed in the Benefits section above, the resulting pay packages may represent either a benefit or a cost to the NEOs themselves depending on whether or not the less optimal compensation structure, including the level of compensation as well as the risk exposure, is preferred by the executives.

As in the case of the potential benefits outlined above, many of these costs are difficult to quantify because the ideal pay-versus-performance analysis for investors, as well as the optimal pay structure, is uncertain and may vary by company and because reactions to the repackaging of information are difficult to predict. Still, because the final rules do not require the disclosure of significant new information, and given the high level of existing attention to pay practices—including the increased engagement on these matters with institutional investors, and the sophisticated methods and processes that many investors and third parties have developed for evaluating pay—we believe that it is unlikely that the final rules will play a significant role in encouraging sub-optimal pay practices. We therefore believe that the final rules likely will have no material detrimental effects on competition or capital formation.

The costs of some specific implementation choices are discussed in more detail in the Implementation Alternatives section below.

4. Implementation Alternatives

In this section, we present significant implementation alternatives and a discussion of their benefits and costs relative to the implementation choices in the final rules.

i. Registrants and Filings Subject to the Disclosure Requirement

An alternative to the final rules would be to fully exempt SRCs from the disclosure requirement. Exempting SRCs generally would be consistent with the overall scaled disclosure requirements that apply to SRCs. While the final rules subject SRCs to scaled requirements in order to limit the incremental burdens such companies may face relative to other registrants, some such burdens remain. For example, SRCs are currently not required to disclose their TSR in annual reports, so they would face a higher burden than other registrants to calculate and include this measure in the pay-versus-performance disclosure. SRC pay-versus-performance disclosure, under the final rules, may also benefit investors to a lesser degree than that for other registrants, because the scaled requirements reduce the content and comparability of the disclosures. Also, in the absence of CD&A disclosure, investors will have less information with which to interpret pay-versus-performance disclosures from these registrants. As discussed above, some commenters agreed that SRC pay-versus-performance disclosure would generate greater burdens and/or lesser benefits than that for other registrants.

On the other hand, it is possible that investors may particularly benefit from the required pay-versus-performance disclosure for SRCs, precisely because these registrants currently provide less extensive disclosure about compensation. For example, some investors may believe that the long-term performance of younger, high-growth companies may be highly sensitive to the design of executive compensation. Such investors may be particularly interested in compensation structures at SRCs but may find it difficult to assess these structures in the absence of CD&A disclosure for SRCs. These investors may benefit from SRC pay-versus-performance disclosures, even if these disclosures are not directly comparable with the disclosures of other affected registrants. Further, the data that SRCs do currently disclose is less likely to be available in aggregate form from data vendors that collect such data from the proxy statements of larger companies. Investors that are interested in comparing executive compensation across SRCs may particularly benefit from the data in the pay-versus-performance disclosure being tagged in Inline XBRL, to the extent this makes the data more accessible or increases the likelihood that more commercial databases expand their coverage to such registrants. Some commenters agreed that there may be particular governance concerns at SRCs and that investors would benefit from pay-versus-performance disclosures by these registrants.

The final rules permit SRCs to present fewer years of information in the disclosure; to not include peer group performance, a Company-Selected Measure, or a Tabular List; and to exclude items related to pension plans in computing executive compensation actually paid. While these scaled requirements may reduce the benefits of the disclosure, these accommodations should substantially limit the incremental burdens faced by SRCs in providing pay-versus-performance disclosure, while preserving some benefits to investors interested in executive compensation at such registrants.

Another alternative with respect to the applicability of the final rules would be to expand the filings requiring pay-versus-performance disclosure, such as requiring that such disclosure accompany any Item 402 of Regulation S-K disclosure, including in Form 10-K or Form S-1. Such an approach would make pay-versus-performance disclosures more consistently available for Section 12(g) registrants subject to the final rules and broaden the disclosure requirement to include Section 15(d) registrants other than EGCs. However, the required disclosure may be most useful to shareholders when they are deciding whether to approve the compensation of the NEOs through the say-on-pay vote, voting on the election of directors or acting on a compensation plan. The adopted approach requires pay-versus-performance disclosure in proxy statements in each of these cases. As discussed above, one commenter agreed that this approach would provide “relevant information” when it is “most useful.” Nonetheless, shareholders making voting decisions at a particular registrant may benefit from broader and more consistent availability of pay-versus-performance disclosures on an annual basis at other registrants. Specifically, these disclosures may allow shareholders to more easily compare pay practices across registrants when deciding how to vote at a particular registrant, particularly, for example, in the case of smaller companies whose peers may be more likely to be Section 12(g) or Section 15(d) registrants. Such disclosures may also be of use to some investors in making investment decisions, irrespective of any matters that are up for a vote.

However, registrants with reporting obligations only under Section 12(g) or Section 15(d) do not have securities that are registered on national securities exchanges, so the markets for their shares are likely to be comparatively less liquid. Estimates of share values and therefore of TSR for such registrants may be less precise and less readily available, potentially making pay-versus-performance comparisons based on this measure less meaningful across such registrants. Also, as in the case of SRCs, Section 15(d) registrants are not subject to Item 201(e) of Regulation S-K requirements for stock price performance disclosure. Similarly, Section 12(g) registrants may not be required to disclose Item 201(e) of Regulation S-K information in some or all years, so Section 15(d) registrants and some Section 12(g) registrants would bear an additional burden of calculating their own TSR and, except in the case of SRCs, the TSR of a peer group for this purpose. One commenter supported requiring the new pay-versus-performance disclosure in all filings that discuss compensation, but this commenter also acknowledged that shareholders would most likely only read those materials assembled for an annual meeting, which would include the new disclosure under the final rules.

ii. General Disclosure Requirements

We have considered several reasonable alternatives to the general disclosure requirements of the final rules.

Many commenters recommended a more principles-based approach that would permit registrants to determine which measures of pay and performance to disclose or how to disclose the relationship between these measures based on what they deem to be appropriate for their individual situations. Such an approach could have the potential to allow investors to more directly observe how management views the alignment of pay and performance at a given registrant, and might reduce reporting costs because registrants need only report what they believe to be appropriate given their unique circumstances. To the extent that the prescribed measures may be less meaningful at particular registrants, a principles-based approach could reduce shareholder confusion in understanding the relationship between pay and performance at a particular registrant. A principles-based approach would also reduce the risk that the disclosure requirements could lead registrants to change their compensation structures in ways that are less than optimal for the sake of achieving what they perceive to be more favorable pay-versus-performance disclosure.

On the other hand, a principles-based approach may reduce comparability of the disclosure and could increase shareholder confusion because the choice of pay and performance measures, and the disclosure time horizon, may vary significantly across registrants. Also, a principles-based approach may allow registrants to selectively choose the measures or time horizon that result in the most favorable disclosure. Several commenters indicated that scrutiny by sophisticated investors and proxy advisory firms, as well as the incentive effect of say-on-pay votes, would motivate registrants to produce effective disclosures within the flexibility of a principles-based regime. However, we note that investors continue to express discontent with existing disclosures despite these factors. The adopted approach of specifying some uniform requirements for the disclosure, requiring certain elements that will vary across registrants (the Company-Selected Measure and Tabular List), allowing registrants to choose the format for describing the relationship between different measures, and permitting the inclusion of additional measures, additional years of data, or other supplemental disclosure should promote comparability while preserving flexibility to tailor the disclosure to a registrant's individual situation. Registrants will also continue to have significant latitude in presenting additional compensation analyses, which provides further opportunity for registrants to clarify their unique circumstances and considerations in designing compensation.

Conversely, we also considered prescribing a uniform format or some minimum requirements for the descriptions of the relationships between different measures. Under the final rules, registrants may apply a wide range of formats when presenting these relationships. For example, some registrants may discuss percentage changes in the measures in narrative form while others may present the levels of the measures in graphical form. Investors' ability to easily interpret and compare the disclosure across registrants could be increased by requiring a uniform format for presenting the relationship, such as a standardized graphical presentation, or some minimum standards for the presentation format, such as a requirement that the disclosure be in the form of a graph. The cost of these more prescriptive approaches would be the restrictions on the ability of registrants to tailor the format of the required disclosures to best reflect their individual circumstances, which may vary significantly. For example, with a prescribed format, registrants might not be able to scale a required description to reflect the relevance of a particular measure at that particular registrant, which could result in lengthy disclosure about relationships that are not meaningful. Under the final rules, the tabular disclosure of the annual values of the required compensation and performance measures should facilitate comparisons of the underlying content of the disclosures across registrants regardless of the format for the required descriptions. It is also possible that these descriptions could become more comparable as registrants gain experience with the requirements; as one commenter predicted, “[o]ver time best practices will emerge, and investors will encourage companies to follow those best practices.”

We also considered alternatives with respect to the extent of the required descriptions. As discussed above, the final rules require, for non-SRCs, the description of seven different relationships (and four in the case of SRCs) in graphical or narrative format. An alternative would be to not require the description of some of these relationships, such as that between net income and executive compensation actually paid of the PEO or the other NEOs. Such an approach could help to mitigate commenter concerns about the costs and length of the required disclosure, given that the description of a specific relationship might require the application of significant discretion and involve more space in the proxy statement than a particular column in the required table. Reducing the number of mandated descriptions may reduce the extent of disclosure in cases where the measures in question may not be relevant in the context of a particular registrant. A more focused set of required descriptions could reduce compliance costs and make it easier for investors to more quickly review the disclosures. The underlying measures would still be available in tabular form for investors to consider; for example, investors might refer to net income as a benchmark to gauge the adjustments in a non-GAAP profitability measure presented as a Company-Selected Measure. However, investors may benefit from understanding the registrant's perspective on each performance measure, and, as discussed above, we expect that the descriptions can be scaled depending on their relevance to a particular registrant.

We also considered alternative approaches to presenting the pay and performance data. For example, several commenters suggested that, instead of requiring the presentation of year-by-year data, we could require registrants to aggregate pay over a three to five year horizon and compute the cumulative TSR over a similar horizon, and then either present a single pair of statistics or a set of rolling values of these multi-year statistics. As noted by these commenters, such an approach could help to smooth any lumpiness in pay (such as when certain awards or payments are not made every year) or short-term volatility in the performance measure. However, it would also make it harder to discern how pay has been associated with year-by-year changes in performance. Further, for investors preferring this approach, a form of aggregate analysis should be relatively straightforward to construct from the disclosure required under the final rules, by adding the values of executive compensation actually paid over multiple years and comparing this to the cumulative TSR over that horizon. In contrast, presenting aggregate statistics would not reduce compliance costs over time because new computations for the latest fiscal year would still be required each year that the disclosure is produced.

Other commenters suggested that we require registrants to isolate pay granted in a particular year and provide an updated valuation of that pay, for each grant year in the time horizon of the disclosure, at the end of the latest fiscal year (or possibly at vesting), and relate those updated values to cumulative performance. Such a focus on the pay granted in a particular year, and how its value has changed, may provide insight specific to the compensation decisions by the board in each year. However, given that grants have overlapping performance periods, it may be difficult under this approach to judge the overall association of pay with performance, and the relationship between the performance in a particular period and all of the associated pay.

We also considered alternatives with respect to the required structuring of the disclosures. Alternatives to the adopted approach include not requiring that the underlying data disclosed in tabular form be provided using a structured data language ( i.e., tagged in Inline XBRL), requiring more or less of the information to be tagged, or requiring a different structured data language. Not requiring that the disclosure be provided in a structured data language would reduce the costs of compliance. Some commenters indicated that the tagging requirements would increase the costs and time to produce the disclosure or delay the filing process. The affected registrants are familiar with Inline XBRL because they are required to provide information in other filings in this data language, but the exact specifications differ and, with limited exception, they are not required to provide any structured data in proxy or information statements. The Inline XBRL requirements would impose additional burdens on registrants, beyond what they currently spend on producing structured data for other purposes, because their contracts with outside data tagging vendors and/or the responsibilities of their in-house staff that works on data tagging would have to be expanded to include the new tagging requirement. In addition, a few commenters anticipated some difficulties because staff preparing proxy statements would be unfamiliar with Inline XBRL. One commenter stated the cost of XBRL tagging can be up to tens of thousands of dollars. A few commenters remarked that the costs of XBRL tagging outweigh the benefit to investors, and questioned whether there was sufficient evidence that such structured data was being used by, or would benefit, investors.

Since the time of the Proposing Release, the market has had significantly more experience with structured data languages, including XBRL. We expect that this experience, along with the adoption of Inline XBRL, will reduce the costs of implementing the requirements and enhance the quality of the data made available. While costs will remain, the Inline XBRL requirements should facilitate the extraction of the tagged data across large numbers of filings. These requirements may therefore benefit investors interested in analyzing and comparing the information in the disclosure across large numbers of registrants or, eventually, a large number of years. The tagging of compensation information under the final rules may be particularly beneficial to investors, in that several widely-used commercial databases collect compensation data only for large companies. Some commenters agreed that tagging the disclosures would enhance the benefits to investors, by increasing the efficiency with which large amounts of data could be filtered and analyzed, by enhancing the ability of investors to compare the data across companies or over time, and by allowing investors to obtain this data efficiently or at lower cost. There is also increased evidence that structured data is used by investors and generates benefits. For example, one study found that XBRL has helped to reduce the informational advantage of large institutions over small ones, in that small institutions' trading responsiveness to Form 10-K information and stock-picking skills improved relative to large institutions after the adoption of XBRL. Other studies provide evidence consistent with XBRL tagging of financial statement disclosures leading to an increase in stock price informativeness ( i.e., the extent to which market prices reflect company-specific information).

We considered not requiring some or all of the block tagging that the final rules will require, such as: the graphical or narrative disclosure that would follow the tabular disclosure; the disclosure of deductions and additions used to determine executive compensation actually paid; and the disclosure regarding vesting date valuation assumptions. While the nature and potential variation in format of these disclosures may make them less suitable for large-scale analysis than the numerical data in the main table, the incremental costs of tagging these disclosures as block-text should be low and such tagging could benefit investors interested in extracting these parts of the disclosure from a large number of filings We also considered, as proposed, not requiring that each numerical item in the deductions and additions used to determine executive compensation actually paid and the vesting date valuation assumptions be tagged separately. While such tagging will require incremental compliance costs, it may benefit investors interested in using this data, such as for constructing alternate pay measures.

We also considered requiring registrants to provide the data an XML-based data language specific to the pay-versus-performance disclosures (“custom XML”) rather than Inline XBRL. As discussed in the Proposing Release, a custom XML requirement could increase the ease of implementation of the structured formatting requirement for the main table, and could thus reduce costs of structuring, particularly for smaller registrants. However, the Commission's custom XML data languages are generally unsuitable for tagging large blocks of information or implementing detail tags within such blocks, and are therefore not as appropriate for implementing the requirements of the final rules.

iii. Compensation Measures

We have considered several alternative approaches to the compensation measures to be included in the disclosure, particularly with respect to the definition of executive compensation actually paid. The final rules define this compensation measure generally in line with the approach described as “incremental compensation earned” in the discussion of implementation alternatives in the Proposing Release. We also considered adopting definitions that would treat equity awards and pensions differently, such as in the proposed definition, or that would include different elements of compensation.

With respect to equity awards, the proposed approach would have required registrants to include the fair value of stock and option awards in executive compensation actually paid at the time of vesting. As discussed in more detail above, some commenters agreed with arguments in the Proposing Release that certain features of this approach, such as the fact that it would reflect the change in value of equity awards based on performance outcomes after they are granted, would be beneficial for this purpose. However, many commenters raised concerns that the proposed definition would generate a mismatch between the period in which pay was reported and the period of the associated performance, and that this would significantly reduce the potential usefulness of the disclosure.

Specifically, as discussed in the Proposing Release, under the proposed definition of executive compensation actually paid, the measure may be subject to volatility based not on performance but on the vesting pattern of equity awards, because it includes, in the year of vesting, the original grant-date value and all gains (or losses) related to returns in all years since the grant was made. A number of commenters highlighted concerns of this nature. Similar issues that commenters noted include an exacerbation of the misalignment when the size of an award is intended to recognize performance in the year of grant (or prior); when awards formally vest in a different year than the end of the performance period, or when the vesting date of an award is distant from the end of the year. Commenters also noted that the timing mismatch would not apply equally to different types of compensation or across different vesting patterns, leading to difficulties in comparisons across registrants or executives. Consider, for example, a fiscal year in which one PEO receives a $1 million cash bonus and another instead receives a $1 million restricted stock award that vests after one year. Under the definition that was proposed, executive compensation actually paid would have been $1 million and zero, respectively, for the two PEOs in that fiscal year.

As discussed above, the treatment of equity awards in the adopted measure of executive compensation actually paid is expected to preserve the benefits noted by commenters of the proposed approach while substantially reducing the risk of a timing mismatch. Under the adopted approach, the total value reflected in executive compensation actually paid for a given award, when summed across years, will be equivalent by the time of vesting to that which would have been included at vesting under the proposed approach. However, by attributing the change in an equity award's fair value in a given year—which would reflect performance in that same year—to that individual year, rather than ascribing the full value to the vesting date, the revised measure should better align pay with the associated performance.

This improved alignment will limit the volatility associated with vesting patterns, by distributing pay over the full vesting period, as it is earned. It will also reduce the sensitivity to small differences in formal vesting dates, by associating amounts of pay with particular years based on the changes in value attributable to those years rather than solely based on where the vesting date happens to fall. Attributing some of the value of equity awards to the grant year addresses the possibility that the size of awards may be designed to reward grant year performance. The revised approach also improves comparability; for example, the two PEOs discussed above, who receive a $1 million cash bonus and a $1 million restricted stock award, will both be considered to receive $1 million of compensation actually paid in that year, while any change in the value of the second executive's stock until vesting would also be reflected in future years. Overall, the enhanced alignment resulting from the revised definition is expected to make it easier for investors to understand the relationship between pay and performance, though this comes at the cost of increased compliance costs for registrants.

In valuing option awards in executive compensation actually paid, a number of commenters recommended that we use intrinsic values ( i.e., the “in-the-moneyness,” or the amount that would be gained upon immediate exercise) instead of fair values. Those commenters indicated that intrinsic values would be easier and less burdensome to calculate or would more appropriately reflect compensation rather than the effect of an executive's investment decisions. We acknowledge that fair values are more burdensome to compute than intrinsic values. However, intrinsic values can severely understate the values of options. The fair value of an option provides a more accurate picture of the total value of the asset being transferred, which includes both the current intrinsic value and the ongoing time value of the option: the ability to potentially capture additional upside while not taking the commensurate downside risk. By granting an option with significant remaining time to maturity after vesting, boards are consciously awarding executives with value beyond the vesting-date intrinsic value. As such, this transfer of value may reasonably be considered to be compensation. While an executive might not wait until maturity to exercise an option, the fair value calculation should generally incorporate an assumption regarding typical exercise behavior. Whether the executive chooses to exercise earlier or later than is typical (and therefore expected by the board) can reasonably be considered an investment decision.

Some commenters also suggested that we consider valuing equity awards as of alternate dates, such as the grant date or, for options, the exercise date. Valuations as of these alternative dates may be less burdensome to calculate, as grant date fair values are already included in the Summary Compensation Table and the amount realized on exercise of options is already included in the Stock Vested and Options Exercised Table. However, grant date valuations would not reflect the performance sensitivity of unvested equity awards. As discussed above, because the empirical relationship between pay and performance is driven by changes in the value of executive stock and option holdings, considering only grant-date values may ignore one of the primary channels for relating pay and performance. Exercise date valuations, in turn, reflect the effect of performance after the grant date, but also reflect the executive's decision of when to exercise awards, which may reasonably be considered an investment decision rather than a compensation decision. For example, as one commenter noted, “executives who hold their options to the full term before exercise may be unjustifiably seen as being overpaid compared to executives who exercise their options quickly.”

With respect to pensions, the final rules require that executive compensation actually paid include the pension service cost for the year as well as the prior service cost (or credit) due to any plan amendments or initiations in the year, rather than just the pension service cost, as proposed. Some commenters alternatively suggested that we include the present value of pension benefits that were earned in the last fiscal year, or, similarly, the change in present value of accumulated pension benefits while holding the beginning and ending valuation assumptions constant. All of these approaches—including what is being adopted, what was proposed, and the commenters' suggestions—should reduce the volatility in reported pay caused solely by changes in assumptions relative to the pension component of the Summary Compensation Table, because the latter includes the change in value of all previously accumulated benefits with changes in interest rates and other actuarial assumptions. Thus, any of these approaches should make it easier for investors to evaluate the relationship of pay with performance. We considered, as an alternative to the adopted approach, including only pension service cost (as proposed) or the present value of pension benefits that were earned in the last fiscal year (as suggested by, or similar to what was suggested by, various commenters).

Pension benefits may be a function of compensation levels, as in the case of pay-related, final-pay, final-average-pay, or career-average-pay plans. They are also a function of the terms of the plan. Service costs are based on estimates of future benefits that assume plan terms remain fixed and that may already incorporate projections about future compensation levels. Service costs are also smoothed over time relative to how the future benefits are actually earned or change over time. As a result, the effect of plan amendments and actual changes in current compensation levels on the value included for pensions under the proposed approach may be dampened. For example, if a plan were amended, current and future service costs would be adjusted upwards, but there would be no corresponding adjustment for service costs reported for previous years. The adopted approach would more fully reflect the effect of any plan amendments by including a catch-up adjustment for the impact on service costs reported in previous years.

The adopted approach does not fully account for changes in actual compensation levels from the estimated compensation levels used to estimate service cost. Because actual changes in current compensation may be related to performance, and these changes in compensation may be magnified by pension benefits that are a function of compensation levels, the alternative approach of including the present value of pension benefits earned in a given year may be more useful in evaluating the relationship between pay and performance. This alternative approach would fully reflect plan amendments as well as unexpected increases in pay, whose impact on pension benefits may reflect an important source of increased compensation. Under this alternative, registrants may be able to make the required computations based on the information already available to them, rather than through their actuarial services provider, which could marginally reduce compliance costs. Such an approach may also further increase the comparability between compensation provided through defined benefit and defined contribution plans, because registrant contributions to defined benefit plans may also be directly related to current compensation levels or other such metrics with respect to the last fiscal year. However, the amount included with respect to pensions under this alternative would not have as direct of a relationship with the values included in the audited GAAP financial statements as the service cost (and prior service cost or credit) included under the adopted approach.

Some commenters suggested excluding components of pay that may be considered unrelated to performance—such as perquisites and values related to retirement benefits—from the definition of executive compensation actually paid. As discussed in the Proposing Release, restricting the definition of executive compensation actually paid in such a way would not provide investors with a complete picture of compensation and how it relates to financial performance. While compensation committees may rely mainly on particular components of compensation in order to provide performance incentives, the other components of compensation may still vary with company performance and, even if they do not vary with performance, may be important to consider in order to understand how sensitive the totality of compensation is to performance. Restricting the types of compensation included in executive compensation actually paid may also reduce the comparability of disclosures across registrants that rely more heavily on types of compensation that would be excluded from the prescribed measure versus those that rely more heavily on compensation types that would be included.

We also considered adjusting the definition of executive compensation actually paid to account for executives' continued exposure to registrant performance after an equity award vests, due to restrictions on the transfer or monetization of such equity, by continuing to reflect such awards in executive compensation actually paid until these other restrictions lapse. In some cases, the relationship of executives' wealth accumulation to registrant performance may be driven by their vested holdings of equity. When such holdings are mandated, the resulting exposure to registrant performance after vesting may reflect a compensation decision rather than an active investment decision by the executives, and could be helpful to consider in order to better understand the total required sensitivity of an executive's income and financial assets to the registrant's performance.

However, different sets of restrictions on the transfer or monetization of equity can have different effects on the degree of continued required exposure. For example, some non-transferable holdings could be monetized by executives through contractual agreements with a broker-dealer, if the registrant's hedging policies permit such transactions. There is therefore uncertainty as to how best to reflect such restrictions for the purpose of the new disclosure. While the adopted definition of executive compensation actually paid does not include adjustments for restrictions on the transfer or monetization of equity awards, registrants can choose to provide supplemental measures of pay if they believe that those measures better demonstrate the effects of these features.

The final rules require registrants to include the Summary Compensation Table measure of total compensation together with executive compensation actually paid in the tabular disclosure of pay and performance measures. We considered excluding this measure. Some commenters indicated that it would be extraneous or confusing in the pay-versus-performance disclosure. However, as discussed above, some current pay-for-performance analyses used by investors use grant-date measures of pay, similar to total compensation from the Summary Compensation Table. To the extent that some investors may be interested in considering the relationship of performance with a measure of pay that excludes changes in the value of equity awards, they would be able to refer to the Summary Compensation Table measure of total compensation in the tabular disclosure. Further, as discussed above, this existing total compensation measure may be a useful benchmark for understanding executive compensation actually paid, such as in the case where infrequent grants designed to provide multi-year incentives may cause sharp increases in the latter measure in the years when such grants are made.

We considered also requiring the disclosure of a measure of realizable pay, a type of measure that a number of commenters indicated may be useful in this context. The adopted measure of executive compensation actually paid is quite similar conceptually to realizable pay measures, with a few key differences. For example, realizable pay is typically computed based on equity awards granted over a fixed period. This approach may make it easier to evaluate the compensation decisions made by a board over such fixed period. However, equity awards can have long vesting periods and typically have overlapping performance periods, so considering all unvested awards, regardless of when they were granted, may provide a more complete picture of pay for the purpose of evaluating its alignment with performance. Realizable pay is also typically computed over a multi-year period, with outstanding equity awards valued as of the end of the period (or sometimes at vesting or exercise, if earlier). As discussed above, such aggregated, multi-year pay measures can smooth certain outliers but can also obscure the year-to-year relationship of pay and performance. Registrants may voluntarily include measures of realized or realizable pay in the disclosure if they deem them to be helpful to explaining the relationship of their pay with performance.

Lastly, we considered also requiring the disclosure of peer group compensation. While TSR for a peer group is required to be included under the final rules, also incorporating pay information for a peer group in order to produce relative pay-versus-performance disclosures may be useful to investors as it would provide further context in which to evaluate the pay-versus-performance alignment of a registrant. However, requiring further comparisons to a peer group may reduce the comparability of disclosures because of registrant discretion in selecting the peer group or variation in the availability of a closely comparable peer group. There are also practical implementation considerations, as peer compensation for the last fiscal year is not likely to be available at the time a registrant is compiling the disclosure. Further, even if these practical considerations could be mitigated ( e.g., by permitting peer information to be excluded when unavailable), requiring relative pay-versus-performance disclosures would most likely impose higher compliance costs. Under the final rules, investors can construct relative pay-versus-performance analyses on their own by comparing the separate pay-versus-performance disclosures of each of a registrant's peers, based on the peer group reported by a registrant under Item 201(e) of Regulation S-K or in the CD&A, if such peers have filed their disclosures as of the time of comparison.

iv. Performance Measures

We have considered several reasonable alternatives with respect to the performance measures to be included in the disclosures. For example, commenters raised, and we have considered, many different approaches to computing and presenting TSR. As discussed above, common suggestions included, among others, presenting a rolling average of TSR ( i.e., for each year, registrants would report the cumulative TSR for the previous five years) or an annualized TSR ( i.e., for each year, registrants would report TSR for that single year). While a rolling average could present a broader view of performance to those taking a longer-term perspective, it could also obscure the performance specific to a given year. A five-year rolling average TSR could change from year to year because of performance in the current year being newly included in the rolling average or because of the performance six years ago being newly excluded from the rolling average. An annualized TSR would provide greater clarity and align with the revised definition of executive compensation actually paid, which will reflect, in a given year, changes in the value of outstanding equity awards over that specific year. Also, according to one commenter, “most investors and proxy advisors generally look to an annualized approach when they assess a company's TSR.”

However, the adopted approach of computing cumulative TSR, and presenting it as the changing value of an initial fixed dollar investment, will be familiar to both investors and registrants because it aligns with the Item 201(e) of Regulation S-K performance graph requirement. We also expect this approach will make the trend in performance easier to understand for less sophisticated investors, given concerns about financial literacy among investors and, particularly, a common difficulty in appropriately combining percentage changes ( e.g., recognizing that a negative 50 percent return followed by a positive 50 percent return represents a negative 25 percent return on a cumulative basis). A cumulative return, scaled to a fixed investment, will still make the return attributable to a given year apparent, and sophisticated investors can easily use this return to compute other variations of TSR that they may prefer.

We also considered not requiring any registrants, including non-SRCs, to include peer group TSR in the disclosure. As discussed above, a number of commenters had concerns about the peer group TSR requirement, including that it would be costly and yet the benefits could be limited because variation in peer group selection, and in the degree of relevance of peer group performance, could reduce comparability and mislead investors. We acknowledge that peer group TSR will not provide an equally relevant benchmark across all registrants. However, it may nonetheless provide helpful context for assessing registrant TSR by providing some indication of broader market or industry conditions, and may help to address the concerns of commenters that registrant TSR could reflect a number of factors outside of the control of the executives of the registrant. We continue to expect the costs of including peer group TSR to be limited, even if a registrant does not use the same peer group as in the Item 201(e) of Regulation S-K peer group TSR disclosure, because the required data is readily available and the required computations are relatively straightforward.

Another alternative to the final rules would be, as in the proposed rules, to not require any other prescriptive performance measures, beyond TSR and peer group TSR, to be included in the disclosure. As some commenters noted, it is not clear that any single measure other than TSR would be relevant across most registrants. Declining to prescribe additional measures would reduce costs and limit the risk that registrants would have to include and discuss a measure that could be misleading or which investors may not find to be useful. This approach could thereby increase the likelihood that investors could process the disclosures quickly, while not decreasing the total amount of underlying information available from public disclosures. At the same time, if the addition of another performance measure would better explain the pattern in executive compensation actually paid, registrants would be able to voluntarily provide such measures, and would likely be motivated to do so.

However, as discussed in the Benefits section above, the inclusion of net income as an additional measure may provide investors with useful context for interpreting the disclosure. Even if required to include a Company-Selected Measure, registrants might not always provide a measure of profitability, in which case net income may help to provide a more complete picture of registrant performance. Further, as discussed above, measures of profitability are commonly used as performance metrics in executive compensation contracts. Yet, if registrants provide measures of profitability in the disclosure, they may be non-GAAP or adjusted measures, and investors may benefit from having net income beside these measures as a benchmark to better understand the effects of such adjustments. Finally, limiting the additional prescribed measures to a single, readily available measure should help to contain the costs and risks of expanding the required measures that are noted above.

We also considered other financial measures as alternatives to net income. As discussed in the Baseline section above, the measures presented by third parties as part of pay-for-performance analyses in recent years—which may reflect investor interest in or demand for the measures—include operating cash flow growth, earnings per share growth, EBITDA growth, return on equity, return on invested capital, return on assets, and various ratios and growth rates using “economic value added.” Measures that commenters suggested we consider include EBITDA, free cash flow, revenue or profit growth, return on investment, shareholder value added, or the ratio of enterprise value to either EBITDA or earnings before interest and taxes (“EBIT”). Overall, these suggestions and the measures presented in third party analyses differ from net income in that many involve some form of scaling—that is, some are ratios, which can help to account for the capital or assets used to generate profits, while others are growth rates—and many include adjustments to focus on operating items or cash flows. It is possible that investors may benefit more from a prescribed measure with these characteristics, rather than net income. However, it is not obvious that there is a single preferred measure, and net income has the benefit of being a clearly-defined, widely-understood measure. Registrants may supplement the disclosure with other measures if they feel they would be useful or if their investors demand them.

Another alternative to the final rules would be, as in the proposed rules, to give registrants the option to include additional performance measures but not to require a Company-Selected Measure of any registrant. As discussed above, if the addition of another performance measure would better explain the pattern in executive compensation actually paid, registrants would likely be motivated to include such a measure on a voluntary basis. Not requiring a Company-Selected Measure would also eliminate any costs or difficulties associated with isolating a single most important measure and give registrants more flexibility to include only the measures that they expect may be most useful to investors. For example, investors may benefit if registrants are able to present a different measure than the Company-Selected Measure in cases where the measure that drove compensation in the last fiscal year may not be the most important for explaining the pattern in executive compensation actually paid over the full five-year horizon of the disclosure. On the other hand, requiring a Company-Selected Measure may elicit additional helpful context in cases where registrants would not otherwise supplement the required performance measures.

As an alternative to the Tabular List, we also considered other approaches to providing context about the measures that were critical in linking pay to performance at a given registrant. For example, we could have required registrants to disclose all of the measures actually used to link pay to performance, with or without quantitative disclosure of the outcomes of the quantifiable measures, any applicable thresholds and targets, and the associated payouts. Such disclosure may provide a more complete view of how pay is linked to performance at a given registrant, and the potential quantitative element may allow investors to more readily assess the sensitivity of pay to particular measures and the rigor of performance goals. Some investors commented that they would benefit from this information being more readily available. However, depending on the specific requirements, such disclosure could be more costly to produce than the Tabular List and may take more time for investors to review, rather than providing simple context and framing for an investor's review of the main table and associated descriptions. There may also be implications of increased transparency of quantitative targets and thresholds, such as pressuring registrants to limit discretion in their pay programs, which may or may not be beneficial. Finally, we note that several commenters mentioned that some registrants are already providing such disclosures, with one indicating that the market does not seem to have coalesced around a consistent format for such disclosures. We expect that market practices in this area may continue to develop.

VI. Paperwork Reduction Act

A. Background

Certain provisions of our regulations and schedules that would be affected by the final rules contain a “collection of information” within the meaning of the PRA. The Commission is submitting the final rules to the Office of Management and Budget (“OMB”) for review in accordance with the PRA. The Commission published a notice requesting comment on changes to these collection of information requirements in the Proposing Release and submitted these requirements to the OMB for review in accordance with the PRA. The hours and costs associated with preparing, filing, and distributing the schedules constitute reporting and cost burdens imposed by each collection of information. An agency may not conduct or sponsor, and a person is not required to comply with, a collection of information unless it displays a currently valid OMB control number. Compliance with the final rules is mandatory. Responses to the information collections will not be kept confidential and there is no mandatory retention period for the information disclosed.

The titles for the collections of information are:

“Regulation 14A and Schedule 14A” (OMB Control No. 3235-0059); and

“Regulation 14C and Schedule 14C” (OMB Control No. 3235-0065).

We adopted the above-referenced regulations and schedules pursuant to the Securities Act or the Exchange Act. The regulations and schedules set forth the disclosure requirements for proxy and information statements filed by registrants to help investors make informed investment and voting decisions. The final rules are intended to satisfy the requirements of Section 14(i).

A description of the final amendments, including the need for the information and its use, as well as a description of the likely respondents, can be found in Section II above, and a discussion of the expected economic effects of the final amendments can be found in Section V above.

B. Summary of Comment Letters and Revisions to PRA Estimates

In the Proposing Release, the Commission requested comment on the PRA burden hour and cost estimates and the analysis used to derive such estimates. While several commenters provided comments on the potential costs of the proposed rules and of the potential requirements discussed and analyzed in the Reopening Release, only one commenter specifically addressed our PRA estimates, stating that the Commission's estimates of the man hour and cost burden of the rule on companies were “grossly underestimated.” As discussed, above, we have made some changes to the proposed amendments as a result of comments received in response to the Proposing Release and the Reopening Release. We have revised our estimates from the Proposing Release accordingly, taking into account the changes and the comments received.

C. Summary of Collection of Information Requirements

We are adding new Item 402(v) to Regulation S-K. This item requires registrants to provide a table containing the Summary Compensation Table measure of total compensation and the values of the prescribed measure of executive compensation actually paid for the PEO and as an average for the other NEOs, TSR both for the registrant and its peer group, the registrant's net income, and a Company-Selected Measure. Item 402(v) of Regulation S-K also requires a registrant to provide a clear description of (i) the relationships between executive compensation actually paid to its PEOs and, on average, to its other NEOs and the registrant's TSR, (ii) the relationship between executive compensation actually paid to the registrant's PEOs and, on average, its other NEOs, and the net income of the registrant, (iii) the relationships between executive compensation actually paid to the registrant's PEOs and, on average, its other NEOs and the registrant's Company-Selected Measure, and (iv) the relationship between the registrant's TSR and its peer group TSR, in each case over the registrant's five most recently completed fiscal years. A registrant will also be required to disclose an unranked Tabular List of its most important financial performance measures used by it to link executive compensation actually paid to its PEOs and other NEOs during the fiscal year to registrant performance. The final rules require registrants to separately tag the values disclosed in the table in Inline XBRL, block-text tag the footnote and relationship disclosure and the Tabular List in Inline XBRL, and tag specific data points (such as quantitative amounts) within the footnote disclosures in Inline XBRL.

The disclosure is required in proxy statements on Schedule 14A and information statements on Schedule 14C in which executive compensation disclosure pursuant to Item 402 of Regulation S-K is required. EGCs, registered investment companies, and foreign private issuers are not required to provide the disclosure. SRCs are subject to scaled disclosure requirements, under which they will not be required to provide a peer group TSR or a Company-Selected Measure (or any related relationship disclosures), nor will they be required to provide a Tabular List or disclose amounts related to pensions; and will only be required to provide three (two in the first applicable filing after the rules become effective) years of disclosure. SRCs must provide the Inline XBRL data beginning in the third filing in which they provide the required pay-versus-performance disclosure.

Much of the information required to produce the pay-versus-performance disclosure is based on items that are already required elsewhere in the executive compensation disclosure and financial statements provided by registrants. In particular, we believe that using as a starting point the total compensation that registrants already are required to report in the Summary Compensation Table and making adjustments to those figures will help reduce the burden on registrants in preparing the disclosure required by new Item 402(v) of Regulation S-K. As discussed above, the final rules are not expected to require registrants to collect significant new data, relative to current disclosure requirements. All of the individual components needed to calculate executive compensation actually paid already must be reported under existing disclosure requirements, with the exception of the values to be included with respect to equity awards and the values to be included with respect to pension benefits for registrants other than SRCs, which are not required to include such pension amounts in their calculation of executive compensation actually paid. Information about net income for all registrants is already required to be disclosed in the registrant's financial statements. Further, information about TSR and peer group TSR is already required to be disclosed in a registrant's annual report to shareholders under Item 201(e) of Regulation S-K, and the measures that make up the Tabular List and the Company-Selected Measure are already considered by registrants when making executive compensation determinations, and may already be discussed, in a different form, in the CD&A. SRCs are not required to provide disclosure under Item 201(e) of Regulation S-K or a CD&A, but also are not required under the final rules to provide disclosure of peer group TSR, the Tabular List, or the Company-Selected Measure. However, SRCs, which currently are not required to disclose their TSR in annual reports, will need to calculate this measure under the final rules.

We arrived at the estimates discussed below by reviewing our burden estimates for similar disclosure and considering our experience with other tagged data initiatives. In addition, the estimates discussed below reflect our belief that much of the information required to prepare the pay-versus-performance disclosure will be readily available to registrants because the information is required to be gathered, determined, or prepared in order to satisfy the other disclosure requirements of our rules, including Item 402 of Regulation S-K. We believe that the amendments regarding pay-versus-performance disclosure will enhance the already required compensation disclosure.

The following PRA Table 1 summarizes the estimated effects of the final amendments on the paperwork burdens associated with the affected collections of information listed in Section VI.A.

The estimated burden increase associated with the final rules for both SRCs and non-SRCs reflects an increase from the estimated average burden increase of 15 hours for all registrants that was included in the Proposing Release. The increase reflects adjustments made due to comments received and accounts for several modifications relative to the proposed rules, including with respect to the calculation of executive compensation actually paid, the addition of net income and the Company-Selected Measure as performance measures to be included in the table, and related relationship disclosures with respect to those performance measures, and the requirement to provide the Tabular List. Because these estimates are averages of the burdens for all such companies in each respective category, the burden could be more or less for any particular company, and may vary depending on a variety of factors, such as the complexity of companies' compensation plans or the degree to which companies use the services of outside professionals, or internal staff and resources, to tag the data in Inline XBRL. This burden, as discussed in more detail below, will be added to the current burdens for Schedule 14A and Schedule 14C.

D. Incremental and Aggregate Burden and Cost Estimates for the Final Amendments

We anticipate that new disclosure requirements will increase the burdens and costs for the affected registrants. We derived our new burden hour and cost estimates by estimating the total amount of time it would take a registrant to prepare and review the disclosure requirements contained in the final rules, as well as the average hourly rate for outside professionals who assist with such preparation. The burden estimates were calculated by multiplying the estimated number of responses by the estimated average amount of time it would take a registrant to prepare and review disclosure required under the final amendments. For purposes of the PRA, the burden is to be allocated between internal burden hours and outside professional costs. For the proxy and information statements on Schedule 14A and Schedule 14C, we estimate that 75% of the burden of preparation is carried by the company internally and that 25% of the burden of preparation is carried by outside professionals retained by the company at an average cost of $400 per hour. The portion of the burden carried by outside professionals is reflected as a cost, while the portion of the burden carried by the company internally is reflected in hours.

We estimate that about 1,275 EGCs are required to file proxy statements on Schedule 14A or information statements on Schedule 14C, in which executive compensation disclosure pursuant to Item 402 of Regulation S-K is required. We have adjusted the estimates to deduct the filings attributed to these companies from our estimate because EGCs are not subject to the final rules. The table below sets forth our estimates of the number of current filings on the schedules that will be affected by the final rules. We used this data to extrapolate the effect of these changes on the paperwork burden for the listed collections of information.

In deriving our estimates, we recognize that the burdens will likely vary among individual registrants based on a number of factors, including the size and complexity of their executive compensation arrangements. We believe that some registrants will experience costs in excess of this average (particularly in the first year of compliance with the final rules) and some registrants may experience less than the average costs. PRA Table 3 below illustrates the incremental change to the total annual compliance burden of affected collections of information, in hours and in costs, as a result of the final amendments.

The following PRA Table 4 summarizes the requested paperwork burden, including the estimated total reporting burdens and costs, under the final amendments.

VII. Final Regulatory Flexibility Analysis

The Regulatory Flexibility Act (“RFA”) requires the Commission, in promulgating rules under Section 553 of the Administrative Procedure Act, to consider the impact of those rules on small entities. We have prepared this Final Regulatory Flexibility Analysis (“FRFA”) in accordance with Section 604 of the RFA. An Initial Regulatory Flexibility Analysis (“IRFA”) was prepared in accordance with the RFA and was included in the Proposing Release. This FRFA relates to the amendments to Item 402 of Regulation S-K, Item 405 of Regulation S-T, Schedule 14A, and Schedule 14C.

A. Need For, and Objectives of, the Final Rules

The final rules are designed to implement the requirements of Section 14(i), which was added by Section 953(a) of the Dodd-Frank Act. Section 14(i) mandates that the Commission adopt rules addressing specified disclosure requirements. Specifically, as described in detail in Section II above, the final rules will require registrants (other than EGCs, registered investment companies, and foreign private issuers) to disclose in any proxy or information statement for which disclosure under Item 402 of Regulation S-K is required, the relationship between executive compensation actually paid to the registrant's PEO and, on average, its other NEOs and the financial performance of the registrant for the three most recently completed fiscal years in the case of a registrant that qualifies as an SRC (or the five most recently completed fiscal years in the case of a non-SRC), taking into account any change in the value of the shares of stock and dividends of the registrant and any distributions.

The final rules require registrants to present pay-versus-performance disclosure that can be readily compared across registrants, while also providing investors with disclosure reflecting the specific situation of the registrant. We believe that the final rules will, among other things, allow investors to assess a registrant's executive compensation actually paid relative to its financial performance more easily and at a lower cost to investors. The need for, and objectives of, the final rules are described in greater detail in Sections I and II.

B. Significant Issues Raised by Public Comments

In the Proposing Release, we requested comment on all aspects of the IRFA, including the nature of any impact on small entities and empirical data to support the extent of the impact. In addition, the Reopening Release included a discussion of the potential impact on SRCs of requiring disclosure of the additional performance measures discussed in that release and also requested comment on a number of matters with respect to SRCs in relation to the proposed rules and the additional requirements considered in that release. We did not receive any comments specifically addressing the IRFA. However, we received a number of comments on the proposed rules generally, and have considered these comments in developing the FRFA. In addition, as discussed in detail above in Section II.G.2, we received a variety of comments on whether SRCs should be subject to the proposed rules. Some commenters supported fully exempting SRCs from the pay-versus-performance disclosure requirements, while another suggested that the pay-versus-performance disclosure be voluntary for SRCs. Other commenters stated that we should not exempt SRCs from the disclosure requirements, some noting that a lack of transparency could have negative market effects for SRCs. Commenters also made a variety of suggestions with respect to the timing of the disclosure for SRCs, including that SRCs be subject to the full pay-versus-performance disclosure requirement but with a one year “grace period,” or that SRCs provide five years of data, but with a three year transition period. One commenter also suggested that the Commission exempt SRCs from the disclosure requirements for five years so that the Commission could first analyze the impact of the disclosure requirements on larger registrants.

C. Small Entities Subject to the Final Amendments

The final rules will affect some companies that are small entities. For purposes of the RFA, under our rules, an issuer, other than an investment company, is a “small business” or “small organization” if it had total assets of $5 million or less on the last day of its most recent fiscal year. The final rules will affect issuers that have a class of securities that are registered under Section 12 of the Exchange Act but are not foreign private issuers, registered investment companies, or EGCs. We estimate that there are approximately 450 issuers that may be considered small entities and are potentially subject to the final amendments. An investment company, including a BDC, is considered to be a “small business” if it, together with other investment companies in the same group of related investment companies, has net assets of $50 million or less as of the end of its most recent fiscal year. We believe that the final rules will affect some small entities that are BDCs that have a class of securities registered under Section 12 of the Exchange Act. We estimate that one affected BDC may be considered a small entity.

D. Projected Reporting, Recordkeeping, and Other Compliance Requirements

We expect the final rules to have an incremental effect on existing reporting, recordkeeping and other compliance burdens for all issuers, including small entities. Under the final rules, SRCs are permitted to provide disclosure in accordance with Item 402(v) of Regulation S-K that is scaled for small companies, consistent with SRCs' existing scaled executive compensation disclosure requirements. Specifically, SRCs are not required to provide a peer group TSR, a Company-Selected Measure, a Tabular List, or to disclose amounts related to pensions. Because SRCs are not required to provide a peer group TSR or Company-Selected Measure, they are similarly not required to provide relationship disclosure with respect to those performance measures. In addition, because the existing scaled definition of NEO in Item 402 of Regulation S-K applicable to SRCs applies for purposes of the new Item 402(v) disclosure, SRCs are required to provide disclosure about fewer NEOs than non-SRC registrants. SRCs also will only be required to provide three years of disclosure (two in the first applicable filing after the rules become effective). Both SRCs and non-SRC registrants are required to separately tag the values disclosed in the table in Inline XBRL, block-text tag the footnote and relationship disclosure and the Tabular List in Inline XBRL, and tag specific data points (such as quantitative amounts) within the footnote disclosures in Inline XBRL, but SRCs are required to provide the required Inline XBRL data beginning in the third filing in which they provide pay-versus-performance disclosure.

Much of the information required in the pay-versus-performance disclosure is based on items that are already required elsewhere in the executive compensation disclosure and financial statements provided by registrants, and the final rules are not expected to require registrants to collect significant new data, relative to current disclosure requirements. Compliance with certain provisions affected by the amendments will require the use of professional skills, including accounting, legal, and technical skills. The final amendments are discussed in detail in Sections I and II above. We discuss the economic impact, including the estimated compliance costs and burdens of the final rules on all registrants, including small entities, in Sections V and VI above.

E. Agency Action To Minimize Effect on Small Entities

The RFA directs us to consider alternatives that would accomplish our stated objectives, while minimizing any significant adverse impact on small entities. In connection with the final rules, we considered the following alternatives:

• Establishing different compliance or reporting requirements or timetables that take into account the resources available to small entities;
• Clarifying, consolidating or simplifying compliance and reporting requirements under the rules for small entities;
• Using performance rather than design standards; and
• Exempting small entities from all or part of the final rules.

As noted above, the final rules will require clear disclosure of prescribed measures of executive compensation actually paid and the company's financial performance and the relationship between these measures. All of the individual components needed for SRCs to calculate executive compensation actually paid already must be reported by SRCs under current disclosure rules, with the exception of the values to be included with respect to equity awards. In addition, net income is required under existing financial disclosure. As discussed above, we do not believe that it is necessary to exempt small entities from the final rules entirely, as we believe the benefit to investors of small entities providing pay-versus-performance disclosure outweighs the costs to them of preparing the scaled disclosure. We have provided some different and simplified compliance requirements for small entities, taking into account their resources. In particular, we have scaled the disclosure requirements for SRCs in an attempt to limit the compliance burden to which such companies will be subject. Accordingly, registrants that are SRCs will be subject to the final rules, but will be permitted to provide only three years of disclosure, instead of five years as required for all other registrants. Also, the final rules will require SRCs to disclose their company TSR and their net income, but they will not be required to disclose peer group TSR, a Company-Selected Measure, or a Tabular List. In addition, because the scaled compensation disclosure that applies to SRCs under existing Item 402 of Regulation S-K does not include pension plans, the pension plan adjustment otherwise required under the final rules will not apply to SRCs. To the extent that a small entity is a registrant, we believe that there are few, if any, small entities that do not qualify as SRCs because it is unlikely that an entity with total assets of $5 million or less would have a public float of $75 million or more. Under the final rules, a small entity, therefore, will likely be subject to the scaled disclosure requirements described above that will apply to SRCs. We believe this will minimize any adverse impact on small entities of providing new disclosures which they generally do not currently provide.

With respect to compliance timetables, the final rules also provide SRCs with transitional relief under which they may provide two years of disclosure, instead of three, in the first applicable filing after the rules become effective, and three years of disclosure in subsequent proxy and information statement filings. The final rules also provide SRCs with a phase-in of the requirement to provide the disclosure in Inline XBRL, under which SRCs need not comply with the Inline XBRL requirement until the third filing in which they provide pay-versus-performance disclosure.

Although the final rules will require disclosure of prescribed measures of executive compensation actually paid and registrant financial performance, they will permit issuers significant flexibility in presenting the relationship between these measures. For example, issuers, including small entities, can describe the relationships in narrative form or by means of a graph or chart, or a combination of both forms. In this respect, the final rules make use of both design and performance standards as a means of balancing the investors' need for uniform disclosure across registrants while also providing registrants, including small entities, with flexibility to describe their pay-versus-performance relationship in a format that is best suited to their particular circumstances.

Statutory Authority and Text of Amendments

The final amendments contained in this release are being adopted under the authority set forth in Section 953(a) of the Dodd-Frank Act and Sections 3(b), 14, 23(a) and 36 of the Exchange Act.

List of Subjects in 17 CFR Parts 229, 232, and 240

• Reporting and Recordkeeping requirements; Securities

For the reasons set out in the preamble, we are amending title 17, chapter II, of the Code of Federal Regulations as follows:

PART 229—STANDARD INSTRUCTIONS FOR FILING FORMS UNDER SECURITIES ACT OF 1933, SECURITIES EXCHANGE ACT OF 1934 AND ENERGY POLICY AND CONSERVATION ACT OF 1975—REGULATION S-K

1. The authority citation for part 229 continues to read as follows:

Authority: 15 U.S.C. 77e, 77f, 77g, 77h, 77j, 77k, 77n, 77s, 77z-2, 77z-3, 77aa(25), 77aa(26), 77ddd, 77eee, 77ggg, 77hhh, 77iii, 77jjj, 77nnn, 77sss, 78c, 78i, 78j, 78j-3, 78 l, 78m, 78n, 78n-1, 78o, 78u-5, 78w, 78 ll, 78mm, 80a-8, 80a-9, 80a-20, 80a-29, 80a-30, 80a-31(c), 80a-37, 80a-38(a), 80a-39, 80b-11, and 7201 et seq.; 18 U.S.C. 1350; sec. 953(a), Pub. L. 111-203, 124 Stat. 1904 (2010); sec. 953(b), Pub. L. 111-203, 124 Stat. 1904 (2010); and sec. 102(c), Pub. L. 112-106, 126 Stat. 310 (2012).

2. Amend § 229.402 by adding paragraph (v) to read as follows:

PART 232—REGULATION S-T—GENERAL RULES AND REGULATIONS FOR ELECTRONIC FILINGS

3. The general authority citation for part 232 continues to read as follows:

Authority: 15 U.S.C. 77c, 77f, 77g, 77h, 77j, 77s(a), 77z-3, 77sss(a), 78c(b), 78 l, 78m, 78n, 78o(d), 78w(a), 78 ll, 80a-6(c), 80a-8, 80a-29, 80a-30, 80a-37, 80b-4, 80b-10, 80b-11, 7201 et seq.; and 18 U.S.C. 1350, unless otherwise noted.

4. Amend § 232.405 by:

a. Revising the introductory text and paragraphs (a)(2) and (4).

b. In paragraph (b)(1)(i), removing the word “and” from the end of the sentence;

c. In paragraph (b)(1)(ii), removing the period from the end of the sentence, and adding “; and” in its place;

d. Adding paragraph (b)(1)(iii);

e. In paragraph (b)(3)(i)(A), removing the word “and” from the end of the sentence;

f. In paragraph (b)(3)(i)(B), adding “and” at the end;

g. Adding paragraphs (b)(3)(i)(C) and (b)(4); and

h. Revising Note 1 to § 232.405.

The additions and revisions read as follows:

PART 240—GENERAL RULES AND REGULATIONS, SECURITIES EXCHANGE ACT OF 1934

5. The general authority citation for part 240 continues to read as follows:

Authority: 15 U.S.C. 77c, 77d, 77g, 77j, 77s, 77z-2, 77z-3, 77eee, 77ggg, 77nnn, 77sss, 77ttt, 78c, 78c-3, 78c-5, 78d, 78e, 78f, 78g, 78i, 78j, 78j-1, 78k, 78k-1, 78 l, 78m, 78n, 78n-1, 78o, 78o-4, 78o-10, 78p, 78q, 78q-1, 78s, 78u-5, 78w, 78x, 78dd, 78 ll, 78mm, 80a-20, 80a-23, 80a-29, 80a-37, 80b-3, 80b-4, 80b-11, and 7210 et seq., and 8302; 7 U.S.C. 2(c)(2)(E); 12 U.S.C. 5521(e)(3); 18 U.S.C. 1350; and Pub. L. 111-203, 939A, 124 Stat. 1376, (2010); and Pub. L. 112-106, sec. 503 and 602, 126 Stat. 326 (2012), unless otherwise noted.

6. Amend § 240.14a-101 by adding paragraph D.5 to the Notes to read as follows:

7. Amend § 240.14c-101 by revising Item 1 to read as follows: