Additional Dissenting Comments on Pay Ratio Disclosure
Commissioner Michael S. Piwowar
Aug. 7, 2015
 The pay ratio disclosure rulemaking has flaws throughout. I further enumerate a number of those defects below.
I. The Proposing Release did not provide sufficient notice under the Administrative Procedure Act.
When engaging in informal rulemaking, the Securities and Exchange Commission (the “Commission”) must satisfy three procedural requirements under the Administrative Procedure Act (“APA”): general notice of the proposed rulemaking, an opportunity for interested persons to participate in the rulemaking, and a concise statement of the basis and purpose of the rules adopted after consideration of the relevant matter presented. These procedural requirements impose necessary discipline upon the Commission as we make policy.
As recognized by the U.S. Court of Appeals for District of Columbia Circuit (“D.C. Circuit”), the aforementioned procedural requirements are intended to result in “an exchange of views, information, and criticism between interested persons and the agency.” The general notice required under the APA “must disclose in detail the thinking that has animated the form of a proposed rule and the data upon which the rule is based.” An agency “has an obligation to make its views known to the public in a concrete and focused form so as to make criticism or formulation of alternatives possible.”
Contrary to the APA, the Proposing Release on the CEO pay ratio disclosure rule failed to identify any objective, goal, or benefit that the Commission believed the rulemaking was intended to accomplish. Rather than stating our thinking on Section 953(b)’s objectives, the Proposing Release merely obliquely indicated that our proposal was intended to achieve “what we believe to be the potential benefits, as articulated by commenters, of the disclosure” mandated by Dodd-Frank.
Pre-proposal comments pointed to a widely dispersed set of conjectured benefits, such as (i) considering vertical pay equity within companies, (ii) improving employee morale and productivity, (iii) evaluating the relative worth of a chief executive officer (“CEO”); (iv) increasing board accountability; (v) facilitating identification of a board’s strengths and weaknesses; and (vi) providing insight into a board’s relationship with its CEO. The Commission never stated which of these concerns – if any, all, combination, or none – it was addressing when it promulgated the pay ratio proposal. This failure to describe the Commission’s thinking about the precise goals or objectives was criticized by at least one commenter in connection with the proposal.
Without knowing what the Commission intended to accomplish with the pay ratio disclosure, it was impossible for the public to comment on whether the regulatory choices made in the proposal were arbitrary and capricious. For example, if the Commission believed that the purpose of Section 953(b) of Dodd-Frank was to “name and shame” companies with a high pay ratio and had articulated that objective, then commenters could have evaluated whether the Commission had a rational basis for its discretionary choices with that objective in mind. In other words, the public would have had the opportunity to comment on whether the Commission’s proposal achieved the maximum amount of shame or whether there were alternative approaches that might achieve desirable levels of shame but at lower cost. Unfortunately, the public was deprived of that opportunity. Thus, the requirements under the APA for the Commission to provide general notice of this rulemaking and an opportunity for comment were not satisfied.
II. Once the Commission decided what objectives Section 953(b) was intended to accomplish, it failed to publicly disclose such understanding prior to adoption.
In the Proposing Release, the Commission correctly recognized that “neither the statute nor the related legislative history directly states the objectives or intended benefits of the provision or a specific market failure, if any, that is intended to be remedied.” The Commission further observed that “there is limited understanding of the legislative intent behind Section 953(b).” As a result, the Proposing Release sparingly made references to Congress or congressional intent.
At some point in time between the proposal and adoption, the Commission interpreted what it believed to be the congressional intent behind Section 953(b) based on its analysis of the statute and review of comments. In contrast to the Proposing Release, the Adopting Release relies significantly on congressional intent – making 76 references to Congress versus only three in the Proposing Release. The Adopting Release asserts that “Section 953(b) was intended to provide shareholders with a company-specific metric that can assist in their evaluation of a registrant’s executive compensation practices” and that “Congress intended Section 953(b) to supplement the executive compensation information available to shareholders.” The Adopting Release further states that the new data about pay ratio may be relevant and useful to shareholders when exercising their say-on-pay voting rights under Section 951 of Dodd-Frank. Therefore, the Commission concluded “we believe the primary benefit of the pay ratio disclosure is to provide shareholders with a company-specific metric that they can use to evaluate the [principal executive officer’s] compensation within the context of their company.”
The purpose behind Section 953(b) goes to the heart of what the Commission is trying to achieve with this rulemaking. In fact, the Commission admitted the importance of finding a purpose for Section 953(b) in the Adopting Release, stating that understanding Congress’ apparent goal was “a significant consideration for us in fashioning a final rule.” Commenters, however, were not given the same opportunity to consider congressional intent, at least as being interpreted by the Commission. It is not sufficient for the Commission to say commenters were on notice because it was merely implementing Section 953(b). For example, the public could not comment on the level of precision required by shareholders to utilize the pay ratio calculation and the estimated costs required to achieve such precision.
By withholding until adoption its views on what Section 953(b) was intended to accomplish, the Commission effectively precluded public discussion on whether its views of Section 953(b) were correct and whether, and to what extent, the Commission’s proposal satisfied those objectives. There was not “enough time with enough information” for the public to provide a meaningful opportunity for comment.
The proper remedy for this failure to provide enough information for the public to respond to key aspects of the rule would be to re-propose the rule and make the public aware of the Commission’s views. I object to the Commission’s inclusion of a hollow, self-serving statement in the Adopting Release that “we do not believe it is necessary to solicit additional public input before adopting the final rule.”
III. The Commission failed to consider what the quantitative effects of providing flexibility would be on the accuracy of the pay ratio. By not evaluating such quantitative effects, the Commission acted in an arbitrary and capricious manner when it limited the de minimis exclusion of non-U.S. employees to 5%.
In both the Proposing and Adopting Releases, the Commission recognized that implementation of Section 953(b) might impose significant costs on companies. Therefore, the Commission purported to adopt flexible rules that would reduce the costs and burdens on companies, while “preserving what we perceive to be the purpose and intended benefits” of the pay ratio. For example, under the rule as adopted, companies can use reasonable estimates in identifying the median employee and calculating annual total compensation or any element of total compensation. Companies may use statistical sampling or other reasonable methods to determine the group of employees from which the median employee is identified. Companies have flexibility to identify the median employee using any compensation measure that is consistently applied to all employees, such as information derived from tax and/or payroll measures, rather than be limited only to using annual total compensation as calculated under Item 402(c)(2)(x) of Regulation S‑K.
Assuming, arguendo, that the most accurate pay ratio would be determined by calculating the total compensation for all employees in accordance with Item 402(c)(2)(x) and finding the median employee based upon such calculations, then the alternative methodologies, using samples and estimates, likely deviate from that result. The Commission, however, did not quantitatively evaluate the potential magnitude of those deviations from the “true” measure, nor did it explain why such deviations were impossible to quantify. So we do not know whether those deviations might significantly undercut the purpose of the pay ratio disclosure, either on a quantitative or qualitative basis.
How much of a deviation in the pay ratio will matter to shareholders? The Adopting Release takes great pains to avoid any finding by the Commission that describes the pay ratio disclosure as material. Rather, the Adopting Release only ascribes findings of materiality to commenters, who provided no data or empirical evidence to support their opinions. The closest the Commission came to addressing the significance of the pay ratio is a statement that the disclosure will provide “new data points” that shareholders “may find relevant and useful” when exercising their say-on-pay vote.
As stated in the Adopting Release, the Commission does not “regard precise comparability as the primary objective of the final rule.” If the “true” pay ratio for a company is 15 to 1, will it no longer satisfy the purposes of the rule if the reported CEO pay ratio, calculated using sampling and estimates, is 20 to 1, 18 to 1 or 16 to 1? If the purpose of the pay ratio disclosure is to be a company-specific metric to evaluate the CEO’s compensation, then the “true” pay ratio may matter much less relative to the ability to observe trends in the pay ratio for that specific company over time. Notably, any discussion about the necessary quantitative level of precision is absent from the Adopting Release.
Thus, the Commission had no rational basis to conclude that the de minimis exemption for employees located out of the United States should be set at 5% because higher amounts would result in unacceptable deviations. In rationalizing the 5% limit, the Commission relied on the illusion of false precision. Having already decided that 5% should be the limit for the de minimis exemption, the Commission only then requested a study from our Division of Economic and Risk Analysis (“DERA”) about possible effects. DERA subsequently concluded that such exclusion might cause the pay ratio to decrease by up to 3.4% or to increase by up to 3.5% (an aggregate range of 6.9%). Nonetheless, the Adopting Release dismissively rejects other alternatives, including one which the DERA study also showed that one measure might result in a decrease in the pay ratio of up to 10.77% or an increase of up to 12.08%. The Commission simply asserted that such amounts are “significant” and “more than negligible.” The fallacy of that reasoning, however, is that no threshold has been determined as either “de minimis” or “significant.”
Even assuming, arguendo, that an aggregate pay-ratio range of 6.9% would be a reasonable measure for determining the appropriate de minimis exclusion threshold, the decision to exclude 5% of employees is arbitrary and capricious. According to the DERA analysis, the 5% exclusion limit is achieved only when the Commission uses the most extreme assumption for wage variance (“sigma” = 0.55). This is based on the DERA literature review and the most extreme scenarios for the percentage of excluded pay observations that are below the median (“Scenario I” assumes all excluded observations are below the median) and the percentage of excluded pay observations that are above the median (“Scenario II” assumes all excluded observations are above the median). If, for example, the Commission had assumed a lower wage variance (“sigma” = 0.25) and more reasonable scenarios (“Scenario I(a)” assumes 75% of the excluded observations are below the median, and “Scenario I(c)” assumes 75% of the excluded observations are above the median), then an aggregate range of 6.9% would have yielded a 20-25% exclusion threshold.
In its discussion of the de minimis exemption, the Commission obtusely rejected any higher level, such as 10%, 15%, or 20%, on the grounds that it “could result in an impact on pay ratio that is greater” than at the 5% level. Of course a higher exclusion level may have a greater impact on the pay ratio; but the question is whether the relevance or usefulness of the pay ratio disclosure would be diminished to the extent that it would no longer serve the purpose of providing a company-specific metric that can assist shareholders in their evaluation of a company’s executive compensation practices. For this question, the Commission gave no answer.
IV. The Commission acted arbitrarily and capriciously when it defined “employee” to exclude contract workers only if they are employed by an unaffiliated third party.
As proposed, the rule text defined “employee” to include “an individual employed by the registrant or any of its subsidiaries.” The Proposing Release elaborated that workers who are “not employed by the registrant or its subsidiaries, such as independent contractors or ‘leased’ workers or other temporary workers who are employed by a third party” (collectively, “contract workers”) would not be covered by the definition. The proposed rule text itself did not directly address contract workers. Although the Commission asked a question as to whether contract workers employed by a third party should be included in the definition of “employee,” the Proposing Release made no distinction between affiliated and unaffiliated third parties.
A number of commenters opposed the inclusion of contract workers in the pay ratio calculation. Only two commenters supported the express inclusion of contract workers. Those commenters who supported inclusion made no distinction between contract workers employed by affiliates and non-affiliates of the company – they would have required inclusion of both types. Thus, nowhere in the public discussion was there any mention of distinguishing the inclusion of contract workers based on whether the third party was an affiliate or non-affiliate of the company.
As adopted by the Commission, the final rule creates a distinction between the treatment of contract workers depending on whether they are employed by a third party and whether that third party is affiliated or not affiliated with the reporting company. The rule text now includes a specific sentence that states the “definition of employee or employee of the registrant does not include those workers who are employed, and whose compensation is determined, by an unaffiliated third party but who provide services to the registrant or its consolidated subsidiaries as independent contractors or ‘leased’ workers.” To exclude contract workers from the definition of “employee,” a company must satisfy at least three requirements: (i) the contract worker is employed by a third party; (ii) the contract worker’s compensation is determined by a third party; and (iii) the third party is unaffiliated with the company. Absent compliance with these three requirements, a contract worker appears to be an “employee,” because any other interpretation would render this part of the definition meaningless and superfluous.
The requirements for excluding contract workers from the definition of employee fundamentally changed the nature of the pay ratio disclosure rule. As adopted, any self-employed contract worker would be counted as an “employee,” because no third party is involved. The following types of self-employed contract workers may be considered “employees” under the pay ratio disclosure:
- a stringer, or freelance journalist or photographer, who has pieces published by a publicly-traded media organization;
- a cybersecurity or computer consultant who does projects for a publicly-traded company;
- a doctor or physician who receives payments from a publicly-traded hospital;
- a lawyer who is a solo practitioner and has a publicly-traded company as a client; and
- an artist who is commissioned to produce artwork for a publicly-traded company.
The change from the Proposing Release also unnecessarily injects the Commission into the current debate on employee and independent contractor classifications for businesses based on a “sharing economy” model, such as Uber. As proposed, the Commission took a neutral approach in this debate, using a principles-based approach as to whether the individual was “employed by” the company. This approach, in essence, deferred any issue regarding classification as an employee or independent contractor to a more appropriate legislative or regulatory body. As adopted, the Commission has firmly come down on the side of those parties who assert that contract workers who work for companies like Uber must be counted as employees.
The Adopting Release is devoid of any economic analysis on the implications of including in the pay ratio contract workers who do not satisfy the three requirements identified above. The comment file is also silent. So in estimating the compliance costs, commenters presumably did not factor in the need to include contract workers or to calculate the cost to determine whether a third party employing the contract worker is an affiliate. While companies have tax and payroll systems in place for employees, records for payments to contract workers may be handled quite differently. For example, for tax reporting purposes, many independent contractors receive a Form 1099 instead of a Form W-2. Amounts recorded on Form 1099 as “nonemployee compensation” often include payments for parts or materials used to perform the services provided, and are not readily equivalent to wages and salary. How a company would reconcile Form W-2 data and Form 1099 data is an important question that the Commission did not address or consider.
The inclusion of the unaffiliated third party requirement for contract workers in the pay ratio also runs afoul of the logical outgrowth principles under the APA. As the D.C. Circuit has observed, while “an agency may promulgate final rules that differ from the proposed rule, a final rule is a logical outgrowth of a proposed rule only if interested parties should have anticipated that the change was possible, and thus reasonably should have filed their comments on the subject during the notice-and-comment period.” The fact that no person commented on a non-affiliation requirement for third parties serves as an indicator that it did not constitute a logical outgrowth.
The Commission’s decision to require that contract workers be employed by unaffiliated third parties also contradicts decisions made elsewhere in the very same rulemaking. In particular, it is directly at odds with the Commission’s reasoning to limit the definition of employee to include only employees of the company and its consolidated subsidiaries rather than the company and its “subsidiaries.”
In the context of pay ratio disclosure, the definitions of “affiliate” and “subsidiary” are linked. Securities Act Rule 405 and Exchange Act Rule 12b-2 define an “affiliate” as “a person that directly, or indirectly through one or more intermediaries, controls or is controlled by, or is under common control with, the person specified.” The same rules define a “subsidiary” as “an affiliate controlled by [an entity] directly, or indirectly through one or more intermediaries.”
Common to both terms is the concept of control. The rules define “control” as the possession, directly or indirectly, of the power to direct or cause the direction of the management and policies of a person, whether through the ownership of voting securities, by contract, or otherwise. As recognized in the Adopting Release, whether an entity is controlled by another is based on the facts and circumstances of each situation, so affiliate status is not always clear. More importantly, the control test includes a potentially broad application of control by contract, with its reach depending on whether the terms of the contract direct or cause to direct the management and policies of another.
The concept of control by contract has potentially very significant consequences for calculation of the pay ratio. For example, publicly-traded companies using a franchise system, such as fast food restaurants and hotels, often have extensive contractual requirements that govern the operation of such establishment. Under the Commission’s definition of “control,” it would be a facts and circumstances determination as to whether such contracts might be deemed to impart control over the third party by the publicly-traded company and potentially require inclusion of contract workers in the pay ratio if they are determined to be providing services to the publicly-traded company.
In considering whether to use “subsidiary” or consolidated subsidiary in the definition of “employee,” the Commission noted that a consolidated subsidiary standard would generally result in a smaller pool of employees being included in the median employee determination. Even though it would result in a smaller pool of employees, the Commission concluded that limiting the rule’s application only to companies and their consolidated subsidiaries “in identifying their median employee should not limit the usefulness of the pay ratio disclosure as a data point for shareholders to use in making their voting decisions on executive compensation under Section 951 of the Dodd-Frank Act in any significant way.”
The requirement that contract workers be employed by an unaffiliated third party in order to be excluded from the definition of employee eviscerates the Commission’s decision to use consolidated subsidiary in lieu of “subsidiary.” Employees of unconsolidated subsidiaries would be swept back into the definition of employee to the extent that they provide services to the company or its consolidated subsidiaries because they are not employed by an unaffiliated third party. This is a contradiction in a fundamental part of the final rule.
In addition, it will be difficult and costly to determine which contract workers should be included in the pool of employees and who is the median employee given that the type of compensation differs significantly between regular employees and contract workers. Nonetheless, the stakes will be incredibly high for accurately categorizing those workers because the pay ratio will subject companies to potential liability under the federal securities laws through potential shareholder litigation and certification requirements under the Sarbanes-Oxley Act. Even if Commission staff were to issue guidance, that would only partially mitigate potential effects because such guidance on the rule would not be binding on the courts.
V. The Commission’s economic analysis failed to consider academic studies as to whether the pay ratio might create pressure to increase CEO compensation.
The Commission’s economic analysis was deficient because it failed to consider fully whether pay ratio disclosure might increase CEO compensation. Academic studies have found that it is possible that pay ratio disclosure could exacerbate any upward bias in executive pay by providing another benchmark that could be used in certain situations to increase CEO compensation (i.e., for a CEO whose company’s pay ratio is lower than its peers’ pay ratios). In addition, there is evidence that setting executive compensation through benchmarking practices is practical and efficient, particularly when the market can observe the method used. To the extent that current benchmarking practices and disclosure requirements are efficient, the additional pay ratio disclosure would not likely provide additional benefits in this regard. The evidence that current CEO compensation practices are not efficient or that the benchmarking process causes an upward bias in executive compensation is not sufficiently clear to establish that the pay ratio disclosure would remedy a specific market failure in current compensation practice.
VI. Use of the pay ratio for comparative purposes among companies may violate an investment adviser’s fiduciary duty under the Investment Advisers Act of 1940.
The Adopting Release expressly states the Commission’s belief that it is not necessary for the pay ratio to have precise conformity or comparability across companies. In particular, the Commission concluded that “there are significant limitations to using the pay ratio information for comparative purposes in light of the various factors that could affect employee compensation at a particular registrant and the flexibility we are providing.” Instead, the Commission asserted that a possible benefit of the pay ratio disclosure is providing a company-specific metric to evaluate the CEO’s compensation within the context of his or her own company.
The Commission found that no particular methodology would necessarily improve comparability because of a multitude of factors that could cause the ratios to be less meaningful for company-to-company comparisons, including:
- differences in industry and business type;
- variations in the way companies organize their workforces to accomplish similar tasks;
- differences in the geographical distribution of employees (domestic or international, as well as in high- or low-cost areas);
- degree of vertical integration;
- reliance on contract and outsourced workers; and
- ownership structure.
The Commission concluded that “there are significant limitations to using the pay ratio information for comparative purposes in light of the various factors that could affect employee compensation at a particular registrant and the flexibility we are providing.” Given the Commission’s clear concerns about comparative use of the pay ratio among companies, an investment adviser that uses CEO pay ratios on a comparative basis when providing advice or making recommendations may well find itself in breach of its fiduciary duty, which requires an investment adviser to have a reasonable basis.
I have many objections to the pay ratio disclosure, as set forth in my remarks at the Commission’s open meeting and my comments above. Should the final rule become effective, I have one request for companies. Please keep track of your compliance costs and consider voluntarily disclosing that information alongside your pay ratio. The Commission and others should have an understanding of your actual compliance costs, and voluntary disclosures would make the likely incredibly high costs evident. But even then, be careful; such information must be “clearly identified, not misleading, and not presented with greater prominence than the required ratio.”
 My remarks delivered at the August 5, 2015 open meeting are available at http://www.sec.gov/news/statement/dissenting-statement-at-open-meeting-on-pay-ratio-disclosure.html.
 5 U.S.C. § 551 et seq.
 5 U.S.C. § 553(b)-(c).
 Home Box Office, Inc. v. Federal Communications Commission, 567 F.2d 9, 35 (D.C. Cir. 1977) (emphasis in original).
 Id. at 36.
 Securities Act Release No. 9452 (Sept. 18, 2013) [78 Fed. Reg. 60560 (Oct. 1, 2013)] (“Proposing Release”).
 Id. at 60562.
 Dodd-Frank Wall Street Reform and Consumer Protection Act of 2010, Pub. L. 111-203, § 953(b) (“Dodd-Frank Act” or “Dodd-Frank”).
 Id. at 60562, n.24.
 Letter from David Hirschmann, U.S. Chamber of Commerce (Dec. 2, 2013). The Adopting Release discusses, but does not directly respond to, the commenter’s concern. See Securities Act Release No. 9877 (Aug. 5, 2015) (“Adopting Release”), at 23, 25-26.
 During the course of the rulemaking, the Commission concluded that Section 953(b) of Dodd-Frank was not intended to publicly “shame” companies concerning the disparity between CEO compensation and the compensation of the median employee. Id. at 12 n. 20. The Commission did not publicly disclose this conclusion until adoption of the final rule.
 Proposing Release, at 60582.
 Id. at 60584.
 Adopting Release, at 9.
 Id. at 11.
 Id. at 12.
 Id. at 11-12.
 Prometheus Radio Project v. Federal Communications Commission, 652 F.3d 431, 450 (D.C. Cir. 2011).
 Adopting Release, at 25-26.
 Proposing Release, at 60563; Adopting Release, at 13.
 Adopting Release, at 13-14.
 Instruction 4(i) to Item 402(u) of Regulation S-K.
 Instruction 4(ii) to Item 402(u) of Regulation S-K.
 Instruction 4(iii) to Item 402(u) of Regulation S-K
 With respect to statistical sampling, the Adopting Release contained some quantitative evaluations assuming a theoretical lognormal wage distribution and a 95% confidence interval with a 0.5% margin of error. See Adopting Release, at 99-100. However, in another context, the Commission acknowledged that deviations from the theoretical lognormal assumption in an actual company can decrease the accuracy of the pay ratio. See id. at 227 n.628, 230.
 See generally TSC Industries, Inc. v. Northway, Inc., 426 U.S. 438 (1976) (defining “materiality”); Basic Inc. v. Levinson, 485 U.S. 224 (1988) (same).
 Adopting Release, at 20.
 Id. at 181.
 Id. at 182.
 In other words, determining total compensation for all employees in accordance with Item 402(c)(2)(x) and finding the median employee based upon such determinations.
 These deviations represent changes of 33%, 20%, and 6.7%, respectively. A ±5% precision requirement would limit a pay ratio to a range between 15.75 to 1 and 14.25 to 1.
 Id. at 74 and 229. In an unfortunate return to the Commission’s past practice of using economists to justify decisions that have been already made, the 5% exemption for non-U.S. employees was included in a term sheet circulated to the Commissioners on September 22, 2014, far before the DERA study had been provided to the Commissioners or released publicly.
 Id. at 75.
 See Division of Economic and Risk Analysis, Potential Effect on Pay Ratio Disclosure of Exclusion of Different Percentages of Employees at a Range of Thresholds (June 4, 2015), available at http://www.sec.gov/comments/s7-07-13/s70713-1556.pdf.
 See Division of Economic and Risk Analysis, Extension of the Analysis of the Potential Effect on Pay Ratio Disclosure of Exclusion of Different Percentages of Employees at a Range of Thresholds (June 30, 2015), available at http://www.sec.gov/comments/s7-07-13/s70713-1559.pdf.
 Adopting Release, at 231. The Commission identified possible cost savings of $105 million at the 10% level and $210 million at the 20% level. See id. at 233 n.640.
 Proposing Release, at 60604.
 As used herein and unless otherwise specified, the term “contract workers” is intended to include any worker “not employed by the registrant or its subsidiaries,” including independent contractors, leased workers, and other temporary workers who are not employed by a third party.
 Id. at 60566 (emphasis added). The language indicates that the contract workers employed by third parties were an example and not an exclusion of other types of contract workers, such as those who are self-employed.
 Id. at 60567. The economic analysis in the Proposing Release recognized that a company might alter its corporate structure or its employment arrangements in order to reduce its compliance costs or to alter the reported ratio. Id. at 60590.
 See Adopting Release, at 49 n.114.
 Letter from Miles Rapoport, Demos (Nov. 22, 2013); Letter from Virginia Fischer (Oct. 3, 2013). The Adopting Release mischaracterizes a third comment letter as supporting the inclusion of contract workers in the pay ratio. See Adopting Release, at 50 n.116 and accompanying text. But that commenter only requested that the final rule have flexibility for a company to include full-time contractors. See Letter from Ronald E. Christian, Vectren Corporation (Dec. 2, 2013).
 The original draft of the Adopting Release, which was received by the Commissioners on July 6, 2015, made no such distinction either. The first time the affiliate language appeared in the final rule text was when a revised draft was distributed to the Commissioners on July 27, 2015, nine days prior to the Commission meeting at which approval was sought.
 Item 402(u)(3).
 Uber is not currently a company registered with the Commission and would not be subject to the pay ratio disclosure. See also Kelly Evans, Uber Casts $50 billion Shadow over Public Markets, CNBC (May 12, 2015), available at http://www.cnbc.com/2015/05/12/uber-casts-50-billion-shadow-over-public-markets.html (observing that increased cost and regulatory burdens may discourage companies from going public).
 In addition, the final rule will potentially result in an inconsistency in a company’s Form 10-K disclosure. Form 10-K requires a description of the business, including the number of persons employed by the company. See Item 1 of Form 10-K; Item 101(c)(1)(xiii) of Regulation S-K. As constructed, the pay ratio calculation could be based on a different number of employees than disclosed elsewhere in the Form 10-K.
 As a methodology to identify the median employee, the final rule permits a company to use any compensation measure that is consistently applied to all employees, such as information derived from a company’s tax and/or payroll records. See Instruction 4(3) to Item 402(u).
 For example, would the Form W-2 and Form 1099 amounts be considered a compensation measure that is “consistently applied” when the Form 1099 might include amounts for parts and materials, in addition to labor?
 International Union, United Mine Workers v. Mine Safety & Health Administration, 407 F.3d 1250, 1259 (D.C. Cir. 2005) (internal quotation marks, brackets, and citations omitted).
 See Business Roundtable v. Securities and Exchange Commission, 647 F.3d 1144, 1148-49 (D.C. Cir. 2011) (finding rules on proxy access arbitrary and capricious because, among other reasons, the Commission had contradicted itself).
 Adopting Release, at 84. A company generally consolidates a subsidiary onto its financial statements if it owns over 50% of the outstanding voting shares of the subsidiary. Id. at 86. Because a company already makes a determination as to whether a subsidiary needs to be consolidated for purposes of financial statement reporting, no further analysis would be required for purposes of the pay ratio disclosure.
 17 CFR 240.12b-2 and 17 CFR 230.405.
 For purposes of the rules, “control” includes the terms “controlling,” “controlled by” and “under common control with.”
 17 CFR 240.12b-2 and 17 CFR 230.405.
 Adopting Release, at 85.
 Id. at 86-87.
 Id. at 87.
 The pay ratio disclosure will be considered “filed” for purposes of the federal securities laws. See Adopting Release, at 143-45.
 Certain academic studies analyze this effect on CEO compensation. See, e.g., M. Faulkender and J. Yang, Inside the Black Box: the Role and Composition of Compensation Peer Groups, J. of Financial Economics 96, 257-270 (2010); C. Elson and C. Ferrere, Executive Superstars, Peer Groups and Overcompensation: Cause, Effect and Solution, J. of Corporation Law 38, 487-531 (2013) (providing the view that benchmarking is inefficient because it can lead to increases in executive compensation not tied to firm performance). See also Letter from Center on Executive Compensation (Dec. 2, 2013) (noting that the pay ratio disclosure may put upward pressure on CEO compensation).
 See, e.g., J. Bizjak, M. Lemmon, and L. Naveen, Does the Use of Peer Groups Contribute to Higher Pay and Less Efficient Compensation?, J. of Financial Economics 90, 152-168 (2008) (indicating that benchmarking is a practical and efficient mechanism used to gauge the market wage necessary to retain valuable human capital).
 See, e.g., J. Bizjak, M. Lemmon, and T. Nguyen, Are All CEOs Above Average? An Empirical Analysis of Compensation Peer Groups and Pay Design, J. of Financial Economics 100, 538-555 (2011) (finding that disclosure of peer groups mandated in the 2006 Adopting Release has reduced the bias in peer group choice).
 See, e.g., J. Core, W. Guay, and R. Thomas, Is U.S. CEO Compensation Broken?, J. of Applied Corporate Finance 12, 97-104 (2005); C. Fryman and D. Jenter, CEO Compensation, Annual Review of Financial Economics 2, 75-102 (2010).
 15 U.S.C. § 80b-1 et seq.
 Id. at 177 (emphasis added).
 Adopting Release, at 102-03.
 Id. at 103.
 Id. at 177 (emphasis added).
 Id. at 12 (text accompanying footnote 21).
 See Securities and Exchange Commission v. Capital Gains Research Bureau, Inc., 375 U.S. 180 (1963) (finding that investment advisers are fiduciaries). Investment advisers that retain a proxy advisory firm have ongoing duties to oversee such firm, including sufficient oversight to ensure that the investment adviser continues to vote proxies in the best interests of its clients. See Staff Legal Bulletin No. 20 - Proxy Voting: Proxy Voting Responsibilities of Investment Advisers and Availability of Exemptions from the Proxy Rules for Proxy Advisory Firms (June 30, 2014), available at https://www.sec.gov/interps/legal/cfslb20.htm.
 See Instruction 9 to Item 402(u).