United States Securities and Exchange Commission

Washington, D.C. 20549



Pursuant to Rule 14a-103


Name of the Registrant: PepsiCo, Inc.

Name of persons relying on exemption: National Center for Public Policy Research

Address of persons relying on exemption: 2005 Massachusetts Ave. N.W., Washington, D.C 20036


Written materials are submitted pursuant to Rule 14a-6(g) (1) promulgated under the Securities Exchange Act of 1934. Submission is not required of this filer under the terms of the Rule but is made voluntarily in the interest of public disclosure and consideration of these important issues.



TO: Shareholders of PepsiCo, Inc.

RE: The case for voting FOR Proxy Item No. 11: Shareholder Proposal – Report on Risks Created by the Company’s Diversity, Equity and Inclusion Efforts

This is not a solicitation of authority to vote your proxy. Please DO NOT send us your proxy card.

The following information should not be construed as investment advice.


Stop Claiming the Business Case for Forced Diversity Has Been Proven:

Fiduciary Duty Prohibits Uncritical Reliance on Contested or Unrigorous Studies

(N.B. — Unless otherwise indicated, when the term “diversity” is used in this memorandum it refers to demographic diversity that is measured along lines such as race, sex, or gender identity. It is important to distinguish this from demonstrable viewpoint diversity.)[2]

Have you been told that the business case for racial and gender diversity has been proven? Before you accept that claim as a justification for managerial efforts (very possibly illegal)[3] to redistribute opportunities and resources on the basis of race,

sex, gender identity, etc. – consider that in approving a diversity-related rule for Nasdaq, the SEC was unable to make that claim after considering all the relevant studies.[4] (It  goes without saying that if there was any way the current  SEC could massage the results to claim a business case for diversity, it likely would have done so.) More broadly, consider also a recent piece in The Wall Street Journal that noted that “a recent report commissioned by the U.K. government …. found there’s little evidence DEI efforts such as mandatory antibias training and corporate policy overhauls have any positive effect on corporate culture.”[5] Finally, consider a study published March 2024 that attempted to replicate the findings of prior studies frequently cited to justify diversity efforts but ultimately concluded that “they should not be relied on to support the view that US publicly traded firms can expect to deliver improved financial performance if they increase the racial/ethnic diversity of their executives.”[6]

Getting into the weeds a bit more, the following critical points[7] were made as part of a recent webinar presented by the asset manager Strive,[8] titled Why DEI Data May Contain Lies.[9] Presenters included Alex Edmans, Professor of Finance, London Business School.

1. The primary sources typically cited[10] for the claim that the business case for diversity has been proven are a series of studies by corporate consultant McKinsey & Company,[11] but these studies suffer from two glaring flaws: data mining and causation errors.[12]

2. As to data mining: While there are many ways to measure both diversity and performance, a basic level of rationality and consistency is required for results to be meaningful.[13]

3. On performance, McKinsey chose EBIT,[14] which is odd given shareholders typically are concerned with total shareholder return (TSR).[15] If one uses any other measure of performance besides EBIT, the positive diversity/performance relationship disappears – suggesting EBIT was cherry-picked and that the results are not robust.[16]

4. Even if one accepts EBIT as an appropriate measure of performance, there are still problems of causation that materially mar the relevant studies.[17]

5. For example, the most recent McKinsey study correlated performance in 2017-2021 with diversity in 2022,[18] making it much more likely that correlation is a result of diversity following performance rather than causing it (certainly, the

measured diversity could not have caused the performance that preceded it).[19] This makes sense when one conceives of diversity as a “luxury good” that firms may be able to afford once they are doing well.[20]

6. What explains the willingness of McKinsey and other related actors to cite such shoddy studies as proof of the business case for diversity? One explanation is confirmation bias. For a variety of reasons, these actors are so invested in the diversity proposition that they will overlook all sorts of study design problems if it allows them to advance their preferred narrative.[21]

7. And Strive has seen this in its portfolio companies: Companies rely on the McKinsey studies to justify their DEI agendas.[22]

8. McKinsey is not alone in this. A recent Blackrock study, Lifting financial performance by investing in women,[23] suffers from similar problems.

9. Even when these studies use return-on-assets as their performance metric, the question remains: Why not TSR?

10. One will sometimes see adjusted TSR used, but the adjustments/benchmarking simply shift the focus of the data-mining / cherry-picking concerns.

11. Another causation problem is that of omitted variables, or a failure to control for other possible or even likely explanations for the benefits attributed to demographic diversity. This is an existential flaw, rendering any study effectively useless. For example, some studies of the mining industry have found a correlation between the lack of diversity in that industry and the relatively poor performance of that industry. But it is far more likely that the poor performance is a function of the market’s evaluation of the industry overall rather than a function of it being male-dominated, particularly in light of the fact that there were no mining companies with sufficient females to serve as a control group.

12. Significantly, when robustness checks are applied,[24] then results showing a positive correlation between diversity and performance go away – though sometimes the study authors shockingly claim the positive correlation results remain regardless.

13. These study design problems are not just limited McKinsey and Blackrock.[25]

14. Another question undermining the validity of these studies is highlighted by asking why companies like McKinsey and Blackrock would giving away this information if it was actually valuable? Should we not expect them to keep the

insights secret and seek to profit from them via their investment decisions, etc.? But giving the information away makes sense if it is just for PR/marketing purposes. In other words, the results are not actually robust enough to profit from in application, but if they give potential clients the results they want to hear, then they become valuable for client relationship building.

15. The negative impact of these bad studies cascades through the corporate ecosystem. For example, when Blackrock, one of the largest controllers of shares in many companies and, along with State Street, by far the largest control block of most large U.S. corporations, releases studies claiming a business case for diversity, managers are likely to view these as marching orders. The problem is extensively compounded if these massive managers of other people’s capital use the adopted power of that capital to pressure companies to act as though the studies are competent, valid, and worthy of reliance in making vital business decisions.

16. All of this means the forces pushing these bad studies act with conflicts of interest in addition to confirmation bias and potential failures of diligence, competence, or related duties.

17. Another critique is that these studies typically focus on upper-level management and directors yet are applied to justify diversity initiatives throughout the organization. Using even competently constructed and robust studies to justify actions not relevantly considered by those studies may constitute an independent breach of duty.

18. Legal liability can flow from using these studies if they lead to formal or implicit demographic diversity quotas, which is a potential cost that must be included in any relevant cost-benefit analysis.

19. The negative impact of these bad studies then trickle down to all hiring and promotion and compensation decisions.

20. Search consultants’ work is also distorted, as they are pressured to present demographically diverse slates so that they can demonstrate their commitment to such diversity for future engagements.[26] Sacrificing merit for diversity by limiting the pool of candidates under consideration in this way should be expected to have a negative impact on performance.

21. There are also overboarding implications. Many corporations want to fill their director positions with CEOs and CFOs of other corporations, purportedly to access the relevant experience and expertise that holding those positions conveys.

But the shortage of CEOs and CFOs able to check a “diversity” box means either (1) diversity candidates will be stretched too thin to do a good job or (2) standards will have to be lowered, likely negatively impacting performance.[27]

22. All of this is not say there aren’t good studies in this space – they just show no positive correlation between diversity and performance, and in fact show negative correlation for forced diversity.[28] It follows that the producers and proponents or demandants of reliance on these flawed studies have known or should have known that credible and competent evidence challenged their fundamentally and non-controversially flawed efforts.

23. An example is Jesse Fried’s article: Will Nasdaq's Diversity Rules Harm Investors? In this piece, Prof. Fried concludes that “the empirical evidence provides little support for the claim that gender or ethnic diversity in the boardroom increases shareholder value. In fact, rigorous scholarship—much of it by leading female economists—suggests that increasing board diversity can actually lead to lower share prices.”[29]

24. This appears to be the general consensus of the good studies. For example, a good study of implementation of the Norway female director quota found that impacted companies performed worse, likely due to scarcity of qualified female candidates.

25. Again, it is important to distinguish forced from organic diversity, and apparently the best that the good studies can say is that the impact of unforced diversity is neutral, or more or less in line with general intuition and the broad consensus about the supremacy of character over color and other demographic differences that only recently frayed, and upon fraying, brought forth these problematic studies. There is no evidence that competent, valid and rigorous studies led to a fraying of the consensus; rather, it appears normative conclusions were reached in some parts of the population, with positive claims and the flawed studies cited for their veracity following. Again, we see the vital importance of competent and correct delineations of causation.

26. As noted in Fried’s article, many of the better studies, which undermine the business case for diversity, are written by women – which reduces the likelihood of bias, at least for studies of gender quotas.[30]

In addition to the foregoing review, we note one additional relevant item. Researchers concluded the following in their 2023 paper, Does Greater Diversity

in Executive Race/Ethnicity Reliably Predict Better Future Firm Financial Performance?

In contrast to the equivocal findings in academic research, “the business case for diversity” is the dominant rhetorical paradigm for how U.S. corporations debate actions and policies around racial/ethnic diversity. In this paper, we conduct an empirical test of the paradigm by gathering data on the race/ethnicity of the individuals shown on the leadership pages of S&P 500 firms’ websites as of mid-2011, 2014, 2017, 2020 and 2021, and then determining if any of nine measures of the racial/ethnic diversity of these executives reliably predict cross-sectional variation in any of six measures of their firms’ financial performance over the next fiscal year. We do not find reliable evidence that they do. As such, our results do not support the “business case for diversity” when the claim is assessed using 1-year-ahead financial performance metrics and multiple measures of the race/ethnicity of S&P 500 executives over the last decade.[31]

Discriminating on the basis of protected categories like race and sex is illegal and immoral. This would be true even if one has good reasons for it. But it becomes particularly pernicious when powerful actors like corporations and corporate consultants fuel that discrimination with lies promulgated under the cover of statistics. We urge shareholders to vote FOR Proxy Item No. 11: Shareholder Proposal – Report on Risks Created by the Company’s Diversity, Equity and Inclusion Efforts.







For questions regarding our shareholder proposal, Proxy Item No. 11, on the PepsiCo, Inc. 2024 proxy ballot, please contact NCPPR at info@nationalcenter.org.


[2] Unless otherwise indicated, when the term “diversity” is used in this memorandum it refers to demographic diversity that is measured along lines such as race, sex, or gender identity. It is important to distinguish this from genuine viewpoint diversity. To wit, a rainbow coalition of radical leftists is simply an echo chamber. Cf. “In theory, diversity of thought strengthens organizations by preventing groupthink; in practice, the business world’s pursuit of diversity has become skin-deep.” https://www.strive.com/documents/FG/strive/news/627521_Shortened_McKinsey_White_Paper.pdf

[3] “After the Supreme Court’s decision striking down university affirmative action, courts will likely enforce Title VII’s ban on the corporate version ….” https://www.strive.com/documents/FG/strive/news/627521_Shortened_McKinsey_White_Paper.pdf

[4] Self-Regulatory Organizations; The Nasdaq Stock Market LLC; Order Approving Proposed Rule Changes, as Modified by Amendments No. 1, To Adopt Listing Rules Related to Board Diversity and To Offer Certain Listed Companies Access to a Complimentary Board Recruiting Service, 86 FR 44424-01 (“Taken together, studies of the effects of board diversity are generally inconclusive ….”). Cf. James R. Copland, Ilya Shapiro, Tim Rosenberger, Amicus Brief: Alliance for Fair Board Recruitment vs. Securities and Exchange Commission, MANHATTAN INSTITUTE ( Mar. 28, 2024) (“the SEC conceded that these rules have no impact on corporate performance”), available at https://manhattan.institute/article/amicus-brief-alliance-for-fair-board-recruitment-vs-securities-and-exchange-commission .

[5] Editorial Board, A Lesson in DEI Failure From Britain: Diversity, equity and inclusioncosts billions and doesnt work, WALL STREET JOURNAL (Mar. 26, 2024), available at https://www.wsj.com/articles/diversity-equity-inclusion-united-kingdom-report-d73d47ed .

[6] Jeremiah Green & John R. M. Hand, McKinseys Diversity Matters/Delivers/Wins Results Revisited, 21 ECON JOURNAL WATCH 5 (Mar. 2024), available at https://econjwatch.org/articles/mckinsey-s-diversity-matters-delivers-wins-results-revisited .

[7] All points are paraphrased and presented in the order they were raised in the webinar.

[8] See generally, https://www.strive.com/the-strive-story (“Strive was recognized in June 2023 as the fastest growing asset manager with under a billion dollars in assets under management (AUM), before outgrowing the category just a few months later by eclipsing the billion-dollar AUM mark in September 2023.”).

[9] https://www.youtube.com/watch?v=ERHOP3l4s-g (hereinafter “webinar”)

[10] See webinar at 5:40 (listing Coca Cola, Bank of America, and Exxon as a few examples of companies that have cited the McKinsey studies to justify their DEI efforts).

[11] See, e.g., Diversity matters even more: The case for holistic impact, available at https://www.mckinsey.com/featured-insights/diversity-and-inclusion/diversity-matters-even-more-the-case-for-holistic-impact .

[12] Cf. Jeremiah Green & John R. M. Hand, McKinseys Diversity Matters/Delivers/Wins Results Revisited, 21 ECON JOURNAL WATCH 5 (Mar. 2024) (“Combined with the erroneous reverse-causality nature of McKinsey’s tests, our inability to quasi-replicate their results suggests that despite the imprimatur given to McKinsey’s studies, they should not be relied on to support the view that US publicly traded firms can expect to deliver improved financial performance if they increase the racial/ethnic diversity of their executives.”), available at https://econjwatch.org/articles/mckinsey-s-diversity-matters-delivers-wins-results-revisited .

[13] Cf. “[A]cross the three reports, the underlying dataset is always shifting, displaying arbitrary choices in the timeframes it covers and companies it includes.” https://www.strive.com/documents/FG/strive/news/627521_Shortened_McKinsey_White_Paper.pdf

[14] Earnings before interest and taxes.

[15] Cf. “The authors must have had the data about executive diversity’s relation to revenue, margin, and earnings growth, but chose to show one narrow measure of financial performance and hide the most fundamental ones.” https://www.strive.com/documents/FG/strive/news/627521_Shortened_McKinsey_White_Paper.pdf

[16] Cf. “Cobbling datasets together in this patchwork way to show a strengthening correlation is arbitrary at best, dishonest at worst. It displays all the signs of p-hacking, the statistical trick of adding artificial criteria until one’s data supports a predetermined conclusion. Most people call it cherry-picking.” https://www.strive.com/documents/FG/strive/news/627521_Shortened_McKinsey_White_Paper.pdf

[17] “Boardroom diversity quotas and race and gender-based hiring and promotion policies are founded on the belief that these kinds of diversity cause corporate outperformance. But the research they are based on disavows any causal claims.” https://www.strive.com/documents/FG/strive/news/627521_Shortened_McKinsey_White_Paper.pdf

[18] Cf. “The choice of timeframes raises multiple red flags.” https://www.strive.com/documents/FG/strive/news/627521_Shortened_McKinsey_White_Paper.pdf

[19] Cf. “Without evidence that diversity causes outperformance, why should companies expect diversity measures to improve their performance? The relationship between ice cream consumption and murder is in a sense real, but entirely correlational; both happen more often during warm weather. Causation is what matters.” https://www.strive.com/documents/FG/strive/news/627521_Shortened_McKinsey_White_Paper.pdf

[20] Cf. “facts suggest companies adopted diversity measures during boom times and are backing away from them now that profit is harder to come by”; “one-third of DEI professionals have left their jobs within the last year, compared to a non-DEI attrition rate of 21%. If successful companies thought these employees increased profitability, they would not be among the first to be fired” https://www.strive.com/documents/FG/strive/news/627521_Shortened_McKinsey_White_Paper.pdf

[21] Cf. Washington Examiner, The DEI elites have no clothes, WASHINGTON EXAMINER (Apr. 5, 2024) (“The author of these McKinsey studies, Vivian Hunt, once explained why her research was so important. ‘Neutral is no longer neutral. A neutral position, that is meritocratic, that it is good to treat people evenly, isn’t good enough,’ she said in 2020. ‘You have to proactively stand for an anti-racism environment.’”), available at https://www.washingtonexaminer.com/opinion/2952982/the-dei-elites-have-no-clothes/ .

[22] “The McKinsey studies are influential, partly because they were accepted uncritically by mainstream financial publications like Financial Times and The Wall Street Journal.” https://www.strive.com/documents/FG/strive/news/627521_Shortened_McKinsey_White_Paper.pdf

[23] https://www.blackrock.com/corporate/literature/whitepaper/lifting-financial-performance-by-investing-in-women.pdf

[24] Cf. Validation and Verification of Analytical Testing Methods Used For Tobacco Products: Guidance For Industry, FOOD DRUG COSM. L. REP. P 360070 (Dec. 21, 2021) (“Robustness is the measure of an analytical test method's capacity to remain unaffected by small, but deliberate, variations in method parameters and provides an indication of the reliability of the method during normal usage.”).

[25] Cf. Scott Shepard, Corporations Undermine Their Anti-Discrimination With Support of Discrimination, REALCLEARMARKETS (Nov. 21, 2022), available at https://www.realclearmarkets.com/articles/2022/11/21/corporations_undermine_their_anti-discrimination_with_support_of_discrimination_865769.html


More than 60 corporations signed on to an amicus brief in the U.S. Supreme Court case heard shortly before the midterm elections that will determine whether diversity-excused race discrimination may continue in American higher education. In making their case, these corporations demonstrated the corruption that runs through all their attempts to justify their politically motivated, hard-left “ESG” boosterism: incoherent thinking built on profoundly flawed and goal-sought research…. The brief authors marshalled six studies in favor of its contention that “racial and ethnic diversity enhances business performance.” None of them does that work.


Id. (analyzing studies).

[26] Cf. Association of Executive Search and Leadership Consultants (AESC), Diversity and Inclusion and the Role of Executive Search (citing McKinsey study and reporting on a “a global survey of executive search professionals” that found “more than 300 respondents suggest that D&I is a key agenda item, with over two thirds of respondents stating that it will be ‘highly important’ to their clients in 2019”), available at https://www.aesc.org/insights/magazine/article/diversity-and-inclusion-and-role-executive-search .

[27] Cf. “Lower hiring standards could lead to lower profits. Performance could further suffer if employees lose motivation upon seeing peers receive career opportunities based on race or gender instead of merit…. Treating employees equally and fairly is incompatible with giving out career opportunities based on race or gender.”  https://www.strive.com/documents/FG/strive/news/627521_Shortened_McKinsey_White_Paper.pdf

[28] Cf. “Texas A&M accounting professor Jeremiah Green and UNC business professor John R.M. Hand attempted to replicate McKinsey’s analysis using S&P 500 data from 2015 to 2019. They found no correlation between executive racial diversity and chance of EBIT margin outperformance…. McKinsey declined to comment on their results.” https://www.strive.com/documents/FG/strive/news/627521_Shortened_McKinsey_White_Paper.pdf

[29] https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3812642 (reviewing studies) (summary available at https://corpgov.law.harvard.edu/2021/04/08/will-nasdaqs-diversity-rules-harm-investors/ ).

[30] See, e.g., Renée B. Adams and Daniel Ferreira, “Women in the Boardroom and Their Impact on Governance and Performance.” Journal of Financial Economics, Vol. 94 No. 2, November 2009, 291-309, https://doi.org/10.1016/j.jfineco.2008.10.007 .

[31] https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4576173