野蛮人喜欢私有化什么

QUARTERLY LETTER Part 1 2Q 2026 WHAT BARBARIANS LIKE TO TAKE PRIVATE (Or: The Risks in Your Private Equity Portfolio) Ben Inker and John Pease | Pages 2–12 Part 2 LETTER TO THE INVESTMENT COMMITTEE ON PRIVATE EQUITY The Persistent Dream of Performance Persistence Ben Inker | Pages 13–16 2Q Part 1: QUARTERLY LETTER 2Q 2026 WHAT BARBARIANS LIKE TO TAKE PRIVATE (Or: The Risks in Your Private Equity Portfolio) Ben Inker and John Pease | Asset Allocation Introduction When the economy suffers, so too do stocks. This is both obvious—it would be bizarre for even a diversified set of companies to thrive at a time when people and businesses are demanding and consuming less—and somehow, easy to forget. This fugue is particularly prevalent in the private corners of the U.S. equity market, where years of strong performance, coupled with occasionally stale marks, have erased the idea of “true downside.” Concerns about private equity today center on illiquidity: the inability of many sponsors to offer distributions at the same rate as they have in the past. But the true downside for private stocks—much as with their public counterparts—is insolvency. With private equity portfolios looking ever more concentrated on a small set of risks, investors in these portfolios look more vulnerable to certain types of economic shocks than they ever have.

We say this reasonably confidently because over the past few months we have collected, cleaned, and analyzed data—God bless LLMs1 —for roughly 700 formerly listed developed-market companies that were taken private via a leveraged buyout from 1981 to 2025, including the likes of RJR Nabisco, Hospital Corporation of America (twice!), and TXU. Given that LBOs represent roughly 80% of the capital deployed by private equity funds,2 this data helps us understand what the companies in PE portfolios look like. Our sample is not entirely representative of the buyout space (the average public-to-private LBO is much larger than its private-to-private counterpart), but it is a substantial portion of it, covering 10–30% of the dollar value in global buyout activity over most years (Exhibit 1). Arguably, our sample should also be similar on other traits, like profitability and valuation, assuming sponsors choose rationally between the companies they can possibly purchase and considering that a company’s listed status is irrelevant for how much it can add in value to stakeholders.3 ...which have no role in writing this letter. You still get artisanal human text when you read GMO! Or, at least, the large language model summarizing this letter for you does.

The other 20% are in the “growth” category. See, for instance: Alternative Investments: Private Equity & Venture Annual Report (Bloomberg 2025).

Though this is an argument that holds true controlling for size, our sample might still be quite different from what you would get in smaller private-to-private buyout funds.

Part 1: What Barbarians Like to Take Private | p3 EXHIBIT 1: PUBLIC-TO-PRIVATE LBO SHARE OF TOTAL LBO DEAL VALUE 0.35 0.30 0.25 0.20 0.15 0.10 0.05 0.00 As of 12/31/2025 | Source: GMO, Bain/Dealogic Our initial goal for this letter was to use this data to explain how private equity portfolios have changed over time. The story the data tells is more interesting than we expected, however, so we will therefore cover three things: first, as originally intended, we show how PE companies have evolved from a fundamental point of view; second, we describe how PE portfolios have become more concentrated from a risk perspective; and third, we discuss what allocators with a meaningful slug of private equity in their portfolios can do to mitigate the higher concentration of risks therein.

What’s in the Box?

It is odd that the standard depiction of buyout targets is of small, unleveraged, profitable, and cheap companies with a secular growth tailwind. While such companies do exist, they are far from the typical players that come to mind when LBOs are mentioned. RJR Nabisco was an expensive purchase when contrasted with its peers; Toys"R"Us was leveraged to the hilt even before its leveraged purchase in 2005; Hilton was running razor-thin net margins when it was acquired despite being contrasted against its low-margin industry. In fact, these companies were all over the map on leverage, profitability, and valuations versus comparable 4 businesses at the same point in time (Exhibit 2).

EXHIBIT 2SELECTED LBO CHARACTERISTICS RELATIVE TO MATCHED PEERS Leverage (Net Debt/Assets) Net Margin Price/GMO Fair Value Relative to Peers (%) Difference (p.p.) Relative to Peers (%) 1.2 4% 0.3 1.0 2% 0.2 0% 0.8 0.1 -2% 0.6 0.0 -4% 0.4 -0.1 -6% 0.2 -8% -0.2 -10% -0.3 0.0 RJR Toys"R"Us Hilton RJR Toys"R"Us Hilton RJR Toys"R"Us Hilton Nabisco Nabisco Nabisco We find peers by matching names on date, size, and industry group. As of 12/31/2025 | Source: GMO, MSCI, Bloomberg, Compustat, Worldscope Part 1: What Barbarians Like to Take Private | p4 Looking through a wider lens, the only thing about the buyout industry that is consistent over time is its focus on small caps. In our entire dataset, only RJR Nabisco was among the top 100 companies by size at the time of its purchase. Indeed, only about 5% of the companies were among the top 500 by market cap, and fewer than 15% were within the top 1,000. The remaining 85% were either small caps or, smaller still, veritable microcaps. If anything, our data overstates the size of LBOs (Exhibit 3), given it is easier to capture data on larger buyouts, especially pre-2000, and public-to-private transactions skew larger than private-to-private ones.

EXHIBIT 3PUBLIC-TO-PRIVATE LBO ROLLING 5-YEAR SIZE CATEGORY COMPOSITION Shares of LBO Count (%) 1980 1985 1990 1995 2000 2005 2010 2015 2020 2025 Megacap Large Cap Midcap Small Cap Microcap As of 12/31/2025 | Source: GMO, MSCI, Bloomberg It is not uncommon to read commentary from buyout sponsors arguing that this small bias makes private equity a diversifier, especially when considering that 85% of investable companies5 are private.6 Their reasoning is two-fold: public markets skew larger and are therefore more concentrated, while private markets skew smaller and thus have a broader opportunity set.7 While these two statements are factual, the implication that they make private equity a diversifier is wrong.

Consider a simplified model of reality in which each company is exposed to only two risks: the business cycle, which we’ll assume affects every company, and management, which we’ll declare to be a firm-specific risk. Even if, for any individual company, management contributes twice as much to overall risk as the business cycle, a portfolio of companies exposed to different management teams quickly ceases to have any meaningful risk beyond their exposure to the economy. By the time the portfolio has 100 companies at roughly equal weight, “specific” risk becomes a rounding error. The reason behind this is that the idiosyncratic gains and losses that befall a specific business are, by definition, not shared. If Companies with more than USD 100 million in revenues.

6 you have one company, such idiosyncrasies have nothing to cancel them out; if you have 100, See, for instance: Why Private Equity? Why Now? Reasons the lucky and unlucky will mostly cancel, and the experience of the portfolio will simply be that to Invest in Private Equity in 2026 (KKR 2026).

of owning an average business. Expanding a portfolio from 100 to 10,000 companies, in this Both are a consequence of Zipf’s Law (as applied to firm simplified world, gives you negligible diversification gains (Exhibit 4).8 size).

For the mathematically curious, we are setting for all , σ and σ Changing the parameters, introducing heterogeneous market betas, or adding other factors doesn’t change the underlying results unless you believe private companies are significantly more idiosyncratic than public companies.

Part 1: What Barbarians Like to Take Private | p5 EXHIBIT 4: SPECIFIC RISK VANISHES WITH DIVERSIFICATION Portfolio Volatility (%) Specific Risk 25 (Management) Systematic Risk (Business Cycle) 1 26 51 91 162 288 513 912 1,622 2,884 5,129 9,120 Number of Equally Weighted Stocks Source: GMO The real world is, of course, messier than our simplified model. Businesses that share certain characteristics—leverage, low profitability, a similar industry makeup—have outcomes that are fundamentally more correlated than what is implied by a “business-cycle-only” model. And this is where the argument for private equity as a diversifier is at its weakest. If you owned the entire PE complex to be maximally diversified, you would just end up holding a big basket of small caps. Not only have small caps seldom been a lower-quality group of businesses than they are today, but the buyout space is lower-quality still. If we rank companies from 0–10 on a composite of profitability, the stability of profitability, and leverage (what we call “Quality”), public-to-private LBOs rank a full quality decile below small caps, who in turn are almost four deciles junkier than the S&P 500 (Exhibit 5).

EXHIBIT 5GMO QUALITY SCORE GMO Quality Score 1980 1984 1988 1992 1996 2000 2004 2008 2012 2016 2020 2024 S&P 500 U.S. Small LBOs As of 12/31/2025 | Source: GMO, MSCI, Bloomberg, Compustat, Worldscope The problem with low-quality stocks is that they are more susceptible to adverse economic shocks. A company that is unprofitable and leveraged requires external financing—equity or debt issuance—to pay the bills in normal times. In periods of economic distress, raising capital can be significantly more expensive (or even impossible), and low-quality companies can be forced to write down large parts of their businesses to stave off bankruptcy. If we rank companies by their quality score right before two nasty economic events—the GFC and Covid— and look at their subsequent five-year performance, we can see that low-quality businesses are far less likely to generate positive returns versus their high-quality counterparts.

Part 1: What Barbarians Like to Take Private | p7 EXHIBIT 7: RECENT LBO CHARACTERISTICS VS. PEERS Leverage Valuation (P/FV) (Net Debt/Assets) Net Margin 3.5 1.2 0.6 3.0 1.0 2.5 0.4 2.0 0.8 1.5 0.2 0.6 Log Ratio Difference Difference 1.0 0.4 0.0 0.5 0.0 0.2 -0.2 -0.5 0.0 -1.0 -0.4 -1.5 -0.2 -2.0 -0.4 -0.6 As of 12/31/2025 | Source: GMO, MSCI, Bloomberg, Compustat, Worldscope It defies belief that a highly compensated, profit-maximizing subset of the equity market would be paying up for crappy businesses until we line up three facts. The first is that the value proposition of a good PE manager is to turn a business around. While net margins for companies taken private are low, gross margins tend to be high, and LBO managers operate under the premise that they can bring them closer together.9 Indeed, the fact that PE has managed to outperform public equities while investing in low-quality businesses (which tend to underperform) is a sign that they probably can, to some extent, turn companies around.10 The second fact is that private equity has grown a lot; there is no dearth of managers or capital in the space, and competition has driven up prices. Finally, and most worrisome: the makeup of the private equity industry has changed dramatically. Buyouts are often thought of (by the authors of this piece, at least) as being focused on industrials and food companies. However, over the past 10 years, public-to-private LBOs have been heavily skewed toward software, which accounts for 40% of the recent deals closed (Exhibit 8).

EXHIBIT 8PUBLIC-TO-PRIVATE LBO ROLLING 5-YEAR INDUSTRY GROUP COMPOSITION Share of LBO Count (%) And presumably not by tanking gross margins.

Comparing the performance of private equity and public equity is tricky. The IRRs often quoted by PE firms are not 40 directly comparable to returns in public equity markets, and even capturing all private equity funds and determining which are appropriately aggregated is a difficult task. 20 Cambridge Associates does as good a job as any. On their data (US Private Equity: Index and Selected Benchmark Statistics), U.S. private equity broke even against the Russell 3000 over the decade-ending September 2025, 1984 1989 1994 1999 2004 2009 2014 2019 2024 with some modest alpha over longer periods. This arguably understates PE's alpha relative to its selection universe.

Software & Services Capital Goods Smaller-cap and junkier companies have substantially Health Care Equipment & Services Technology Hardware & Equipment underperformed the Russell 3000 over the last 20 years, Consumer Durables & Apparel Consumer Services which makes the alpha of PE against the universe of Consumer Discretionary Distribution & Retail Consumer Staples Distribution & Retail stocks they were selecting from better to the tune of Media & Entertainment Commercial & Professional Services several percentage points. PE relative performance since Food, Beverage & Tobacco Materials the peak of the 2021 bubble (and the sustained rise in Other interest rates), on the other hand, has been poor against pretty much any U.S. public equity index. As of 12/31/2025 | Source: GMO, MSCI, Bloomberg, Compustat, Worldscope Part 1: What Barbarians Like to Take Private | p8 This industry bet is, by far, the most concentrated we’ve seen in the history of buyouts. It is a consequence, as one buyout manager mentioned to us, of the multiple expansion of software companies, and an acknowledgement that asset-light businesses often have higher operational leverage and are therefore particularly good targets should they be able to grow. Whether a “good” bet or not, the scenario isn’t mimicked in public equities. Small-cap indices, like the Russell 200011 and the MSCI U.S. Small Cap Index,12 have 5% or less in Software and Services as of March 2026. Large-cap indices are likewise light on pure software: the S&P 500 only holds 10% in the group, while MSCI ACWI ex-U.S. holds a measly 5%. The buyout industry is therefore taking a massive relative bet that software will continue to eat the world. But the story in 2026, as you are well aware, is that the AI world will eat software. If investments in small software businesses in the public equity markets (Exhibit 9) are any indication of the broader impact, recent LBO fund vintages are likely in a tough spot.

EXHIBIT 9YTD CUMULATIVE RETURN Software vs. Benchmarks 10% 0% Cumulative Return -10% -20% -30% 12/2025 1/2026 2/2026 3/2026 4/2026 S&P 500 S&P 1500 S&P 1500 Software As of 5/8/2026 | Source: GMO, S&P Global, MSCI, Bloomberg Our base case, to be fair, is not that these private businesses will go to zero. Private equity sponsors have plenty of smart people working hard to understand how to integrate AI into their companies to boost productivity and reduce the risk of disruption. Arguably, given they manage portfolios of businesses, private equity managers are better equipped to learn and adapt than the management teams of individual companies, which have less information about which process changes have proved successful elsewhere. But being smarter and working harder does not insulate PE portfolio companies from exposure to a technological shift that could radically undercut their value proposition. If AI allows competitors to sprout with meaningfully lower launch costs, or enables clients to build their own customized software at a fraction of the cost, some value erosion to the software industry is inevitable. Moreover, given the generally high valuations at which these software companies were bought, a derating of the terminal multiple for such firms will be quite painful—even if the other financial assumptions buyers made at the time of purchase hold true. A concentrated bet on software is therefore a genuine risk, and one allocators should think very seriously about today, especially in the context of venture capital and private credit (the other major private assets investors with significant PE allocations tend to hold). While investors are suddenly shocked (shocked!) to find a large software concentration in their private credit portfolios, venture capital portfolios https://www.ishares.com/us/literature/fact-sheet/iwm- have an even higher concentration in the software space—meaning that overall, the risk to ishares-russell-2000-etf-fund-fact-sheet-en-us.pdf endowment-style portfolios is substantially higher than our pure-buyout findings may suggest. 4197-a951-9605fce4772f Part 1: What Barbarians Like to Take Private | p9 Hedging the Downside Most private equity investors understand the economic downside risk of their portfolios, even if they value the lower measured volatility that comes with the leisurely pace of appraised value changes. When thinking of their overall equity allocation, PE investors account for the existence of their private assets and reduce their public equity allocations accordingly. What they generally fail to do is bake in the characteristics of the companies owned by private equity funds when considering how to manage the public equity (or liquid alternative) parts of their portfolio.

Pairing private equity with passive index exposure in the public markets, as some large investors are increasingly doing, mitigates this risk a bit, but it only stops the public equity book from piling into the same risks as the private equity portfolio. We’d argue that investors may be better served by running their public equity book to actively counter the risks of their private equity holdings.

As an example, let’s consider an institution that gets half of its U.S. equity exposure from private equity and the other half from a passive portfolio mirroring the overall U.S. public equity market. Relative to the overall U.S. equity market (public and private equities combined) the resulting portfolio is hugely overweight small-cap and lower-quality stocks. Correspondingly, it has a huge underweight to both large-cap and high-quality companies. If we were to cuff the aggregate PE holdings of U.S. companies at ~$5 trillion, their size would be approximately half the ~$10 trillion of non-S&P 500 U.S. companies, and only 7% of the ~$69 trillion market cap of the S&P 500 itself. This leaves the aggregate U.S. equity portfolio for our institution with a 57% weight in companies smaller than the S&P 500 and a 43% weight in S&P 500 constituents. "Indexing” the aggregate of public and private equity owned companies would assign an 82% weight to the S&P 500 and a 6% weight to private equity. While the institution presumably believes its private equity holdings will generate substantial alpha relative to public equities, the aggregate small and low-quality bias of the resulting portfolio is not necessarily something one would want given the return and risk characteristics of small caps and low-quality stocks.

To approximately neutralize this bias on an index basis would involve taking a long position in the S&P 500 or S&P 100 equal to 38% of the total U.S. equity exposure inclusive of public and private holdings, and shorting a small-cap index, such as the Russell 2000, by the same amount. While this could be collateralized by the public equity holdings, the mark-to-market movements and consequent swings in posted collateral created by such a position would be massive. And while periods when the Russell 2000 outperforms the S&P 500 should be economically correlated to those in which private equity does particularly well relative to public equity, the private equity portfolio will certainly not be spinning off cash in a manner that helps mitigate the collateral funding problem. It is also worth noting that while this hedge should reduce the downside economic risk of the aggregate U.S. equity portfolio, it neither has a positive expected long-term return nor does it mitigate the AI disruption risk from the software concentration in PE.

We believe a superior way to counter the small and low-quality bias of such a portfolio would be to replace the passive public equity exposure with an equity portfolio that takes advantage of one of the more persistent inefficiencies of the public equity market: the fact that high-quality stocks tend to mildly outperform over time and low-quality stocks tend to substantially underperform. Exhibit 10 shows the performance of high-quality, low-quality, large-cap, and small-cap stocks in the U.S. equity market since 1981.

Part 1: What Barbarians Like to Take Private | p10 EXHIBIT 10: REAL RETURN SINCE 1981 12% 10.4% 10% 8.6% 8% 7.4% 6% 5.2% 4% 2% 0% S&P 500 Small Caps Low Quality High Quality Data from 1981 to 2026 | Source: Worldscope, GMO, S&P Global Small Caps are the 2,000 smallest stocks within the largest 3,000 U.S. stocks by market cap. High Quality is the top 25% of high-quality companies by market cap within the 3,000 largest U.S. stocks, measured by GMO’s quality factor. Low Quality is the bottom 25% of low-quality companies by market cap within the 3,000 largest U.S. stocks, measured by GMO's Quality Score.

The underperformance of small caps relative to the S&P 500 is a relatively recent phenomenon that depends on start and end dates. Low-quality stocks, on the other hand, have consistently underperformed both the broader market and high-quality stocks despite having a consistently high beta. Another benefit of shorting the low-quality group instead of small caps is the fact that software has more than twice the weight in the low-quality group than it does in broad small-cap indices. We believe a hedge of this sort offers three important benefits relative to its index-based counterpart: 1. It is expected to outperform, given that junky stocks are consistently overpriced due to investors’ propensity to pay up for moonshots.

2. It offers downside protection due to the large beta discrepancy between high-quality companies (who navigate drawdowns well and have low downside beta) and their low-quality cousins (who tend to implode spectacularly during corrections as they are often in need of external capital precisely when it is most expensive to obtain).

3. It wastes no capital. Unlike an index-based hedge, you are not purchasing any high-quality small caps or low-quality large caps in the process.

At GMO, we have run such portfolios on and off since 2004, learning many a lesson on how to manage them over the years.13 Our most recent offering, the Quality Spectrum Strategy, is 175% long large, high-quality stocks and 75% short smaller, junkier stocks. Since our Focused Equity team launched the strategy in 2019, it has delivered on its promise to keep up with equity markets over time while protecting capital during particularly stressful equity periods, as shown 13 in Exhibit 11. At the risk of sounding a little shameless, this is exactly the type of portfolio we The trickiest problem of managing a long high-quality/ believe allocators with private equity and venture capital exposure should be seeking.14 short low-quality portfolio is the fact that the short book is massively more volatile and higher-beta than the long book. The rebalancing drag of a long/short portfolio rises as a function of the volatility of the short book relative to the long book in dollar terms, so a dollar-neutral quality vs. junk portfolio suffers much higher rebalancing drag than one where the high-quality long book is larger than the junk short.

Or a version of it that targets a beta of zero (which we are also happy to run).

Part 1: What Barbarians Like to Take Private | p11 EXHIBIT 11: QUALITY SPECTRUM PERFORMANCE IN DISTINCT SCENARIOS S&P 500 Worst 10 Months Since 2019 0.0% -2.0% -4.0% -6.0% -8.0% -10.0% -12.0% -14.0% Quality Spectrum Relative Returns to S&P 500 on Same 10 Months 16.0% 12.0% 8.0% 4.0% 0.0% -4.0% -8.0% Source: GMO Conclusion It can be tempting to treat private equity as a separate animal from stocks and confine your worries to liquidity considerations rather than true economic exposures. Given the reality of stale marks and generally short or shallow recessions in most of its history, it can seem as though private equity’s downside is not losses but occasional inconvenience. But while the true downside of risk assets doesn’t bite that often, the art of building a good long-term portfolio requires an honest look at what risks that portfolio really takes.

Our analysis of 45 years of LBO data tells us in part what most investors should already know (that private equity portfolios skew massively smaller than public equity indices), as well as some things that we, at least, were not yet aware of: that private-equity-held companies skew less profitable than public equivalents, that their industry concentration in software today is much higher than any concentration they have held in any industry in the past, and that their leverage is high even before layering on the extra debt from a buyout. This means portfolios with significant private equity exposure face a large downside economic risk that is not obviously compensated by the market. While private equity managers have historically delivered significant alpha relative to the small-cap and low-quality company universes they select from, that’s no reason to passively accept a substantial bias toward a riskier universe of companies.

Part 1: What Barbarians Like to Take Private | p12 Ben Inker Owning a purely large cap public equity portfolio may mitigate private equity risk to some Mr. Inker is Co-Head of extent, but PE’s massive, small-cap bias cannot simply be diversified away. To reduce the GMO’s Asset Allocation small-cap bias of a portfolio with significant private equity exposure, it is necessary to actively team and a portfolio short small caps. Such a hedge could be built out of long megacap, short small-cap passive manager for the team’s exposures, but it would likely be inconvenient to manage. We believe that inefficiencies in products. Mr. Inker is a the public markets when it comes to high- and low-quality stocks suggest an active hedge member of the GMO Board of Directors and could be achieved that has better risk characteristics, more capital efficiency, and a positive a partner of the firm. He joined GMO in 1992 expected return. Either way, allocators would do well to manage the large factor bets they are following the completion of his bachelor's degree in Economics from Yale University. In taking in their equity portfolios. his years at GMO, Mr. Inker has served as an analyst for the Quantitative Equity and Asset Allocation teams, as a portfolio manager of several equity and asset allocation portfolios, AVERAGE ANNUAL TOTAL RETURN (NET) IN USD as Co-Head of International Quantitative As of 3/31/2026 Equities, and as CIO of Quantitative Developed Equities. He is a CFA charterholder.

Inception 1-Year 3-Year 5-Year Since Inception Quality Spectrum John Pease Composite 11/30/2019 3.75% 15.09% 15.87% 13.91% Mr. Pease is the Head of Asset Allocation 50% ACWI 50% 3Mo Research on GMO’s TBill Blend 12.02% 10.81% 6.72% 7.40% Asset Allocation team and a partner of the firm. Prior to joining GMO full-time MSCI ACWI 20.01% 16.58% 9.49% 11.39% in 2016, Mr. Pease was an intern with the Asset Allocation team. Mr. Pease earned a bachelor’s in Economics and a master’s Performance data quoted represents past performance and is not predictive of in Economics from Pontificia Universidade future performance.

Catolica do Rio de Janeiro.

Net returns are presented after the deduction of a model advisory fee and incentive fee if applicable. These returns include transaction costs, commissions and withholding taxes on foreign income and Disclaimer capital gains and include the reinvestment of dividends and other income, as applicable. Fees paid The views expressed are the views of Ben by accounts within the composite may be higher or lower than the model fees used. GMO LLC claims Inker and John Pease through the period compliance with the Global Investment Performance Standards (GIPS®). A Global Investment ending May 2026 and are subject to change Performance Standards (GIPS®) Composite Report is available on GMO.com by clicking the at any time based on market and other GIPS® Composite Report link in the documents section of the strategy page. GIPS® is a registered conditions. This is not an offer or solicitation trademark owned by CFA Institute. CFA Institute does not endorse or promote this organization, nor for the purchase or sale of any security and does it warrant the accuracy or quality of the content contained herein. Actual fees are disclosed in should not be construed as such. References Part 2 of GMO's Form ADV and are also available in each strategy’s Composite Report. The portfolio to specific securities and issuers are for is actively managed, is not managed relative to a benchmark and uses an index for performance illustrative purposes only and are not intended comparison purposes only and, where applicable, to compute a performance fee. to be, and should not be interpreted as, recommendations to purchase or sell such securities.

Copyright © 2026 by GMO LLC.

All rights reserved.

Part 2: QUARTERLY LETTER 2Q 2026 LETTER TO THE INVESTMENT COMMITTEE ON PRIVATE EQUITY The Persistent Dream of Performance Persistence Ben Inker | Asset Allocation My day job at GMO does not directly involve private equity beyond being an observer. But I do wind up discussing private equity reasonably regularly, both with investment committees that I serve on and when invited to speak to the investment committees of other institutions. And in those situations, I’ve started to notice something a little jarring that may not be as obvious to investment committee members who only experience the performance of one or two institutions.

It is well known that private equity has failed to keep up with the public markets over the last several years. But I also seem to be hearing from a number of institutions that the performance of their particular PE portfolio, which in the past might have done substantially better than the Preqin, Cambridge Associates, or other composite, no longer seems to be doing so. There is usually an excuse that feels specific to the institution in question—“we focused too much on co-investment opportunities and failed to keep a high enough bar on our expectations for the actual fund performance,” or “we were too slow to react to our GPs’ loss of focus and mission creep.”

The implication of those explanations is that fixing a particular problem they diagnose will lead to better relative performance in the future. But there is another explanation for this phenomenon that is less fixable and feels awfully plausible to me: if the persistence of performance for PE managers has gone away, or even significantly deteriorated, the performance difference between the best institutional PE portfolios and the mean is doomed to collapse to low levels.1 For private equity allocations predicated on a belief in the investment There will still be a fair bit of performance dispersion, staff’s ability to find and secure the very best private equity managers, such an explanation since most investors invest with a relatively small number would call into question the rationale for the allocation in the first place. of PE managers, and there will still be plenty of variability in actual fund returns. But without persistence of returns, that variability will wind up mostly owing to chance, and The original handbook for the endowment model, David Swensen’s Pioneering Portfolio longer-term returns will tend to converge. Management (2009), made no claims about an inherent return premium for private equity. 2 While Swensen acknowledged some advantages of private equity in principle—better alignment Paraphrased from the 2009 edition, which made basically the same points as the original 2000 edition (pp. 224–233) with investors, longer time horizons, the focus on operating efficiency that comes along with a with some updated data. greater debt load—he pointed out that private equity also suffers from high fees, principal-An illiquidity premium for leveraged buyouts (LBOs), agent problems, and the tendency for successful managers to raise ever-larger funds only at least, never made any sense in the first place. If for them to underperform their earlier, smaller ones. He concluded that private equity was you voluntarily take a public company private and pay riskier than public equities due to its high leverage and, to the best of his knowledge, achieved a premium to do so, there is no plausible mechanism by which you could possibly get paid for taking on the disappointing median returns over its history (pp. 220–235).2 The case for private equity, illiquidity. The illiquidity might be a means to an end for rather than resting on some vague “illiquidity premium,”3 was all about finding extraordinary some other mechanism to achieve higher returns, but the idea that you would generally get paid for the fact that the managers. He believed private assets were a good place to do that, given their much wider asset is no longer liquid is just silly when the illiquidity is range of performance across managers relative to public equities or fixed income. entirely self-imposed.

Part 2: Letter to the Investment Committee on Private Equity | p14 In practice, generating this alpha for an institution would involve finding extraordinary portfolio managers or firms who can consistently outperform their peers. So the first question any investment committee should ask when discussing an allocation to private equity or any other private asset is: what makes us confident we can find these extraordinary managers and get meaningful allocations to their funds? If the committee can’t credibly answer that question, it makes little sense for them to try to replicate the asset allocation of institutions that can. But even for institutions that have reason to claim such a selection ability, private equity fund performance really needs to be significantly persistent for the game to work. And it is far from clear that such persistence exists.

Several academics have done interesting work on the topic, noting that persistence of performance has fallen notably since 2000, and more so for private equity than venture capital.4 A particularly relevant finding is that the interim performance of funds that have not completed their life cycles is entirely unhelpful in predicting future fund returns, a real problem since those are the only returns recent enough to feel relevant when considering a manager’s next fund. While we all know “past performance is not indicative of future results,” it is extremely hard to overstate how central past performance is to investors' decision-making when choosing private asset managers. You are buying into a blind pool, and almost the only thing you know is what the manager did in the past.

While the performance of the investments in that previous pool is not the only thing you can analyze, it feels like the most salient piece of data there is. But what if that is an illusion? A mature private equity portfolio will consist of multiple funds from multiple managers, so the total number of different funds owned by an institution will generally be pretty large, easily a couple of dozen or more, even if the institution has relationships with a relatively small number of firms. If there truly is little persistence in private equity fund returns, it implies that even if the range of returns between the best- and worst-performing funds remains large, the aggregate returns for an institution will almost always be close to the median. The chart below shows the implied alpha of a diversified PE portfolio across several levels of performance persistence (Braun, Jenkinson, and Stoff 2017).

EFFECT OF PERFORMANCE PERSISTENCE ON EXPECTED PE AND VC ALPHA 3.5% 2.9% 3.0% 2.5% 2.3% 1.9% 2.0% 1.5% 1.20% 1.3% 1.0% 0.55% 0.5% 0.26% 0.03% 0.0% Pre-2001 Full Sample full sample Post-2000 Post-2000,but But Amazing at amazing at new New 4 Managers managers I’m not going to pretend to give a comprehensive listing Expected PE Alpha Expected VC Alpha of the research, but a couple of studies that stood out to me included Braun, Jenkinson, and Stoff (2017), which looked at performance by deal rather than by fund, Source: Braun, Jenkinson, and Stoff (2017) helping to abstract away from some of the fund return Assumed alpha for quartiles of performance is 8%/3.5%/-3.5%/-8% for PE and 12%/4%/-4%/-12% for calculation problems; and Harris, Jenkinson, Kaplan, and Stucke (2023), which looked at the problem of interim VC. “Amazing at New Managers” assumption is 40%/30%/20%/10% odds of new managers being in performance calculations that investors are forced to rely each alpha quartile, and 20% of assets in PE/VC invested in such new managers. on given the long lives of funds.

Part 2: Letter to the Investment Committee on Private Equity | p15 Ben Inker I’ve put the venture capital results in as well. While there was basically no evidence of Mr. Inker is Co-Head of persistent performance in the post-2000 sample for private equity, venture capital did show GMO’s Asset Allocation a decent amount of persistence, even if it, too, shows substantially less persistence than team and a portfolio the early sample. I added a fourth column in which I made a friendly assumption about the manager for the team’s new funds that an institution hires. I assumed that the institution had an amazing record in products. Mr. Inker is a backing new managers, and that those new managers had a 40%/30%/20%/10% chance of member of the GMO Board of Directors and being in the 1st through 4th quartiles of performance. I further made the (probably insanely a partner of the firm. He joined GMO in 1992 friendly) assumption that the institution’s full 20% PE or VC allocation was invested in such following the completion of his bachelor's degree in Economics from Yale University. In funds (such an institution could still not expect very much alpha from a PE portfolio, though 55 his years at GMO, Mr. Inker has served as an basis points is a whole lot better than the 3 basis points of implied alpha for an institution that analyst for the Quantitative Equity and Asset simply reupped with its strongest performers).

Allocation teams, as a portfolio manager of It’s possible I’m being unfair in assuming that the basic due diligence in choosing to invest several equity and asset allocation portfolios, as Co-Head of International Quantitative in the new funds of current managers is to look at the interim performance of their previous Equities, and as CIO of Quantitative funds, but for institutions whose current alpha relative to the PE composite does not look Developed Equities. He is a CFA charterholder. particularly impressive, I think it’s fair to ask why you think it will get better in the future.

I’m not trying to make the case that institutions should abandon private equity. Actually, if one believes, as I do, that private equity is choosing from a small, junky group of firms, the Disclaimer industry's performance has been somewhat better than it looks over the last decade.5 I also The views expressed are the views of Ben believe that investing skill exists,6 and that it makes sense for well-resourced institutions to Inker through the period ending May 2026 and are subject to change at any time based on invest with private equity managers they truly have high conviction in. The difference between market and other conditions. This is not an the best and worst performers among private equity funds remains large, and an institution offer or solicitation for the purchase or sale that can truly tilt the odds in favor of top-quartile results will reap substantial benefits. of any security and should not be construed But the bar to invest in a private equity manager should be high—arguably even higher than it is as such. References to specific securities for active public asset managers, since you’ll be stuck paying PE managers high fees for a long and issuers are for illustrative purposes only and are not intended to be, and should not be time, even if you lose conviction in the interim. And if individual fund allocations truly do have interpreted as, recommendations to purchase a high bar, a target PE allocation may not even make sense (at least not beyond establishing or sell such securities. an upper limit).

Copyright © 2026 by GMO LLC. If, for example, you target 25% of your portfolio in U.S. public equities and can only come up All rights reserved. with 10% worth of allocations to active managers you truly believe in, you have the option to allocate the other 15% passively. That passive option is not available to you in private equity. If you max out on high-caliber PE managers short of an overall allocation target, you will wind up investing the rest of your allocation in managers you have less confidence in. Paying high fees to managers you have less confidence in is unlikely to be a good use of capital.

How can the investment committee help? I think a good start would be for the investment committee to ask the investment staff to discuss their beliefs about each asset class in which 5 the institution invests, the purpose each serves in the portfolio, how much (if any) alpha they See part 1, What Barbarians Like to Take Private, for evidence of a small, low-quality bias in private equity. expect to add in each asset class, and, crucially, how they intend to test those beliefs over 6 time. They should document their beliefs for each asset class and compare them periodically, Admittedly, I’m highly likely to be biased toward such a belief. perhaps every three to five years.7 The risk in doing this is that it just turns into a referendum At the end of the day, the role of the investment committee is to help the investment staff do on which assets have done well or badly in the trailing a better job managing the portfolio. That should not be about second-guessing individual period, which would be a profound mistake. There is already too much performance chasing in the investment manager decisions, but pushing the investment staff to think critically about what they do world. But putting your beliefs down on paper is extremely and why it is absolutely in the committee’s wheelhouse. Private equity programs are not important to avoid the narrative creep that it is all too easy meant to run on autopilot; there are critical questions to answer and, for many institutions, to fall into. If ”private real estate is a great place to add alpha” turns into “private real estate is an inflation hedge,” disappointing results to grapple with. then into “private real estate is an under-owned asset class,” and so on—each rationale replacing the last as the thesis fails to play out—while the target allocation remains fairly static, something has gone very wrong.

Part 2: Letter to the Investment Committee on Private Equity | p16 References Braun, R., Jenkinson, T., & Stoff, I. (2017). How persistent is private equity performance? Evidence from deal level data. Journal of Financial Economics, 123 (2), 273–291.

Harris, R.S., Jenkinson, T., Kaplan, S.N., & Stucke, R. (2023). Has persistence persisted in private equity? Evidence from buyout and venture capital funds. Journal of Corporate Finance, 81 (102361).

Swensen, D. (2009). Pioneering Portfolio Management: An Unconventional Approach to Institutional Investment, Fully Revised and Updated. Free Press.