Private Markets: The BS in IRR

The Most Quoted Metric Isn't What You Think

Apollo’s annual report states that its private equity funds have generated a 39% gross internal rate of return (IRR) since 1990, 24% after fees, "on a compound annual basis."¹ Take the firm at its word. Its first fund raised roughly $400 million. Compound that at the after-fee figure for 36 years and the single fund would be worth about $900 billion today, essentially the entire $938 billion the firm managed at the end of 2025, after four decades of fundraising.² Compound it at the gross figure instead and you get $56 trillion, roughly half of what the world economy produces in a year.³ Apollo has run excellent funds. It has not done that.

Numbers like these contributed to the belief that private equity dominates every other asset class. This misconception has pulled trillions of institutional dollars into these funds, and the same numbers now headline the brochures arriving in front of individual investors, the industry’s declared next frontier. Ludovic Phalippou, an Oxford professor and one of private equity’s most persistent academic critics, has been documenting the metric’s flaws since 2008. His 2024 paper, The Tyranny of IRR, argues that the misuse of since-inception IRRs helped fuel both the institutional flood into private equity and its accelerating push into retail portfolios.⁴

Let’s walk through some of his key arguments.

Not a Rate of Return

When an index fund reports a compound annual growth rate (CAGR) of 10% for a decade, the outcome is straightforward: a dollar left in it grew at 10% per year, compounding annually. An IRR answers a different question. It is the discount rate that makes a fund’s net present value (NPV) equal to zero. In laymen’s terms, it measures how fast money grew only while the manager happened to be holding it. CAGR and IRR should not be treated equally. In fact, they’re vastly different.

Let’s use a hypothetical example to show the distinction. You commit $100. The fund does not take it all at once; it "calls" $25 a year for four years as it finds companies to buy. Distributions begin in year 4 and total $200 by year 9. Your money doubled in nine years, and a dollar that becomes two over nine years compounded at 8.0% per year. That’s CAGR. The same cash flows produce a reported IRR of 16.2%. Both numbers are computed correctly, yet only one is the rate at which your wealth grew.

How can a fund report 16.2% while its investors compound at 8.0%? The answer is in what the formula assumes. An IRR treats every dollar the fund returns as if that dollar kept earning the IRR after it left. The $30 you receive in year 4 is presumed to earn 16.2% for the rest of time; so is every check after it. In reality, the money lands in your account and waits, and nothing waiting in an account earns 16.2%. The formula assumes away the hardest problem in private markets, which is finding the next investment. Nor is the gap an artifact of assuming your cash sits idle: invest the waiting dollars at 4% in Treasury bills and your wealth compounds at 9.8% rather than 8.0%. The 16.2% stays out of reach unless every returned dollar re-earns 16.2% the day it lands.

KKR’s fiscal 2025 annual report runs the same experiment at full scale. ⁵The table below covers every private equity and real assets fund the firm has raised since 1976 and reports a since-inception IRR of 25.5% before fees, 18.6% after, implying they’ve generated 18.6% annual returns for their investors since 1976. If that number meant what it seems, $1 billion entrusted to those funds in 1976 would be worth $4.3 trillion by 2025. The IRR figure assumes every dollar returned kept re-earning 18.6% elsewhere. It did not. The filing itself shows where recycled dollars would actually have landed: the funds raised since 1997 earned 12.2% net, not 18.6%. No investor turned a billion into trillions, and we’ll soon see that 12.2% net figure itself is deceptive.

The Immortal Number

A rate that is not a rate is misleading. A rate that can never fall is a marketing department’s dream. Since inception IRR discounts every cash flow back to day one, which hands the earliest profits a vastly disproportionate weight. Think of a grade point average in which freshman year counts disproportionately more than subsequent years. After a strong enough start, the following years have a faint impact on the average.

We see this effect in KKR’s table above. Of the $221 billion its funds invested since 1976, the funds raised before 1997 account for $16.5 billion, about 7% of the total. That 7% earned a gross IRR of 26.1%. The other 93%, everything the firm has invested in the modern era, earned 15.9%. Pool them and cumulative IRR comes out to 25.5%. The headline number an investor reads today describes what 7% of the money did a generation ago.

This is why the figure barely moves. The firm’s own annual reports disclosed since inception IRR of 25.7% in 2013, 25.6% in 2018, 25.5% in 2023, and 25.5% in 2025.⁶ Twelve years, hundreds of billions of dollars of new investing, 0.2% of movement.

To illustrate how early funds set the tone, we built a hypothetical manager and gave it one hot fund: 40% a year for five years. Every fund it raises over the following 25 years earns 8%. Three decades in, its since-inception IRR still reads 35.2%. An investor who put $100 into the first fund and rolled every dollar of proceeds into each successive fund would have $3,683 after 30 years, a compound growth rate of 12.8% per year. The brochure would state 35.2%.

Engineering the Clock

An IRR is computed on the dates cash moves, and in a private fund the manager chooses the dates. Let’s return to our hypothetical fund and consider three versions of a single decision: when to sell the portfolio’s best company. In Scenario 1, the manager sells it in year 6 for $50. In Scenario 2, the same $50 arrives in year 2. In Scenario 3, the manager holds the company through year 9 while it grows at a modest 5% a year, and sells it for $57.90. Every other cash flow is identical.

Scenarios 1 and 2 deliver identical wealth: $200 back on $100 committed, growing at 8.0% a year. Yet the reported IRR jumps from 16.2% to 23.9%, nearly eight points, by moving one cash flow up by four years. Scenario 3 is the uncomfortable one. It hands investors the most money and the fastest wealth growth of the three, and it prints the lowest IRR, because the formula punishes every year the winner is held at less than the fund’s own rate. A manager graded on IRR chooses Scenario 2 every time. An investor would prefer Scenario 3. Quick exits are celebrated in this industry, but you cannot reinvest the proceeds of a quick exit at the stated IRR, which is the very assumption the formula makes.

Borrowing plays the same game from the other end of the clock. Compare two versions of our fund, identical in every investment. One calls capital the ordinary way. The other delays each call by 12 months using a credit line at 6% interest, a practice now standard under the name "subscription line." The delay starts your measured clock later, so the reported IRR rises. The interest, however, is a real cost, so the money you receive per dollar invested falls despite the higher IRR. Without the credit line, the fund reports an IRR of 16.2%. With a six-month delay it reports 17.5%; with a twelve-month delay, 19.3%. KKR’s filing concedes the mechanism in a footnote: the use of such facilities "generally decreases the amount of time that would otherwise be used to calculate IRRs, which tends to increase IRRs when fair value grows over time."⁷ Every reported IRR inherits this flattery, including the 12.2% net that the firm’s modern funds report.

There is one more reason the clock games matter: the manager’s paycheck. The standard arrangement pays a manager 20% of profits once investors have earned a preferred return, typically 8%, and that hurdle is itself an IRR: it accrues on capital only from the day it is called until the day it is returned. A credit line that delays the calls shrinks the preferred return owed. An early exit clears the hurdle while the clock is short. The maneuvers that flatter the reported return also lower the bar for earning the performance fee, which goes some way toward explaining their popularity.

Why It Survives

Which returns us to Professor Frankfurt. His distinction is that the liar knows the truth and works against it, while the BS artist simply does not care. Nothing in an IRR is false. It is audited and standardized, and it endures because nobody in its chain of custody needs it to describe reality. The manager markets it. The allocator carries it to a board. The advisor shows it to clients. The consultant ranks funds with it and builds a business on the ranking, and some investment staff are paid bonuses tied to it. Strathern’s warning is clear: once the measure becomes the target, it ceases to be a good measure.

Asked about the permanence of its number, KKR told Institutional Investor in 2024 that it is "confident that our investors have the requisite information to understand the performance of the funds and strategies they are invested in."⁸ For a pension fund with a floor of analysts, that may be true. The households now receiving private markets brochures do not employ analysts. They are being handed the industry’s most seasoned number with none of the industry’s seasoning.

What We Ask Instead

There is no single honest replacement: private market performance takes more than one number to assess. Start with the multiple on invested capital, net of fees: everything the fund has delivered per dollar in, counting both the cash already returned and the stated value of companies not yet sold. Its blind spots are time, a 2.0x in six years and a 2.0x in thirteen print the same, and that unsold piece, which is an estimate until the last company is sold. Its stricter sibling is DPI, distributions to paid-in, which counts only the cash and gives no credit for the unsold remainder. It is the hardest number in the industry to argue with, though young funds look poor by construction, and a late-life DPI can be dressed up by borrowing against the portfolio rather than selling anything.

The cleanest test is the public market equivalent: run the same dollars, on the same dates, through a plain index fund and divide. Above 1.0, you were paid for the illiquidity and the fees. Below it, you were not. It requires the fund’s full dated cash-flow history, which pitchbooks rarely show. Two repairs exist for the IRR itself. The modified IRR replaces the formula’s reinvestment fantasy with an honest rate on distributed cash, and it is almost never quoted in a pitchbook. And the horizon IRR, the form Phalippou proposes, computes the figure over a rolling 10- or 20-year window, which at least lets the 1980s retire.

The IRR is not a lie. It is BS in Frankfurt’s exact sense: a number produced and repeated with indifference to whether anyone understands what it describes. Our work is translation, converting the pitch into the actual growth a family’s wealth experiences. Ask that of "25.5% since 1976" and the most quoted number in private markets has nothing to say.

Razmig Der-Tavitian, CFA, CAIA
Chief Investment Officer & Managing Partner
Evolve Private Wealth

Evolve Private Wealth Management, LLC (“Evolve”) is a Registered Investment Adviser. The information provided is for educational and informational purposes only and does not constitute investment advice and it should not be relied on as such. It should not be considered a solicitation to buy or an offer to sell a security. It does not take into account any investor's particular investment objectives, strategies, tax status or investment horizon. The illustrative funds are hypothetical, demonstrate calculation mechanics only, represent no Evolve strategy or actual investment, and are not indicative of any achievable result. This content is intended to provide general information about Evolve. It is not intended to offer or deliver investment advice in any way. Advisory services are only offered to clients or prospective clients where Evolve and its representatives are properly licensed or exempt from licensure. Information regarding investment services are provided solely to gain an understanding of our investment philosophy, our strategies and to be able to contact us for further information. All information has been obtained from sources believed to be reliable, but its accuracy is not guaranteed. There is no representation or warranty as to the current accuracy, reliability or completeness of, nor liability for, decisions based on such information and it should not be relied on as such. The views expressed in this commentary are subject to change based on market and other conditions. These documents may contain certain statements that may be deemed forward‐looking statements. Please note that any such statements are not guarantees of any future performance and actual results or developments may differ materially from those projected. Any projections, market outlooks, or estimates are based upon certain assumptions and should not be construed as indicative of actual events that will occur. Past performance is not a guarantee of future results. Any discussion of prospective investment strategies, opportunities, or potential fund launches is provided for informational purposes only and does not constitute an offer or solicitation to invest in any security. Any such offering would be made only pursuant to formal offering documents, which would include additional information regarding risks, fees, and investment objectives.