Why IRR can be misleading

01 The number private equity loves

Private equity has many performance metrics, but one number still receives more attention than almost everything else: IRR. It sits in pitchbooks, fund reports, investment committee memos and track record tables. It is clean, annualised and easy to compare at first glance. A 25% IRR looks better than a 15% IRR. That is the attraction.

The problem is not that IRR is fake. It is not. The problem is that IRR answers a narrower question than many readers think. It tells you the discount rate that equates a stream of cash flows to zero net present value. It is highly sensitive to when money leaves the investor's pocket and when it comes back.

That sensitivity is useful in private markets because cash flows are irregular. Capital is not deployed on day one. It is called over time. Assets are sold when buyers exist, not when a spreadsheet wants them to be sold. Distributions can come from exits, refinancings, dividend recaps or secondary transactions. In that world, a metric that accounts for timing is necessary.

But the same feature that makes IRR useful also makes it dangerous when used alone. Speed can look like skill. Early cash can make a modest multiple look impressive. Delayed capital calls can improve LP-level IRR without changing the underlying asset. Unrealised marks can keep the number alive long after cash distributions have slowed.

This matters more in 2026 than it did during the easy money years. Distributions are under pressure, holding periods have stretched and LPs are asking a more basic question: how much cash actually came back? In that environment, an elegant annualised percentage is not enough.

02 What IRR actually measures

IRR is the rate r that makes the net present value of an investment equal to zero. In simple terms, it is the annualised discount rate that balances the cash paid in with the cash received back.

Technically, IRR is a money-weighted return. That means the size and timing of each cash flow matter. If a large distribution comes early, it has a powerful effect. If a capital call is delayed, the investor's measured holding period is shorter. If a fund holds a large residual NAV for years, the result depends heavily on what that residual value eventually becomes.

This is why IRR is standard in private markets. Public funds can usually be measured with time-weighted returns because external cash flows are often outside the manager's control. Private equity funds are different. The GP controls capital calls, exits and distributions. A money-weighted measure is therefore relevant.

There is a catch. IRR compresses a full cash flow history into one percentage. The more irregular the cash flows, the more that single percentage hides. Two funds can have the same IRR and very different cash outcomes. Two deals can have very different IRRs even if the one with the lower IRR created more absolute value.

That distinction sounds technical. It is not. It is the difference between a fund that quickly returns part of your money and a fund that compounds a larger amount over a longer period. Both can be attractive. They are not the same investment experience.

03 The speed illusion

The cleanest way to see the issue is to compare two simple deals. Deal A doubles the money over five years. Deal B returns 1.5x over two years. Most people instinctively prefer the double, because 200 is more than 150. IRR prefers the faster outcome.

Mathematically, Deal B has the higher IRR. Economically, Deal A created more total value. The answer depends on what happens after year two. If the investor can redeploy the 150 from Deal B into another attractive opportunity, the shorter deal may be excellent. If the cash sits idle, or is redeployed into weaker opportunities, the headline IRR overstates the practical value of that speed.

This is where the reinvestment point is often explained badly. The issue is not that IRR mechanically assumes every interim distribution is reinvested at the same IRR in every possible interpretation. The safer and more relevant point is this: early cash creates a redeployment problem for the investor. A high short-period IRR only becomes powerful if the capital can keep working after it comes back.

Private equity professionals know this. LPs know it too. That is why multiple-based metrics remain important. A 3.0x fund with a lower IRR may still be more valuable to a long-term investor than a quick 1.4x fund with an impressive annualised figure. Speed is good. But speed is not the same as total value creation.

04 The early distribution effect

The same mechanism becomes more subtle when a deal returns cash early but leaves only a modest total gain. Consider a simple cash flow profile: invest 100, receive 90 after one year, then receive another 40 in year five. Total cash received is 130. The multiple is 1.3x. Not terrible, but hardly spectacular.

The IRR is approximately 13.8%. That is not a fake number. It is simply telling you that most of the capital came back quickly. The investor had less money at risk after year one, so the annualised return looks much better than the 1.3x multiple might suggest.

This is common in private equity. A company may be refinanced after a good first year. A dividend recap may return a large part of the sponsor's equity. A partial exit may crystallise early proceeds while the fund keeps a residual stake. None of these are automatically bad. In some cases, returning capital early is disciplined portfolio management.

But the presentation matters. A deal with a 13.8% IRR and 1.3x MOIC is a very different outcome from a deal with a 13.8% IRR and 2.0x MOIC. The percentage is identical. The wealth creation is not.

This is why sophisticated LPs look at distributions, not only rates. They want to know whether the fund returned meaningful cash or whether the number is being flattered by timing, leverage, interim marks or a small early realisation.

05 The subscription line effect

Subscription lines add another layer. A subscription line is a short-term credit facility secured against LP commitments. Funds use them for many legitimate reasons: operational flexibility, faster execution, smoother capital call administration and liquidity management. The problem is that they can also change reported LP-level IRR.

Imagine a fund buys an asset for 100 and sells it for 200 after five years. At the asset level, the result is simple: 2.0x MOIC and approximately 14.9% IRR.

The company did not become better. The purchase price did not change. The exit price did not change. The operating improvement did not change. What changed is the starting point of the LP cash flow clock.

From the LP's perspective, the capital was called one year later and returned in year five. The measured duration of invested capital is shorter, so the LP-level IRR improves. That is why regulators have paid attention to whether performance is shown consistently when subscription facilities are used. The issue is comparability, not the existence of the facility itself.

A subscription line can be a useful treasury tool. It can also make two funds with identical asset-level performance look different at the LP level. If one fund calls capital immediately and another delays the call, the second fund may report a higher IRR even though the underlying investment made the same money.

06 Why DPI changes the conversation

IRR becomes much less dangerous when it is read next to cash multiple metrics. The most important one in today's market is DPI, or distributions to paid-in capital. DPI measures how much money has actually been distributed back to LPs relative to how much they have paid in.

This cuts through a lot of noise. A fund can report a strong IRR and still have weak DPI if much of the performance sits in unrealised NAV. That does not mean the NAV is wrong. It means the value is still a mark, not cash. Until the asset is sold or distributed, the LP cannot recycle that capital into new funds, meet spending needs, or show realised liquidity to its own stakeholders.

TVPI, DPI and RVPI split the story into three pieces. TVPI is total value to paid-in capital. It combines realised distributions and remaining value. DPI is the realised part. RVPI is the residual value still held in the fund. The equation is simple.

A mature buyout fund with 2.0x TVPI and 0.3x DPI is not the same as a mature buyout fund with 2.0x TVPI and 1.6x DPI. The headline total value is identical. The first fund is still mostly a promise. The second has returned a large amount of cash.

This is the reason DPI has become a sharper metric in a slower exit environment. When distributions are scarce, LPs stop treating paper value and cash value as interchangeable. They want to see realisations. They want to see whether the fund can turn marks into money.

07 What investors should actually look at

The answer is not to throw away IRR. That would be lazy. IRR remains useful because private markets have irregular cash flows and because timing genuinely matters. A 2.0x in three years is not the same as a 2.0x in ten years. Annualisation has value.

The mistake is to let IRR carry the full argument. A serious performance review should separate four questions.

The better question is not simply: what is the IRR? The better question is: how much cash came back, how fast, and how much of the remaining value is still just a mark?

That question forces the discussion back to economics. A high IRR with low DPI may be early, promising or inflated by timing. A lower IRR with high DPI and a strong multiple may be a more useful outcome for an LP that needs liquidity. A high TVPI with high RVPI may still become a great fund, but the conclusion should wait until exits prove the marks.

Private equity does not need fewer metrics. It needs less worship of a single one. IRR is a good tool when it is placed in context. Alone, it can make timing look like performance. Beside MOIC, TVPI, DPI, RVPI and PME, it becomes what it should have been all along: one lens, not the verdict.