IRR (Internal Rate of Return)
The industry's primary return metric, its calculation, its variants, and its pitfalls
~30 min read
If you spend any time around private equity, you will hear 'IRR' more than almost any other term. Internal Rate of Return is the industry's dominant performance metric, used by GPs to market their track records, by LPs to evaluate fund managers, and by compensation committees to determine carried interest payouts. Understanding IRR deeply, including its strengths and its very real limitations, is essential for anyone working in or allocating to PE.
At its core, IRR answers a simple question: What annualized rate of return did this investment generate, accounting for the timing of every cash flow? Unlike a simple multiple (2x, 3x), IRR gives credit for returning money quickly and penalizes investments that tie up capital for long periods. A 2.0x return in 3 years is a roughly 26% IRR. A 2.0x return in 7 years is only about 10% IRR. Same multiple, radically different performance.
What IRR Actually Measures
IRR is the discount rate that makes the Net Present Value (NPV) of all cash flows, both in and out, equal to zero. In PE, the 'cash flows in' are capital calls (money going from LPs to the fund) and the 'cash flows out' are distributions (money returning from the fund to LPs). IRR implicitly assumes that all interim distributions are reinvested at the IRR itself, which is a significant assumption. If a fund generates a 30% IRR, the calculation assumes that every dollar distributed back to LPs is immediately reinvested at 30%, which rarely happens in practice. This reinvestment assumption is one of IRR's most important limitations.
Internal Rate of Return (IRR)
CF(t) represents the net cash flow at time t. Capital calls are negative cash flows (money leaving the LP) and distributions are positive cash flows (money returning to the LP). The IRR is the rate 'r' that makes this sum equal to zero. There is no closed-form solution; it must be solved iteratively (Excel's XIRR function or equivalent). For a simple case: invest $100 at time 0, receive $200 at time 3 years. The IRR is approximately 26%, because $100 * (1.26)^3 is roughly $200.
Gross vs. Net IRR
This distinction is critical and frequently misunderstood.
- Gross IRR is calculated on the fund's investment-level cash flows, before deducting management fees, carried interest, and fund expenses. It measures how well the GP picked and managed investments.
- Net IRR is calculated on the LP-level cash flows, after all fees, carry, and expenses are deducted. It measures what the LP actually earned.
The gap between gross and net IRR is typically 500-800 basis points for buyout funds. A fund reporting a 25% gross IRR might deliver a 17-20% net IRR after fees and carry. When a GP markets a track record, they will often lead with gross IRR. When an LP evaluates a manager, they care about net IRR, because that is the return they actually received.
Always ask: Is this gross or net? The number is meaningless without that context.
Since-Inception IRR vs. Point-to-Point IRR
Since-inception IRR (SI-IRR) measures the fund's performance from its very first cash flow through the reporting date. This is the standard metric for PE funds and the one you will see in quarterly LP reports, pitch decks, and benchmark databases. SI-IRR captures the full history of the fund, including the early years of capital deployment when returns are typically negative (the J-curve, covered in Lesson 3).
Point-to-point IRR, also called horizon IRR, measures performance over a specific time window, for example the trailing 1-year or 3-year period. This is less common in PE than in public markets (where trailing returns are standard), but some LPs use it to evaluate how a fund has performed recently versus over its entire life.
The key insight: a fund with a strong SI-IRR may have actually performed poorly in recent years (or vice versa). Sophisticated LPs look at both to distinguish between 'legacy track record' and 'current execution quality.'
Why IRR Can Be Misleading
IRR is sensitive to the timing of cash flows, not just the amount. This creates opportunities for manipulation and creates situations where IRR does not reflect economic reality.
- Subscription lines of credit. GPs can borrow against LP commitments (via a bank credit facility) to fund deals, delaying the capital call to LPs by weeks or months. This artificially compresses the time between when capital is 'called' and when it is returned, inflating IRR by 200-400+ basis points without generating a single extra dollar of profit. A study by the Institutional Limited Partners Association (ILPA) found that subscription lines boost reported IRR by an average of 3-4 percentage points.
- Early exits inflate IRR. A quick flip, buying a company and selling it within 12-18 months, can generate a very high IRR even if the dollar profit is modest. A $50M investment that returns $75M in one year is a 50% IRR but only a 1.5x MOIC, generating $25M in profit. Meanwhile, a $50M investment that returns $200M over five years is 'only' a 32% IRR but a 4.0x MOIC with $150M in profit.
- IRR on unrealized investments. When a fund marks up a portfolio company that has not yet been sold, that unrealized gain boosts the reported IRR. But the gain is based on an estimate, not actual cash. Many LPs discount IRR figures on funds with high RVPI (unrealized value) until the cash is actually in hand.
Time-Weighted vs. Money-Weighted Returns
This distinction matters for understanding why PE uses IRR while public market managers use TWR.
- Money-weighted return (MWR) is essentially what IRR calculates. It accounts for the size and timing of each cash flow. If more capital is invested right before a period of strong performance, the MWR is higher. If capital flows in right before a downturn, the MWR is lower. MWR reflects the GP's decision about when to call and return capital.
- Time-weighted return (TWR) removes the effect of cash flow timing. It measures what $1 invested at the beginning would have grown to, regardless of when additional capital was added or withdrawn. Public market fund managers use TWR because they do not control when investors add or redeem capital.
PE uses money-weighted returns (IRR) because the GP does control timing. The GP chooses when to call capital (invest) and when to distribute it (exit). This timing is part of the GP's skill set. A GP who deploys capital at low valuations and exits at high valuations should get credit for that timing, which MWR captures and TWR does not.
However, this also means that IRR can be 'gamed' via subscription lines and other timing techniques, which is why the most rigorous LP analysis pairs IRR with MOIC and DPI.
IRR vs. MOIC: Two Funds, Different Stories
Consider two buyout funds, both with $500M in committed capital:
Fund A: Deploys capital quickly, exits early. After 4 years, the fund has returned $1.25 billion to LPs. Net IRR: 28%. Net MOIC: 2.5x.
Fund B: Takes longer to deploy, holds investments 6-7 years. After 8 years, the fund has returned $1.5 billion to LPs. Net IRR: 18%. Net MOIC: 3.0x.
Fund A has the higher IRR, and if you rank by IRR alone, it looks like the better fund. But Fund B returned $250 million more in absolute dollars. If you are a pension fund managing a $50 billion portfolio, those extra dollars matter enormously.
This is not a hypothetical scenario. In the real world, Bain Capital's Fund X (2006 vintage) and Advent International's GPE VII (2012 vintage) illustrate how different hold periods and deployment paces can create divergent IRR vs. MOIC rankings in industry league tables. The lesson: always evaluate IRR and MOIC together.
Cambridge Associates PE Benchmark Reports
IRR is indispensable in PE, but it is not sufficient on its own. In the next lesson, we will cover the family of multiple-based metrics (MOIC, TVPI, DPI, RVPI) that complement IRR and give LPs a more complete picture of fund performance. The best practice is to evaluate these metrics together rather than relying on any single number.
Quiz: IRR (Internal Rate of Return)
8 questions · ~4 min