Comparing Fund Performance Across Vintage Years

Comparing the IRR of a private equity fund launched in 2008—right before the financial crisis—to one launched in 2012 is misleading without accounting for vintage year context. Funds launched in different economic cycles encounter radically different exit opportunities, valuations, and leverage environments. Vintage-year adjusted benchmarks correct for this timing bias and allow apples-to-apples performance comparison.

Why Vintage Year Distorts Direct Comparison

Imagine two buyout funds with identical management teams and stated investment thesis. Fund A closed in 2006 and deployed capital into a hot market, acquiring targets at 8× EBITDA with easy leverage. Fund B closed in 2010 and acquired identical-quality targets at 4.5× EBITDA with stricter underwriting. Fund A exited during the 2013–2015 bull market at 12× EBITDA; Fund B exited in 2018–2019 at 10× EBITDA.

On raw IRR, Fund A may report 25% and Fund B 18%. But Fund B did not underperform—it operated in a harder market at entry and exit. The difference is almost entirely timing, not skill. Vintage-year adjusted comparison isolates the GP’s actual investment acumen from the macro tailwind or headwind the fund received.

Timing affects every stage of the fund lifecycle:

• Entry valuations: Funds closing in bull markets buy at peak multiples. Funds closing post-crisis buy at trough valuations, giving them easier return math later.
• Leverage terms and cost: A 2007 fund could use 6–7× leverage cheaply. A 2010 fund faced 3–4× with higher spreads. This alone shifts return profiles by 200–400 basis points of IRR.
• Exit window: A fund maturing in 2014–2016 faced record M&A activity and IPO windows. A fund maturing in 2020 faced COVID lockdowns. Exiting when buyers are flush versus scarce changes returns dramatically.
• Valuation compression or expansion: Sector multiples, cap rates in real estate, or credit spreads may shift. A fund holding through a period of multiple expansion sees free return; one in contraction faces drag.

Without vintage-year adjustment, comparing raw IRRs across different fundraising years is nearly meaningless for assessing GP skill.

Vintage-Year Cohorts and Benchmark Grouping

The solution is to group funds by vintage year and benchmark within cohorts. The most common approach:

Tight cohort: Compare funds within the same vintage year (e.g., all 2015 U.S. buyout funds). This is the cleanest but smallest sample and can be noisy with few funds.

Sliding window: Compare within a 2–3 year range (e.g., 2014–2016 vintages). This balances sample size and homogeneity, as economic conditions within a 3-year window are usually reasonably similar.

Sector and geography controls: Even within a vintage year, a mid-market buyout fund in Texas operates in a different environment than a tech-focused fund in Silicon Valley. Proper benchmarking narrows further: “U.S. midmarket buyouts, 2015–2017 vintage.”

The vintage-adjusted framework then allows the question: “Among all 2015 vintage North American buyout funds, did this GP’s fund place in the top quartile?” That is a meaningful signal of skill.

How Vintage Bias Amplifies With Economic Cycles

The vintage effect is most dramatic across recession-expansion boundaries. Funds closed just before a downturn perform much worse in raw IRR; those closed right after outperform with easier entry prices and subsequent recovery tailwinds.

2005–2007 funds (pre-crisis): Deployed at peak leverage and valuations. Many targets acquired at 8–10× EBITDA sold at 8–9× EBITDA post-crisis without multiple expansion gains. Returns were heavily dependent on operational improvement and leverage paydown, leaving less room for error. Median net IRR typically 12–16%.

2009–2011 funds (crisis/trough): Deployed at recession-low valuations and could use strategic leverage post-Dodd-Frank reset. Benefited from multiple expansion and operational recovery throughout the 2010s. Even mediocre GPs saw solid returns. Median net IRR typically 18–24%.

2014–2016 funds (peak): Deployed into a hot market with elevated entry multiples (9–11× EBITDA), but faced a long runway to exit. Returns depend heavily on operational performance and multiple selection, not macro timing. Early exits posted strong returns; late exits faced 2020 COVID pressures. Median net IRR typically 14–20%.

A 2010-vintage fund posting 20% IRR is not necessarily better than a 2006-vintage fund posting 15% IRR. Once vintage is held constant, the comparison becomes valid.

Adjusting for Vintage: Common Methodologies

Raw IRR within cohort: The simplest approach—sort all vintage-2015 funds by net IRR and rank them. The limitation: IRR itself still has some skew based on exit timing and J-curve effects.

MOIC (Multiple on Invested Capital) with time weighting: MOIC measures the multiple of capital distributed back to LPs relative to capital called. MOIC is less sensitive to timing of distributions, though still somewhat affected by entry-to-exit pace. Some benchmarkers weight MOIC by age (younger funds are allowed more time).

Vintage-year PIPs (Pooled IRRs): Rather than averaging individual fund IRRs, some benchmarkers calculate a Pooled IRR across all funds in a vintage cohort. This weights each fund equally (or by AUM) and shows the blended cohort return. GPs can then see how they rank within the statistical distribution.

PME (Public Market Equivalent): The most sophisticated methodology. PME calculates what an LP’s money would have returned if invested in the public equities market instead. It adjusts for both vintage year and J-curve timing. A fund’s outperformance above PME is called Alpha. This is increasingly the standard used by large LPs.

The J-Curve and Early vs. Late Vintage Performance

One more wrinkle: when you measure a fund’s return matters. Early in a fund’s life, reported IRR is often low or even negative—capital is being called and invested, returns are not yet realized, and fees are dragging returns down. This is the “J-curve” effect. A 2-year-old fund might show a -2% IRR; a 7-year-old fund from the same vintage might show +15%.

When comparing a mature 2012-vintage fund to a younger 2018-vintage fund, the raw IRR difference is partly just lifecycle stage, not performance. Adjusted comparisons account for this by either:

• Measuring IRR at the same point in each fund’s life (e.g., all funds measured at Year 5)
• Using NAV estimates to project returns to maturity
• Comparing dry powder and DPI (Distributions to Paid-In capital) rather than IRR

Practical Use: What LPs Actually Do

Large LP institutions (university endowments, pensions, sovereign wealth funds) use vintage-year adjusted benchmarking as a gate for GP evaluation:

• Initial screen: Did the GP’s 2015 vintage fund beat the median 2015 vintage cohort? If yes, advance to deeper due diligence. If no (or worse), pass.
• Vintage cohort spanning: Strong LPs compare the GP’s last 3–4 vintages to understand consistency. A GP with one great vintage (2012) but two mediocre ones (2010, 2014) is likely a lucky picker, not a skilled operator.
• Sector and geography weighting: A North American buyout GP is not being compared to a Chinese infrastructure fund, even if they closed in the same year. Cohorts are narrowed to apples-to-apples comparables.

Data providers like Cambridge Associates, Preqin, and Burgiss publish annual benchmark reports grouped by vintage year, allowing LPs to see exactly where a fund percentile ranks and how the GP’s track record stacks up.

See also