Benchmarking
Comparing PE fund performance against public markets and peer funds
~25 min read
PE funds do not operate in a vacuum. LPs who allocate to PE are making a choice: instead of investing this capital in public equities (or bonds, or real estate), they are locking it up in an illiquid fund for 10+ years. The fundamental question benchmarking answers is: Was this PE allocation worth it?
Benchmarking PE is harder than benchmarking public equity managers. Public managers can be compared against the S&P 500 on any given day with precise, audited returns. PE funds have irregular cash flows, quarterly (not daily) valuations, and multi-year time horizons. A PE fund cannot be compared to the S&P 500's trailing 1-year return because the capital was deployed and returned over many years, not invested as a lump sum.
This lesson covers the two main approaches to PE benchmarking: peer comparison (how does this fund rank against other PE funds?) and public market comparison (did PE beat what I would have earned in the stock market?).
Public Market Equivalent (PME)
PME is the most rigorous method for comparing PE returns to public market returns. The concept is elegant: it asks, 'What if I had invested and withdrew the exact same dollar amounts, at the exact same times, in a public index instead of this PE fund?'
The most widely used version is the Kaplan-Schoar PME (developed by professors Steven Kaplan and Antoinette Schoar in their 2005 paper). It works as follows:
- Take every capital call the PE fund made to LPs, and instead of sending that money to the PE fund, simulate investing it in the S&P 500 (or another public index) on the same date.
- Take every distribution the PE fund returned to LPs, and simulate selling the equivalent dollar amount from the public index portfolio on the same date.
- At the end of the measurement period, compare the PE fund's remaining NAV to the remaining value in the simulated public index portfolio.
A PME greater than 1.0 means the PE fund outperformed the public index. A PME less than 1.0 means the LP would have been better off in the index fund. The beauty of PME is that it accounts for the exact timing of cash flows, making it a true apples-to-apples comparison.
Kaplan-Schoar PME
Each cash flow (capital call or distribution) is future-valued to the measurement date using the public index return over that period. The ratio of the future-valued distributions plus remaining NAV to the future-valued capital calls gives the PME. A PME of 1.2 means the PE fund generated 20% more wealth than the equivalent public market investment. Most studies find that US buyout funds have historically generated PMEs of 1.15-1.30 against the S&P 500, meaning PE has added real value over public markets for the median fund.
Quartile Rankings and Manager Selection
The PE industry organizes fund performance into quartiles: top quartile (top 25%), second quartile (25th-50th percentile), third quartile (50th-75th percentile), and bottom quartile (bottom 25%). Quartile ranking is the most common way to evaluate a GP's track record.
The dispersion between top-quartile and bottom-quartile PE funds is enormous compared to public equity managers. In public equities, the difference between a top-quartile and bottom-quartile large-cap manager might be 200-300 basis points per year. In PE buyout, the difference can be 1,000+ basis points per year and the gap in MOIC can be 1.0x or more.
This dispersion is why manager selection is the single most important driver of PE returns. Allocating to a top-quartile GP versus a bottom-quartile GP can mean the difference between a 20% net IRR and a 5% net IRR (or worse). It is also why GP track record is the dominant factor in LP due diligence: past performance in PE is more persistent than in almost any other asset class. Top-quartile GPs tend to remain in the top two quartiles in their next fund roughly 35-45% of the time, well above what random chance would predict.
Vintage Year Benchmarking
As we discussed in the J-curve lesson, PE funds must be compared within their vintage year. A 2019 vintage fund is only meaningfully compared to other 2019 vintage funds, because they all deployed capital in the same economic environment and have been investing for the same duration.
Vintage year benchmarks are published quarterly by the major data providers and typically include:
- Median net IRR for the vintage (the 50th percentile fund)
- Upper quartile net IRR (the 75th percentile fund)
- Lower quartile net IRR (the 25th percentile fund)
- Median net TVPI, DPI, and RVPI
When a GP claims that their Fund VII generated a top-quartile net IRR, they are saying that their fund's IRR is in the top 25% of all funds from the same vintage year (and usually the same strategy, e.g., US buyout). LPs verify these claims by cross-referencing against benchmark databases.
Vintage year analysis also reveals market cycles. The 2005-2007 vintages (pre-crisis) generally produced below-average returns. The 2009-2012 vintages (post-crisis) produced exceptional returns. The 2019-2021 vintages are still early but face headwinds from elevated entry valuations and rising interest rates.
| Cambridge Associates | Burgiss | PitchBook | |
|---|---|---|---|
| Data source | Collected directly from GPs and LPs; covers 4,000+ funds | Sourced from LP portfolios (bottom-up); covers 8,000+ funds | Aggregated from GPs, LPs, and public filings; covers 10,000+ funds |
| Strengths | Longest history; gold standard for vintage year benchmarks; extensive PE/VC coverage | Minimal self-selection bias because data comes from LP record-keepers; largest bottom-up dataset | Broadest coverage including small/emerging managers; strong deal-level data |
| Limitations | Potential self-selection bias (better-performing GPs may be more likely to share data) | Coverage may lag for very recent vintage years | Data quality varies for smaller/less-established funds |
| Primary users | Endowments, foundations, institutional consultants | Large pension funds, sovereign wealth funds | Broad market including GPs, LPs, and investment banks |
Survivorship Bias in PE Benchmarks
Survivorship bias occurs when poorly performing funds drop out of the dataset, making the remaining universe look better than reality. In PE, survivorship bias manifests in several ways:
- GPs that fail do not raise subsequent funds. Their track records stop being updated and may eventually be excluded from benchmark calculations. The remaining universe is tilted toward survivors.
- Voluntary reporting. Some benchmark databases rely on GPs voluntarily submitting performance data. GPs with strong track records are more likely to share data than GPs with mediocre or poor results, inflating the reported industry averages.
- Backfill bias. When a GP begins reporting to a database, they may include historical returns from prior successful funds while omitting earlier unsuccessful ones.
Burgiss attempts to address these issues by sourcing data from LP record-keepers (who must report all funds, not just the good ones), creating a more representative sample. Studies comparing Burgiss data to voluntary-reporting databases find that the median PE return in Burgiss is modestly lower, suggesting that survivorship and self-selection biases do inflate reported industry performance by roughly 100-200 basis points.
The practical implication: when someone cites 'PE has outperformed public markets by 300 basis points,' the true premium may be smaller. LPs should use the most comprehensive, least-biased dataset available and apply a healthy discount to industry-wide performance claims.
Performance Persistence: Do Top-Quartile GPs Stay on Top?
One of the most important questions in PE is whether past performance predicts future results. Unlike public equities (where persistence is weak at best), PE shows meaningful performance persistence, though it has declined over time.
A landmark study by Kaplan and Schoar (2005) found strong persistence: GPs whose Fund N was above median had a 60%+ probability of their Fund N+1 also being above median. More recent research by Harris, Jenkinson, Kaplan, and Stucke (2023) found that persistence is still present but weaker than in the 1990s and 2000s, particularly for buyout funds.
The data suggests that top-quartile buyout GPs have roughly a 35-45% chance of remaining top-quartile in their next fund (versus 25% by random chance) and a 55-65% chance of remaining in the top half. This is meaningful but far from guaranteed.
Implications for LPs: track record matters, but it is not destiny. LPs should evaluate team stability (did the partners who generated the track record stay?), strategy consistency, and market conditions in addition to historical returns. A GP whose top-quartile Fund III was driven by two home-run deals may not replicate that outcome in Fund IV.
Kaplan & Schoar (2005), Journal of Finance; Harris, Jenkinson, Kaplan & Stucke (2023), Journal of Financial Economics
Benchmarking PE is as much art as science. PME provides a rigorous public-market comparison, quartile rankings contextualize a fund within its peer group, and vintage year analysis ensures apples-to-apples comparisons. But every benchmark has limitations: survivorship bias, self-selection in reporting, and the fundamental challenge of valuing illiquid assets. Sophisticated LPs use multiple benchmarks from multiple providers and apply judgment rather than relying mechanically on any single number.
In the final lesson of this module, we will look at how all of these metrics are packaged and delivered to LPs through quarterly reporting, capital account statements, and industry-standard templates.
Quiz: Benchmarking
7 questions · ~4 min