Chapter 3 Part 3: Measuring PE Performance - IRR, Multiples and Why Benchmarks Are Tricky
So you want to know if a private equity fund is actually good? Turns out, that’s way harder than it sounds. There is no stock ticker refreshing every second. No public quarterly earnings call. You are stuck with imperfect tools and incomplete data. Welcome to Section 3.3 through 3.5 of Demaria’s book.
The Two Main Yardsticks: Multiples and IRR
Private equity performance comes down to two numbers. First is the multiple, technically called TVPI (Total Value to Paid-In). It answers a simple question: for every dollar I put in, how many dollars did I get back?
You put in $1 million. You got back $2.5 million. That is a 2.5x multiple. Easy.
The multiple breaks into two pieces. DPI (Distributed to Paid-In) is the cash you actually received back. RVPI (Residual Value to Paid-In) is what the fund says the remaining portfolio is worth. Add them up, you get TVPI. The DPI is the honest number because it is real cash in your pocket. The RVPI is an estimate. And estimates can be… creative.
The problem with multiples? They ignore time. Getting 2x your money in 3 years is very different from getting 2x in 10 years. A 2x in 3 years is roughly a 26% annual return. A 2x in 10 years? About 7%. Demaria shows a whole table of this from Coller Capital and the numbers are eye-opening.
That is where the IRR (Internal Rate of Return) comes in. It factors in when the cash actually flows in and out. It is the annualized return that makes the net present value of all cash flows equal zero. Sounds fancy, but it basically tells you how fast your money grew per year.
Why Neither Number Is Perfect
Here is the catch. Both metrics have serious flaws.
The multiple is only truly accurate when the fund is fully liquidated. That takes 8 to 15 years. Until then, you are relying on the fund manager’s own valuation of unsold companies. Some managers keep “zombie” companies alive just to make the unrealized returns look better during fundraising. Not great.
The IRR has its own tricks. Fund managers can use credit lines (called “equity bridge financing”) to delay capital calls from investors. This makes it look like your money was working for less time, which pumps up the IRR. In Europe, these credit lines are capped at 12 months. But that can still shave 25% off the investment duration for a typical 4-year LBO deal. The IRR goes up, but you did not actually make more money. In fact, you might make less because credit lines have interest costs.
Even without intentional manipulation, the order of cash flows matters a lot. Demaria gives a great example: two funds both return 2.75x. Same multiple. But if Fund I gets its best deal done first, it shows a 54% IRR. Fund II, same deals in reverse order, shows 29% IRR. Same total outcome, wildly different IRR.
The PME: Adding Context
The third tool gaining ground is the PME (Public Market Equivalent). It asks: what if you had invested the same cash flows into a public stock index instead? Buy the index when the fund calls capital, sell it when the fund distributes.
This is genuinely useful because it gives context. A 15% IRR sounds great until you find out the S&P 500 did 18% over the same period. Or it sounds even better when you learn the market was flat.
But the PME has limits too. Sometimes there is no good public index to compare against. Biotech startups? Distressed debt? Emerging market infrastructure? Good luck finding a matching benchmark. And like the others, the PME is not risk-adjusted.
The J-Curve and Why Patience Matters
If you plot a PE fund’s returns over time, you get the famous J-curve. For the first few years, returns are negative. The fund is paying management fees and making investments. No money comes back yet. Around year 3-4, companies start getting sold. The curve bends upward. By year 5 or so, you might hit break-even. After that, it is (hopefully) profit.
This means you cannot judge a PE fund for at least 5 years. Probably longer. And funds are compared by vintage year, the year they started investing. A 2006 vintage fund faced a completely different world than a 2010 vintage fund.
The Dispersion Problem
Here is maybe the most important thing Demaria highlights. In public markets, picking the “wrong” index fund barely matters. The difference between the best and worst S&P 500 funds is tiny. In private equity, the gap between top and bottom performers is enormous.
According to Cambridge Associates data, the overall PE market (2000-2009 vintages) delivered about 10.9% net IRR and a 1.69x multiple. That beats the S&P 500 PME of 8.4% and 1.54x. But those are averages. The top quartile managers do much better. The bottom quartile can lose money.
So your returns depend heavily on which fund manager you pick. That is why fund selection is such a big deal in PE.
Pitfalls: When Fund Managers Get Greedy
Section 3.4 covers the traps. The biggest one is fund size creep. A successful manager raises bigger and bigger funds. More assets under management means more management fees. But bigger funds need bigger deals, and bigger deals mean different skills, more competition, and often worse returns. Studies show optimal fund sizes are $150-300 million for venture capital and $300-500 million for mid-market LBOs. Beyond that, returns tend to suffer.
Then there is the diversification trap. Big names like Blackstone and KKR started launching real estate funds, debt funds, hedge funds, and energy funds. The brand promises quality. The results often disappoint. The new strategy team might play it too safe to protect the brand, and the best talent in those areas usually already runs their own shops.
Co-investments sound appealing too. Fund investors get to invest alongside the fund at lower fees. But there is a risk of adverse selection: the fund manager picks which deals to offer as co-investments, and it might not be the best ones. Academic research (Fang, Ivashina & Lerner, 2015) shows co-investing does not actually deliver higher returns than just investing in the fund.
Building a PE Portfolio: Top-Down Meets Bottom-Up
Section 3.5 is about how investors actually build a PE program. It is two approaches merging together.
Top-down: You look at the big picture. How much of your total portfolio goes to PE? What types? Which geographies? What risks can you handle? This includes market risk, liquidity risk (you cannot sell PE fund stakes easily), commitment risk (can you answer capital calls?), and contagion risk (what if other investors in your fund default?).
Bottom-up: You evaluate individual fund managers. Their track record, team stability, strategy, deal sourcing. You do reference calls. You read the due diligence package. You ask the hard questions.
The output is a grid. You want to spread your bets across strategies, geographies, and especially across vintage years. Trying to time PE markets does not work (Brown et al., 2019). Instead, invest consistently over at least 5 years and keep going.
Here is the kicker: a PE program itself has a J-curve. Programs under 5 years old show almost no performance. It takes time for the returns to show up. So if your boss asks “how is our PE portfolio doing?” after 3 years and the answer is “not great yet,” that is actually normal.
One last interesting finding from the book. In public markets, past performance does not predict future results. In PE, it actually does, at least partially. Teams that did well tend to keep doing well, because the same skills, networks, and entrepreneur relationships carry forward. But this only holds if the team stays together and keeps the same strategy.
Private equity performance measurement is messy, imperfect, and full of caveats. But now you know the tools and their limits. And that puts you ahead of most people.
Previous: Chapter 3 Part 2: Fund Governance