Hi {{first_name}},

I turned fifty this past weekend. 

My wife and I rented out my favorite bar in St. Louis' Central West End and invited 40 of my closest friends. 

People from both coasts, a couple who flew in from Europe, my family, even got to introduce my parents to some of my favorite people.

My business partners as well. Together we own a majority of this neighborhood through one of our funds. Though not the bar itself, the owner wanted to keep that one. :)  

Looking around, I was struck by the brilliance in that room. Nearly every person there is excellent at something difficult. And I knew that about each one of them.

That evening it occurred to me why. It certainly has nothing to do with anyone’s resume. I have either worked beside these people or watched their work for 20 years, for some of them 30. I have seen where they choose to double down and what they do when things go south. Proximity over decades is how you come to know whether someone is good, and it is the one thing a fund investor rarely gets to see.

When you evaluate a manager, you get a document. Three funds if you are lucky, a deck, and a conversation. You are asked to judge human capability on a fraction of the evidence you would want before trusting someone.

So I went looking for what the research says about that gap. The answer is more uncomfortable than I expected. In private markets the spread between the best and worst managers is the widest in any asset class, and the tool everyone uses to tell which end they are buying barely works.

That is the Selection Problem.

Fair warning, this doubled as diligence for how I run my own capital. I run vehicles built this way. I'd rather tell you that upfront than have it color how you read the rest. In the last two issues I covered what fund size costs you and what the terms take, both printed somewhere you can read before you wire a dollar. Manager skill isn’t written down anywhere. That changes where your diligence hours should go.

This week I'm getting into where those hours could earn their keep:

— Walker Deibel
WSJ & USA Today Bestselling Author of Buy Then Build
Founder, Build Wealth

P.S. That's my business partner in the photo. There's over $1B in annual revenue represented at this pre-party brunch table, built by people I've watched operate for twenty years. That kind of proximity is what this issue is about. And they just might be involved in our next offering. 

Curious about the deal? Sign up to make sure you don’t miss it: Buildwealth.com

Fewer than one third of private equity managers hold top-quartile performance across three consecutive funds.

That surprising figure comes from a KKR investor piece arguing that manager selection is the most critical decision in private equity. The research puts the gap between top- and bottom-quartile buyout managers above 1,400 basis points. In public markets it's roughly 300. Active managers there hold broadly similar portfolios at different weightings, so their returns can't pull very far apart.

Public equity dispersion is measured through eVestment across the fifteen years to December 2024; buyout dispersion through Preqin net IRR quartile boundaries across vintages to 2022.

Three months before publishing that, KKR closed the largest private equity fund ever raised that invests solely in North America, at approximately $23 billion. PitchBook counted 2025 as the year with the fewest US private equity fund closes in more than a decade.

Consider the commercial position. KKR sells access to large buyout funds. A reader who concludes that manager selection is decisive is exactly what the firm wants. The dispersion figure still appears right, and it lines up with every credible estimate in the field.

Read together, the two numbers are an admission. Dispersion describes a spread that has already happened. Persistence is the question of whether any of it can be anticipated, and fewer than one in three across three funds is KKR's own answer to that.

The academic literature has been working on the same question for twenty years. The answer keeps getting worse. 

The Fund Overlap is an Illusion

For two decades the industry's proof of persistence was a single regression. Take a manager's fund, regress its return against the return of that manager's previous fund, read the coefficient. 

Kaplan and Schoar found a positive one in 2005, and it became the intellectual foundation of the top-quartile pitch. 

Arthur Korteweg and Morten Sørensen took that regression apart in the Journal of Financial Economics. Their observation is a mechanical one. 

Private equity firms raise a new fund every two to five years, which means a manager's Fund IV and Fund V spend somewhere between five and seven years running at the same time, in the same market, occasionally holding pieces of the same companies. Two funds that live through identical credit conditions produce correlated returns whether or not anyone running them is any good.

Roughly 44% of the correlation between a manager's consecutive funds turns out to be a product of timing. 

The conclusion? When two overlapping funds from the same firm have both outperformed, the shared outperformance is more likely explained by the overlap than by the firm.

One question for your next call with a sponsor is how many years did your last three funds run concurrently? Six years of overlap means three funds that lived through the same credit, the same entry prices, and the same exit window. Three track records, one set of conditions.

How Often it Repeats

The cleanest test of it comes from a study by Reiner Braun, Tim Jenkinson and Ingo Stoff. They assembled cash flow data on 13,523 individual buyout deals across 865 funds run by 269 managers, stripped the legal wrapper off the funds, and compared the underlying transactions directly.

Then they reran Kaplan and Schoar's original exercise. Sort every fund into performance thirds against its vintage. Ask what the manager's next fund does.

Kaplan and Schoar were working on an earlier and much smaller sample. 

Coming off a great fund, a manager is just as likely to deliver a terrible one next as they are to deliver another great one.

Working from a separate dataset of 1,924 fully liquidated funds, Korteweg and Sørensen had asked how much of a track record an investor would need before top-quartile past performance reliably indicated top-quartile expected returns going forward. With five observed funds, a top-quartile buyout manager has a 47% chance of genuinely belonging there. For venture, 37%. Push the model out to fifty observed funds, well beyond anything that exists anywhere, and it reaches 53% to 61%.

Of the 831 firms in that data, the median buyout firm managed exactly one fund. The 90th percentile managed four.

The evidence that would make a track record decisive would be larger than the evidence base any manager alive could hand you.

Where Persistence Survived

Persistence has been going away. Split the sample at 2000 and deal-to-deal persistence in the earlier period is strong and statistically significant. From 2001 onward it is neither. Manager fixed effects test whether a manager's identity explains anything at all about returns across their entire history. In the later period they lose significance completely. Braun, Jenkinson and Stoff trace this to competition. As the capital chasing deals rises, persistence falls. Buyout is an auction market now, and auctions do not reward relationships the way proprietary deal flow once did.

What survived the rise in competition sits at the bottom of the distribution. A manager coming off a bottom-quartile deal has a 31.1% chance of producing another bottom-quartile deal, against 24.0% for everyone else, and that holds no matter how competitive the market is. In today's high-competition conditions, a manager coming off a top-quartile deal moves from 24.4% to 28.2%, an effect the authors call weakly significant at best.

Diligence has kept most of its power as a tool for ruling managers out, and lost most of it as a tool for ruling them in. The worst managers keep raising funds anyway, which, as the authors note, raises real questions about the diligence being done on them. The hours belong on the managers you are trying to rule out.

The Three-Fund Objection

So what about the manager you already have? The one who has beaten their benchmark three funds running?

KKR made precisely that argument earlier this year. Announcing the close of North America Fund XIV, the firm pointed to the three predecessor funds behind it:

Together North America Fund XI, Americas Fund XII, and North America Fund XIII delivered a 23% gross IRR and a 2.1x gross multiple on invested capital through the end of 2025. Net of fees, that is 19% and 1.8x, which is a genuinely excellent record measured on the same basis as the dispersion data. The fund closed at approximately $23 billion.

The dispersion is real, and nothing here disputes it. The trouble is what three funds can carry. Three funds is three observations, and Korteweg and Sørensen's model puts five observations at 47% confidence for a buyout manager.

One manager has a genuine 8% annual edge. Another caught three good draws. They hand you the same document, and it can't tell you which is which.

In earlier issues we showed bigger funds cost you in returns and the small end is where the edge hides. Small funds also hold more signal about the manager, which is the one place a track record is worth reading at all.

Reiner Braun, Nils Dorau, Tim Jenkinson and Daniel Urban published a study in the Journal of Finance that measured performance differently. Instead of the rate at which capital compounds, it measures performance in net present value, the dollars a deal created.

The finding cuts against what we’ve said before. Relative returns still fall as funds get larger, but the dollars of value hold and even rise. Larger funds create more value while returning a lower rate on it, and this is the strongest case for scale anyone has published. It holds because, unlike the pooled figures the first issue set aside, it controls for who attracted the capital.

Scale creates dollars for the firm. The rate still degrades, and the rate is what the investor is buying.

Then there’s the number that matters.

Roughly 40% of the value these firms create traces to a single internal decision. Which named manager gets which deal and how much capital they are trusted with. The skilled managers are handed more and go on to better careers, so the firm is grading its own people continuously and accurately.

A track record reports the firm. It aggregates every deal by every manager into a single number with the firm's name on it. The individual who ran the three deals that made Fund III can be gone before Fund V raises, and the document will not say so.

The firm knows which of its people can do this. It has never been asked to print it.

Korteweg and Sørensen close their paper with a prescription. Past fund performance by itself is insufficient, so an investor needs information about a firm's internal organization and culture. Its compensation structure and incentives, its internal processes and deal sourcing, and the individual partners behind the deals in the record. 

One warning before diving into the questions below. At Build Wealth we run twelve funds and not one of them would clear the first two questions outlined below. Of course, neither would a first-time manager holding $300 million and a portfolio compounding beautifully with nothing exited yet. A short record is just not enough information. Treating that as disqualifying would repeat the error this series has spent three issues describing.

But these questions tell you how much weight the document you’ll be holding has. The last three questions actually disqualify. They test what a manager does, and they require a sponsor willing to answer. A refusal tells you something too.

1. How many funds are fully realized, and how many deals?

Count only what is fully exited. A remaining mark is a claim about the future, and a fund still holding is a fund still capable of disappointing. 

If a sponsor has three funds, one liquidated and two still open, that is just one observation. 

Then run the same count at the deal level. Exited deals accumulate faster than exited funds, and the deal count is often the more useful number, particularly for a young firm whose funds are all still open.

2. How many years did your funds run concurrently?

Vintage years and final closing dates are public for most institutional funds, so this is arithmetic rather than diligence. Funds that overlapped shared a credit environment, an entry-multiple regime, and an exit window, which means their returns correlate whether or not anyone running them is skilled.

3. Who ran these deals and are they still here?

Skill sits with individuals, and a track record aggregates individuals into a firm. Ask for deal-by-deal attribution: which partner sourced it, which sat on the board, which ran the exit. Then ask which of those people remain, and which will touch your capital.

A prior issue in this series asked what produced a fund's returns, growth or multiple expansion. This asks who produced them.

4. Show me the deals that lost money.

Poor performance is the more durable signal. A manager coming off a bottom-quartile deal repeats at 31.1% against 24.0% for everyone else, and that pattern holds regardless of how competitive the market gets. Losses carry more diagnostic weight than wins do.

So ask for the write-offs and the underperformers by name. Ask what the firm concluded from each and what changed afterward.

5. How does a partner earn more capital here?

Internal capital allocation accounts for roughly 40% of the value these firms create. That figure comes from the Journal of Finance study. Which partner gets which deal. Who is trusted with more. 

Who decides what a partner has to demonstrate to be handed a larger check, and how the firm handles a partner whose deals stop working?

This test isn’t perfect. A firm can document a rigorous process and execute it poorly. Nothing above produces the confidence a fifty-fund record would produce, because that confidence is unavailable to anyone at any price.

Weighting also depends on what the fund buys. Korteweg and Sørensen measure how informative performance is by asset class, and the spread is wide. Real estate performance is the most informative of anything they study, ahead of both buyout and venture. Buyout sits in the middle. Venture is the worst, carrying about a quarter as much information as real estate, which is why venture needs 25 or more observed funds before a firm's expected return can be estimated with any confidence.

Run all five questions correctly and you still cannot identify the winner. That changes how much you should put behind any one of them.

Concentration is a bet on identification. Write one large check into the manager you believe in and you have assumed you can tell which manager that is. Three funds’ worth of evidence does not support the assumption at the confidence a large position implies. A weak signal calls for holding more managers.

The diversification that matters here operates across your whole portfolio. Any single vehicle can be concentrated, including a one-asset deal, as long as the allocation around it is not.

It also calls for looking where the market isn't. If a long track record carries less information than investors believe, the premium on one runs too high and the discount on a newer manager runs too steep. Capital is flowing the other way right now, which is usually when it costs almost nothing just to look.

So spread the exposure. Disqualify aggressively on process. Then size what survives as though you might be wrong, because roughly half the time you will be.

Let the record tell you how much it knows, and underwrite the people for everything it cannot.

A track record can't tell you who to pick. Flip it and ask the opposite question, who to avoid, and it will answer cleanly every time. 

Underwrite the people for the pick… and everything else.

Did this land in your inbox via a friend or colleague? Subscribe to get your own weekly insight into private markets.

Already a reader? Help other investors open their eyes to the private markets and forward this to someone who would value the insight. Or hit reply to tell us what you think.

By continuing to receive the Wealth Stack newsletter, you acknowledge and accept our Privacy Policy, which includes the sharing of subscriber engagement data with select advertising partners.  

To unsubscribe, click here.

This is not financial advice. Illustrative output of a reasoned thought experiment. Not a backtest, guarantee, or prospectus. Actual results vary based on market conditions, fund selection, timing, fees, taxes, and factors not modeled. Private credit, CRE, and leveraged strategies involve significant risk including loss of principal. Consult a qualified financial advisor.