An earlier BIP Capital essay argued that the efficient frontier is only as honest as the risk numbers fed into it, and that private assets, appraised quarterly with each mark leaning on the one before, feed it the least honest numbers in the portfolio. That essay closed with a question rather than an answer. Once an allocation committee concedes that reported private-market risk is understated, the decision that matters is no longer how much to allocate to the asset class, but whom to trust to mark, monitor, and select within it.
That question has an operational answer, and the answer is more useful than the diagnosis that produced it. Pricing a private holding against economic reality rather than appraisal convention is not a reporting nicety. It is the same habit that separates managers worth backing from those who have merely been fortunate in how infrequently their books are repriced.
Why do private-market returns look less volatile than they are?
Reported private-market return series look smooth because the underlying assets are repriced with a lag, so each quarter carries forward much of the previous quarter’s return. The statistical signature of that lag is serial correlation, which can be removed. Desmoothing methods, including a first-order autoregressive filter and a lagged-beta adjustment that restores the public-market movement a stale appraisal absorbed late, recover a series closer to what an arm’s-length sale would have shown.
The mechanics rest on established finance literature rather than on recent invention. David Geltner’s 1991 work on appraisal smoothing in the Journal of Real Estate Finance and Economics provided the conceptual foundation, and Mila Getmansky, Andrew Lo, and Igor Makarov formalized the autoregressive model of serial correlation in illiquid-asset returns in the Journal of Financial Economics in December 2004. What has changed is not the method but the capacity to run it continuously rather than as a periodic academic exercise.
The output of that exercise is uncomfortable and useful in roughly equal measure. Applied across strategies, the adjustment lifts measured volatility unevenly. Mark Anson’s 2024 study in The Journal of Portfolio Management found that lagged-beta unsmoothing roughly doubled the reported volatility of small buyout funds, to 22 percent from 11 percent, and lifted early-stage venture capital to 87 percent from 29 percent, while private credit was virtually unaffected at 9 percent, compared with a reported 8 percent. Morningstar highlighted those findings in an August 2025 column on what it termed volatility laundering. Anson’s work also reported sharply lower Sharpe ratios after adjustment, with small buyout funds falling to 0.44 from 0.97.
Desmoothing carries its own limitations, and an allocation committee should hold them in view. The adjustment is model-dependent; its magnitude varies with the estimation window and the choice of public-market comparable, and it does not convert an illiquid asset into a liquid one. What it does is remove a known measurement artifact so that risk budgets are built on something closer to economic value. That is an improvement in inputs, not a claim to precision.
What does honest marking reveal that a fund-level NAV hides?
Volatility is only the first thing the exercise exposes. The deeper payoff lies at the level of the individual holding. A fund-level net asset value is an average that obscures its components, while a look-through to the underlying positions shows which marks are supported by recent transactions, which rest on models, and which have not moved in a year for reasons unrelated to value. An allocator who can see that composition is no longer relying on a manager’s self-report. That capability, not a quarterly statement, is what marking honestly means.
Public data does not permit a precise estimate of how much of the private-fund universe is marked to model rather than to transaction, and this article does not offer one. The concern is real enough that the SEC Division of Examinations named private-fund valuation practices and hard-to-value assets among its 2023 examination priorities, as set forth in its February 7, 2023, priorities list. That is a reason for allocators to ask the question at the position level rather than to assume an answer.
Why is smooth performance a risk signal rather than a stability signal?
The instrument that prices a position at entry also monitors it afterward, and it overturns a common intuition. A manager whose reported returns are smooth and highly serially correlated is often read as stable. Under honest marking, that smoothness is a caution rather than a comfort, because it indicates that reported values trail economic ones rather than that the underlying risk is low. Serial correlation belongs in the risk column.
A smooth line on a chart is not evidence of low risk. It is evidence of infrequent repricing, and those are very different things.
Interim appraisal marks are least reliable exactly when they matter most. During dislocations, carried-forward valuations diverge most from any price a buyer would pay, and secondary-market data quantifies the gap. Jefferies reported in its January 2024 Global Secondary Market Review that average LP portfolio pricing fell to 81 percent of net asset value in 2022 before recovering to 85 percent in 2023. In its January 2026 review, Jefferies reported that average pricing finished 2025 at 87 percent of NAV, with tail-end funds more than ten years old changing hands at 73 percent. Secondary pricing reflects liquidity preferences, buyer return targets, and valuation accuracy, so it is an indicator of the gap rather than a direct measure of it.
The corollary is that realized results, meaning capital distributed, are in our view a more reliable read on outcomes than any unrealized mark, because distributed cash cannot be revised later. This is a reasoned position rather than a settled empirical finding, and it matters more than usual in the current environment. Bain & Company’s Global Private Equity Report 2025 found that distributions as a share of net asset value fell to roughly 11 percent down from an average of 29 percent across 2014 to 2017. Bain’s Private Equity Midyear Report 2025 put DPI for 2018-vintage US and Western European funds near 0.6 times against a historical norm of roughly 0.8 times, using data as of the fourth quarter of 2024.
Honest marking also underwrites the governance model institutions are now adopting. As committees move toward judging each holding by its contribution to a total-portfolio risk and liquidity budget, that budget is only as accurate as the marks behind it. The CalPERS board voted on November 17, 2025, to adopt a Total Portfolio Approach that, effective July 1, 2026, replaces fixed asset-class targets with a 75/25 equity-bond reference portfolio and a 400-basis-point active-risk limit. A liquidity budget based on smoothed valuations understates how far a portfolio can move when it must raise cash.
How much does manager selection matter in private markets?
Selection is where outcomes are decided because dispersion between the best and worst private-market managers is wide and varies sharply by strategy. In public equities, the distance between the top and bottom of the manager distribution is comparatively narrow. In private markets, it is not. PitchBook’s Global Fund Performance Report, as of March 31, 2024, covering fund vintages from 2005 to 2019, shows a 38.5 percentage point gap in net IRR between top- and bottom-decile venture capital funds and a 13.3 percentage point gap in private debt, the narrowest spread of any strategy it tracks.
That structure appears durable rather than a feature of a single measurement period. Vanguard’s 2026 private-equity outlook finds that the dispersion of private-market excess returns remains significantly wider than among public-equity funds, a comparison Vanguard draws by using Direct Alpha for private funds across vintages from 1998 to 2024 against ten years of global active public-fund data as of December 31, 2024. Vanguard presents that finding directionally rather than as a single headline figure. When dispersion is this large, access to the asset class matters comparatively little, and selection matters a great deal.
Can a manager’s past returns be trusted to guide selection?
Past returns are a weaker selection guide than their prominence in diligence materials suggests. PitchBook’s persistence research, published in August 2023, found that persistence exists in private markets but that its practical value to allocators is limited, because a predecessor fund is, on average, only 3.5 years old when its successor is being raised, so its interim IRR has not yet settled into a final one. The persistence evidence, therefore, leans on the same interim marks that honest accounting distrusts.
The academic record is split by strategy, and that split matters for how a committee reads a track record. Steven Kaplan and Antoinette Schoar documented strong persistence across successive funds in the Journal of Finance in 2005. Robert Harris, Tim Jenkinson, Steven Kaplan, and Ruediger Stucke, working with information available on fundraising, found little or no persistence for buyout funds, both overall and after 2000, while venture capital persistence survived in weakened form, in NBER Working Paper 28109, issued in 2020. Readers should note that the Harris paper is a working paper rather than a peer-reviewed publication.
Scale compounds the difficulty. Florencio Lopez-de-Silanes, Ludovic Phalippou, and Oliver Gottschalg, writing in the Journal of Financial and Quantitative Analysis in 2015, found a median investment-level IRR of 36 percent in the lowest-scale decile, compared with 16 percent in the highest. A 2025 NBER working paper by Bhardwaj, Gupta, Howell, and Zimmerschied, number 33596, estimates causally that a 1 percent increase in fund size reduces net IRR by about 0.1 percentage point, largely because larger funds do larger deals that perform worse. The picture is not uniformly negative, however. Reiner Braun, Christoph Dorau, Tim Jenkinson, and Daniel Urban, writing in the Journal of Finance in 2026, find that while percentage returns compress with scale, total net present value persists and can increase, with roughly 40 percent attributable to internal capital allocation decisions. A large fund with a decorated record is neither the safe default it appears to be nor automatically compromised by its size.
Selection therefore cannot rest on reported numbers alone. It rests on what honest marking exposes: whether returns came from operational value or from leverage and multiple expansion; whether marks track transactions or models; and whether a manager maintains capacity discipline or gathers assets, because success rarely lowers fees. Those are questions about process, and they are answerable only by an allocator willing to apply the same look-through scrutiny to a prospective manager that it applies to its own book.
How should an institutional RIA evaluate a private-markets partner’s operating model?
BIP Capital built its operating system to enforce the discipline described above, which is why the firm describes itself as a private-market operating system rather than as a source of product. Two capabilities carry the most weight, and both are stated here as process descriptions rather than as claims about outcomes.
The first capability is on-demand look-through. BIP Capital’s reporting system is designed to show allocation partners the underlying positions inside a holding rather than a fund-level mark alone, including how each valuation was derived and how it would move against public-market comparables, with quarterly analysis intended to flag where reported and economic values are drifting apart. The aim is to give an allocator the same view of a position that the investment team holds internally, so that monitoring rests on evidence rather than on assertion.
The second capability is the underwriting behind the marks. The strategies delivered through the platform originate directly in the lower middle market rather than by buying syndicated exposure. They take first lien and senior-secured positions with enforceable covenants where the strategy calls for it, and they maintain capacity discipline, meaning the firm declines to raise or deploy beyond what the opportunity set supports, even when demand would allow more. Allocation partners reach the investment team that makes those decisions directly rather than through a ticket queue.
None of this is a promise about returns, and BIP Capital makes none here. Private-market investments involve substantial risk, including illiquidity, loss of principal, limited transparency, and the possibility that a valuation methodology, however rigorous, proves wrong. What is described above is how a book is marked, monitored, and underwritten. It is the operating evidence an allocation committee should require before trusting any private-market partner, including BIP Capital.
What should an allocation committee ask any private-markets partner?
An allocation committee need not take any of this on faith. It needs a short set of questions that distinguish partners who mark honestly from those who benefit from not doing so. Four are enough to begin.
A partner who answers these questions readily is applying the same discipline to its own book that it asks an allocator to apply to the portfolio. A partner who cannot or will not is offering a smooth line on a chart and hoping it will be mistaken for stability. That distinction is most of the decision.