Princeton’s declining returns: an equity hedge story
Princeton’s experience shows how hedges and changing exposures can reshape an alternative-heavy portfolio. MPI’s returns-based analysis highlights why asset owners need to look beyond allocation labels to total-portfolio behavior.
More than a year ago, the MPI Transparency Lab flagged a striking result: Princeton posted the lowest FY24 returns in our Ivy+ peer set – 3.9% vs. 8.3% Ivy average – and the return pattern was most consistent with very limited effective U.S. equity exposure. We inferred this from returns-based attribution using MPI Stylus Pro on Princeton’s published annual endowment returns.
Contribution by asset class to total return for each Ivy+ endowment and a 70/30 global benchmark, based on asset-class exposures estimated in MPI Stylus Pro from published annual endowment returns (FY24). Princeton has the lowest return in the group and is the only endowment with no U.S. equity contribution in the attribution.
When we then reviewed Princeton’s own disclosures, we found language suggesting that PRINCO had built a meaningful equity-market hedge over roughly 2019–2022. In the FY2022 Report of the Treasurer, PRINCO described a “significant equity market hedge” that it had “established and amplified” as a defensive response to high U.S. equity valuations and elevated effective equity beta. The hedge was implemented primarily via equity index shorts (including NASDAQ futures) and later restructured to include put options, in part to reduce “melt-up” liquidity risk.
Roll forward one year and the same structural issue—near-zero effective exposure to the U.S. equity rally—helps explain why Princeton finished last among Ivy+ peers in FY25. This shift also shows up in longer-horizon optics: after ranking near the top of the peer group on rolling 10-year performance for much of FY15–FY22 (occasionally second only to MIT), Princeton’s rolling 10-year return fell to below the Ivy average by FY25.
Download: Princeton MPI360 report (via the MPI Transparency Lab).
The issue has become consequential enough that Princeton has publicly reset expectations. In a February 2, 2026 letter, President Christopher Eisgruber noted that PRINCO had lowered its long-term expected endowment return to 8.0%, down from 10.2% only three years earlier, citing more limited opportunities in illiquid assets and lower expected future returns. That public explanation focused largely on private-market opportunity sets; it did not address the role of the equity hedge or the portfolio’s effective exposure to the U.S. equity rally.
MPI’s powerful factor discovery engine “Factor Search” deciphers some of the most sophisticated strategies.
Why asset owners need holistic top-down analysis
The attribution results above are derived from MPI-estimated asset-class exposures using annual endowment returns, rather than disclosed policy allocations. That distinction matters. PRINCO’s own rationale for the equity hedge referred to a portfolio with a “bias toward equities or equity-like assets,” noting that 94% of the portfolio was allocated to such investments. But a holistic returns-based analysis tells a more nuanced story: Princeton appears to have had little or no effective U.S. equity exposure even before FY2019.
The chart below compares Princeton’s estimated exposure dynamics with the average of the eight Ivy institutions (including Princeton). While the Ivy average shows a meaningful U.S. equity loading, Princeton’s public-equity exposure appears to have come primarily through emerging-market equities. The chart also shows a small U.S. equity exposure that disappears around FY2019, consistent with PRINCO’s public disclosures about establishing and amplifying an equity hedge.
This is precisely why institutional investors need total-portfolio, returns-based analysis—not just sleeve-level reporting. In alternative-heavy portfolios, exposures can net out in unexpected ways. Hedge funds, private equity, public equity, and overlays may either offset one another, as appears to have happened at Princeton, or unintentionally reinforce the same risk. Private equity beta, leverage, and valuation lags also change over time, so the economic exposure of a 40% private-equity allocation is not static. In Princeton’s case, the combined effect appears to have been a portfolio that missed much of the U.S. equity rally while still carrying substantial illiquid, equity-like risk.
Private equity beta
In FY21, Princeton reported a 98.7% return for its private-equity sleeve, helping drive a record 46.7% total endowment return. The private-equity allocation rose to 41.9%, well above its 30% target. By FY25, the reported allocation remained roughly unchanged at about 41%.
But reported allocation is not the same as economic exposure. MPI’s returns-based analysis of PRINCO’s annual returns suggests that Princeton’s effective PE/VC exposure was above 60% — higher than its reported allocation and the highest among the Ivy institutions, second only to MIT in MPI Transparency Lab’s Ivy+ cohort.
Assess a fund’s potential exposure to complex hypothetical market shocks or historical regimes in MPI Stylus Pro.
Since the private-equity exposures below are measured against a generic diversified index — Cambridge Associates U.S. PE/VC indices in our study — this could mean that Princeton’s actual allocation carried higher risk/beta than the broad index of 800+ funds.
To justify the equity hedge PRINCO was rightfully counting private equity as a part of the 94% equity exposure. Private equity is indeed equity and in many cases is a highly levered equity – both directly, as with many buyout strategies and indirectly – through venture capital optionality. However, this leverage/beta is difficult to see in smoothed (often with mark-ups and mark-downs) and lagged appraisals of private companies resulting in both autocorrelated returns and idiosyncratic noise. Once such noise is filtered out, an interesting pattern emerges.
In the chart below we analyze Morningstar/Pitchbook U.S. PE Index vs. S&P500 Index in MPI Stylus Pro to determine the time-varying beta of the former. The index provides a reasonable proxy for the $14.7B PRINCO’s PE allocation consisting of hundreds of individual LP interests. The analysis shows PE market beta hovering around 1.0 and reaching its peak of ~1.4 around FY21 to fall to ~0.2 in just two fiscal years. Such record low beta for an extended period of time is the product of depressed valuations, low deal volume, rising interest rates, valuation lags/staleness, among others.
With little or no exposure to U.S. equities due to the hedge and historically record low PE beta – the dominant equity exposure after the hedge – the PRINCO portfolio was left with almost no effective equity exposure and trailed all their Ivy peers.
Saves time and money with MPI’s comprehensive software platform that provides both historical and forward-looking analysis capabilities
Had PRINCO been monitoring total effective equity exposure in real time, the changing beta of private equity might have led to a different hedge calibration — including reducing the hedge, removing it, replacing lost equity exposure through more liquid instruments such as futures and/or evaluating potential equity exposure cancellation due to the “Independent Return” sleeve, Princeton’s nomenclature for hedge fund-like investments.
That is why private-market risk has to be monitored at the total-portfolio level — not solely through bottom-up appraisals of hundreds of private companies.
The lesson: allocations are not exposures
Princeton’s experience illustrates why total-portfolio return behavior can reveal what position-level information and reported asset-class allocations may miss. Total return takes everything into account: public and private assets, hedge-fund exposures, overlays, valuation lags, leverage, and the way all of those elements net against one another.
The lesson is not that hedging was wrong, or that private equity was the problem. The lesson is that positions, allocations, and sleeves must ultimately be evaluated through the behavior of the total portfolio. For large asset owners, especially those with substantial private and hedge-fund allocations, the relevant question is not simply what do we own? It is how is the portfolio actually behaving — after all exposures and overlays net out?
In Princeton’s case, MPI’s returns-based analysis suggests that the answer changed materially over time. That change helps explain how an endowment described by PRINCO as 94% allocated to equity or equity-like assets could still miss much of the U.S. equity rally.
Technology Solutions
Trusted by
