Yale Case. By: Daniel Lazier, Jacob Thomas, and A. J. Wasserstein
This Yale case presents one of the most comprehensive empirical studies to date on how CEO demographics in ETA relate to financial outcomes. The authors analyze whether observable characteristics, such as education, professional background, partnership structure, gender, military experience, geography, and industry focus, are associated with differences in investment performance, measured primarily through IRR and MOIC.
Motivation & context
ETA investors make an initial investment decision before a company is acquired, meaning they must largely “bet on the jockey” rather than the business. As a result, demographic and résumé-level attributes often play an outsized role in capital allocation decisions. The authors seek to move beyond anecdote and intuition by testing common beliefs in the ETA ecosystem, such as the superiority of elite MBA programs, partnered searches, or finance backgrounds, using actual exit data.
Data & methodology
The study uses unmasked data from six serial ETA investors, covering 155 unique exited, traditionally funded SFs. This represents approximately 92% of all known traditional ETA exits, making the dataset unusually comprehensive for this asset class. Only exited investments are included; broken searches and still-operating companies are excluded, which inflates reported IRRs and MOICs relative to a full portfolio.
Demographic data were gathered from public sources (e.g., LinkedIn) and include MBA status and school, professional experience, years of work history, partnership status, gender, military background, geography, and industry. The authors conduct:
- Descriptive statistics and visual single-variable analyses (histograms and box plots)
- Single-variable regressions
- Multivariable linear regressions controlling for correlated traits
Throughout, the authors emphasize correlation rather than causation.
Single-variable findings
The descriptive analyses reveal several strong patterns:
- MBA education matters substantially. Search funds led by CEOs with at least one MBA dramatically outperform those without MBAs. Non-MBA searches show far more total losses and significantly lower median and mean IRRs. MBA-backed searches also exhibit less downside risk.
- MBA prestige matters less than expected. While all MBAs perform reasonably well, elite programs (Top 15, M7, Harvard/Stanford) do not consistently outperform other MBA programs. In fact, non-Harvard/Stanford MBAs often show higher median IRRs, and Kellogg stands out with particularly strong results.
- Partnerships outperform solo searches in descriptive statistics, with higher median IRRs and more representation in high-return bins, though this advantage comes at the cost of equity dilution for entrepreneurs.
- Professional background matters descriptively. CEOs with backgrounds in investment banking, PE, or consulting outperform other backgrounds in both IRR and MOIC.
- Military experience correlates with higher median and mean IRRs, though veterans show less extreme upside.
- Gender outcomes are broadly similar. Searches involving at least one woman perform comparably to all-male teams, though the sample size for female-led searches is small.
- Geography is highly consequential. ETA projects outside the U.S. and Canada significantly underperform, with lower medians, more negative skew, and weaker upside.
- Industry choice is less decisive. Software-focused acquisitions, despite their popularity, deliver IRRs similar to non-software businesses.
- Top-decile performance (outliers) is spread across many MBA programs and backgrounds, with no single demographic group dominating once adjusted for sample size.
These results suggest that demographics affect the distribution of outcomes more than they deterministically drive returns.
Multivariable regression results
When controlling for multiple characteristics simultaneously, many apparent advantages disappear. The multivariable model explains about 26% of IRR variation (adjusted R² ≈ 11%), indicating that most performance differences remain unexplained by observable traits.
Only two variables consistently retain statistical significance:
1/ Geography: SFs based in the U.S. and Canada outperform those elsewhere by approximately 36 percentage points of IRR, even after controlling for education, experience, and industry. This likely reflects structural advantages such as deeper capital markets, better financing terms, stronger ETA infrastructure, and more robust exit environments.
2/ MBA education: Teams with no MBA representation underperform MBA-trained teams by roughly 25 percentage points of IRR. Differences across specific MBA programs are not statistically significant once other factors are controlled for.
Other variables (partnership status, finance background, military experience, gender, years of experience, and industry) show weak or statistically insignificant effects in the multivariable model. Notably, the apparent advantage of partnerships largely disappears once education and geography are controlled for, suggesting that partnerships are correlated with (rather than causal of) better outcomes.
Interpretation & implications
The central takeaway is that there is no demographic silver bullet in ETA. While MBA education and North American geography are meaningful correlates of performance, most résumé-level traits explain only a small fraction of outcomes. Differences in performance appear to be driven more by unobservable factors, such as leadership ability, judgment, resilience, adaptability, and luck, than by demographics alone.
For investors, the results argue for widening the talent aperture, especially beyond elite MBA programs. For entrepreneurs, the findings suggest that lacking a prestigious background is not a fatal disadvantage, though pursuing ETA without any MBA is associated with significantly higher downside risk.
Limitations
The authors acknowledge several limitations, including survivor bias (only exited deals are included), omitted variables (e.g., company quality, timing, strategy), small sample sizes for certain subgroups, and reliance on data from a limited number of investors. As a result, the findings should be interpreted as directional rather than definitive.
Conclusion
In sum, this Yale Case provides strong evidence that MBA education and geography matter, but that most demographic characteristics do not reliably predict success in ETA once correlated factors are considered. The ETA ecosystem remains highly heterogeneous, and individual execution and judgment likely dominate demographics in determining long-term financial outcomes.
Read the full case in: https://som.yale.edu/sites/default/files/2026-02/A-Detailed-Analysis-of-ETA-CEO-Demographics-and-Financial-Outcomes.pdf
