Factor Zoo
The "factor zoo" is a critical nickname for the explosion of investment factors that academics and practitioners have proposed over the years. Hundreds of factors now claim to predict returns, and the term captures a pointed concern: that many of them are not genuine return drivers but statistical illusions produced by searching the same data too many times.
The phrase reframes a celebrated research program as a warning. Factor investing rests on the idea that a handful of measurable characteristics explain returns, yet the sheer number of published factors raises the question of how many would survive honest, out-of-sample scrutiny. The factor zoo is shorthand for that crisis of credibility.
Definition
The factor zoo refers to the large and growing collection of factors reported in the academic and practitioner literature, many of which overlap, contradict, or fail to replicate. The term was popularized by John Cochrane, who described a "zoo of new factors" and called for a higher standard before any new factor is accepted. At its core, the critique is about reliability: a factor is only useful if its predictive power persists in data the researcher did not use to discover it.
Key Principle
The factor zoo is fundamentally a multiple-testing problem. If researchers test thousands of candidate factors against the same historical data, some will appear profitable by pure chance. Without correcting for the number of tests run, the literature accumulates false discoveries that look impressive in-sample but carry no real edge. The deeper issue is the gap between statistical significance and economic truth.
The Data-Mining Problem
The root cause is multiple testing: running many statistical tests inflates the chance that at least one looks significant by luck alone. A factor that clears a standard significance threshold may be one of dozens that were quietly tried and discarded, a practice known as p-hacking. Because failed tests rarely get written up, publication bias compounds the problem, filling journals with the lucky survivors while the failures vanish from the record.
The practical consequence is a literature that overstates how many real factors exist. Harvey, Liu, and Zhu argued that, given the hundreds of factors tested, the usual significance bar is far too low, and that a much stricter threshold is needed to separate genuine signals from noise. This connects the factor zoo directly to broader concerns about overfitting, where a model fits historical quirks rather than durable structure.
| Warning Sign | Why It Raises Concern |
|---|---|
| No economic rationale | A factor with no plausible risk or behavioral story is more likely a data artifact |
| Fails out of sample | Predictive power that disappears in new data suggests the original result was luck |
| Highly correlated with known factors | A "new" factor may simply repackage an existing one rather than add information |
| Marginal statistical significance | Barely clearing the usual bar is suspect once the number of tests is considered |
Known Limitations of the Critique
Limitations to Keep in Mind
- Some factors are real. The factor zoo critique does not mean all factors are spurious. A small set, including value, momentum, and profitability, has strong evidence across markets and time periods, and dismissing every factor would be an overcorrection.
- Stricter thresholds can hide real signals. Raising the significance bar to filter out false discoveries also risks discarding genuine but modest factors. The trade-off between false positives and false negatives has no clean solution.
- Correcting for testing is itself uncertain. Estimating how many factors were truly tested across the whole profession is hard, so the corrections proposed are approximations rather than exact fixes.
- Economic rationale can be retrofitted. A plausible-sounding story can be constructed after the fact for almost any pattern, so requiring an explanation is a useful but imperfect filter.
- Out-of-sample data is finite. Markets evolve, so even a factor that holds up in new data offers no certainty about the future. The critique improves discipline but does not deliver guarantees.
Academic Origin
The term gained prominence through John Cochrane's 2011 presidential address to the American Finance Association, where he described a "zoo of new factors" and questioned how many were independent. Campbell Harvey, Yan Liu, and Heqing Zhu followed with a systematic 2016 study cataloging hundreds of proposed factors and applying multiple-testing corrections, concluding that most published findings would not survive a proper statistical bar. Their work tied the factor zoo to the wider replication crisis affecting empirical research across many fields.
Further Reading
- Cochrane, J.H. (2011). "Presidential Address: Discount Rates." The Journal of Finance, 66(4), 1047–1108.
- Harvey, C.R., Liu, Y. and Zhu, H. (2016). "...and the Cross-Section of Expected Returns." The Review of Financial Studies, 29(1), 5–68.
- Hou, K., Xue, C. and Zhang, L. (2020). "Replicating Anomalies." The Review of Financial Studies, 33(5), 2019–2133.
Related Terms
Foxholm Financial is a fee-only registered investment adviser serving Georgia. We bring quantitative rigor to every client engagement. Explore our services or get in touch to discuss how we can help. To see how this kind of analysis informs real client work, explore a Strategic Portfolio Review.
Are you an institution or FinTech firm? Learn about our Quantitative Consulting Services.
Foxholm Financial trains the next generation of quantitative analysts. Students and early-career researchers can explore our quantitative investment fellowships.
This content is for educational and informational purposes only and does not constitute an offer to sell or a solicitation of an offer to buy any securities. Nothing herein constitutes investment advice or recommendations tailored to your individual situation. All investments involve risk, including the potential loss of principal. Past performance is no guarantee of future results. Information presented is believed to be factual and up-to-date, but Foxholm Financial does not guarantee its accuracy and it should not be regarded as a complete analysis of the subjects discussed. Before making investment decisions, consult with a qualified financial advisor who can evaluate your specific circumstances.