What is the 'peso problem' in finance and how does it distort historical return estimates?
CFA Level III mentions the peso problem when discussing why historical returns can be misleading. I understand the basic idea — something bad almost happened but didn't — but I want a clearer explanation of how this creates estimation bias and what analysts should do about it.
The peso problem describes a situation where asset prices persistently reflect the possibility of a major negative event that ultimately does not occur during the sample period. The result: ex post (after-the-fact) returns look artificially high, and ex post risk looks artificially low.
The Mechanism:
Classic Example — Peso Devaluation:
From 1992 to 2001, the Argentine peso was pegged 1:1 to the US dollar. During this period, peso-denominated deposits offered higher interest rates than dollar deposits — the spread compensated investors for the risk of devaluation. For nine years, the peg held and investors earned consistently higher returns.
An analyst looking at 1992–2001 data would see:
- Stable exchange rate (no volatility from currency moves)
- Consistent interest rate premium over USD
- Apparently superior risk-adjusted returns
But in January 2002, the peg collapsed. The peso crashed from 1.0 to 3.8 per dollar within months, wiping out the entire accumulated premium many times over.
The opposite problem also exists. If a rare catastrophic event IS included in your sample (even just once), you may dramatically overstate its likelihood. Consider a stock that dropped 14% on a single day during the March 2020 panic. Based on that one-month sample, a 5% daily Value-at-Risk estimate would suggest a 13%+ loss once every 20 trading days — but the stock experienced no such loss in the subsequent 19 months.
Implications for CME Setting:
| Scenario | Ex Post Returns | Ex Post Risk | True Ex Ante Profile |
|---|---|---|---|
| Disaster priced but didn't happen | Overstated | Understated | Lower return, higher risk |
| Rare disaster did happen (in sample) | Understated | Overstated | Higher return, lower risk |
Practical Defenses:
- Adjust historical returns for the estimated probability and magnitude of priced-in events that didn't materialize
- Use scenario analysis that explicitly models tail events rather than relying solely on historical distributions
- Compare across markets — if a similar asset in a less risky environment earned significantly lower returns, the difference may represent an unrealized risk premium
- Be skeptical of asset classes with unusually smooth, high-return histories — they may be accumulating hidden tail risk
The peso problem is one of the most important conceptual challenges in CME setting because it can't be fixed simply by collecting more data — the bias is structural until the event occurs.
Practice peso problem scenarios in our CFA Level III question bank.
Master Level III with our CFA Course
107 lessons · 200+ hours· Expert instruction
Related Questions
What exactly is the Capital Market Expectations (CME) framework and why does it matter for asset allocation?
How do business cycle phases affect asset class return expectations?
Can someone explain the Grinold–Kroner model step by step with numbers?
How do you forecast fixed-income returns using the building-blocks approach?
PPP vs Interest Rate Parity for forecasting exchange rates — when do I use which?
Related Articles
Join the Discussion
Ask questions and get expert answers.