The Value Premium in G10 Currencies
Buy "cheap" currencies (below PPP fair value), sell "expensive" ones, expecting reversion.
A documented currency factor that, like the broader value factor, delivered weak-to-negative returns through much of the post-2008 period.
- Why it fails
- Cheap currencies can stay cheap for years; the post-2008 stretch was an extended drawdown for value across asset classes, and the currency version shared in it.
- When / how it stopped
- Currency value was established as a factor (Asness, Moskowitz & Pedersen, 2013; Menkhoff, Sarno, Schmeling & Schrimpf, 2017), but it — and the broader value premium — delivered weak-to-negative returns through much of the period after 2008.
The currency value premium applies a classic factor idea to FX: estimate a currency’s fair value — usually via purchasing-power parity — then buy the currencies trading cheap relative to that anchor and sell the expensive ones, expecting reversion over time.
It has real academic standing. Asness, Moskowitz & Pedersen (2013), in “Value and Momentum Everywhere”, showed that value (and momentum) earn premia consistently across markets and asset classes, currencies included, and that the two factors are negatively correlated — a genuinely useful combination. Menkhoff, Sarno, Schmeling & Schrimpf (2017) studied “Currency Value” in its own right within the Review of Financial Studies.
The reason it sits in the graveyard is not that the factor was never there — it is regime fragility. Value as a style went through an extended, well-documented drawdown across much of the period after 2008, and the currency expression of value was not spared. The core weakness of any value approach is that cheap can stay cheap: a currency can trade below its PPP-implied fair value for years, propped up by rate differentials, policy, or flows, without reverting on any tradable horizon.
So the gap between the long-run academic premium and the lived post-2008 experience is wide. The factor is documented; the recent payoff is not. The reader can decide whether a premium that disappears for a decade-plus is one worth waiting on.
Sources
- Asness, Moskowitz & Pedersen (2013), "Value and Momentum Everywhere", Journal of Finance
- Menkhoff, Sarno, Schmeling & Schrimpf (2017), "Currency Value", Review of Financial Studies
Frequently asked
Does the currency value premium still work in 2026?
It has been an unreliable performer. Currency value — buying currencies cheap relative to fair value (often PPP) and selling expensive ones — is a documented factor; Asness, Moskowitz & Pedersen (2013) showed value works across markets including currencies, and Menkhoff, Sarno, Schmeling & Schrimpf (2017) studied "currency value" directly. But the broader value factor delivered weak-to-negative returns through much of the post-2008 period, and the currency version was not insulated from that environment. The long-run academic premium and the recent realized experience diverge sharply.
Why did buying cheap currencies stop paying?
Value strategies rely on reversion toward fair value, and that reversion can be slow or absent for extended stretches. The post-2008 era — shaped by unconventional monetary policy and persistent divergences — was an extended drawdown for value as a style, documented broadly across asset classes. A currency can trade below PPP for years without correcting, and during such regimes a value tilt simply does not get paid.
Not investment advice — your mileage may vary, but the burden of proof is on the person claiming an edge. This entry describes general research and published evidence (or its absence), not a recommendation. See the full disclaimer.