Currency Risk Premiums Redux

Centre for Endowment Asset Management (CEAM)
Centre for Endowment Asset Management (CEAM)

Overview

This project investigates what really drives risk and return in currency markets, the researchers study a large currency cross-section using asset pricing methods that account for omitted-variable and measurement-error biases. 

Currency rates on large floor to ceiling screen.

Authors

Federico Nucera

Gabriele Zinna

Lucio Sarno

Cambridge Judge Business School, University of Cambridge

About the research

Investors demand compensation for holding risky currencies. Over the last two decades, many different “currency factors” have been proposed to explain these risk premia: dollar, carry, momentum, value, macro variables, volatility indices, news‑based measures of uncertainty, and more. This has created a small “factor zoo” in FX. One of the aims of this research project is to help tame this proliferation of FX risk factors by identifying which non‑tradable factors genuinely earn a statistically significant currency risk premium once omitted‑variable and measurement‑error biases are controlled for. 

The project asks 2 key questions: 

  1. How many fundamental risk factors actually price currency returns? 
  1. Which non‑tradable macro‑financial forces truly earn a currency risk premium? 

The project uses monthly data for 49 currencies against the US dollar from 1983–2017 and 46 currency portfolios built from nine well‑known strategies (carry, short‑ and long‑term momentum, value, global imbalances, term structure, output gap, etc.) and over 130 non‑tradable candidate factors, grouped into: Financial conditions (e.g. global FX volatility, liquidity, intermediary leverage), Macro variables (growth, inflation, unemployment, etc.) and Text‑based indices of economic and policy uncertainty. The researchers also combine 2 recent econometric advances: 

  1. Risk‑Premium PCA, which extracts latent factors that fit both time‑series and cross‑sectional returns, and 
  1. A three‑pass estimation procedure that corrects for omitted factors and measurement error, problems that bias traditional two‑pass Fama–MacBeth tests. 

Findings

The key findings of this research show that: 

  • the pricing kernel includes at least three latent factors that resemble (but are not identical to) a strong U.S. “dollar” factor and two weak high Sharpe ratio “carry” and “momentum” slope factors. Evidence for an additional “value” factor is weaker.  
  • Using this pricing kernel, they find that only a small fraction of the over 100 non-tradable candidate factors considered have a statistically significant risk premium, mostly relating to volatility, uncertainty, and liquidity conditions, rather than macro variables. 

The project helps to “tame the FX factor zoo”. It shows that a three‑factor structure – dollar, carry, momentum – is central to pricing currency portfolios, and that the key non‑tradable drivers are related to financial market volatility, uncertainty and liquidity, rather than traditional macro fundamentals. This provides a cleaner benchmark for both academic modelling and the design of currency investment strategies. 

Top