Currency Factors
Last update: 01/11/2024.
Here, we compute currency risk factors and portfolios from the FX literature.
1 Factors
Factors are the monthly excess returns of currency portfolios. The ones we build here are:
- Carry and Dollar factors, as described by Lustig, Roussanov and Verdelhan (2011).
- Momentum, as described by Menkhoff, Sarno, Schmeling and Schrimpf (2012).
- Value, as described by Menkhoff, Sarno, Schmeling and Schrimpf (2017).
2 Methodology
- Carry factor: at the beginning of month \(t\), sort currencies into six portfolios based on their forward premium in \(t-1\), then compute the equally-weighted excess currency return of each potfolio in \(t\). The Carry factor will be the HML portfolio (Potfolio 6 - Portfolio 1). In the case of developed countries only, we sort currencies into five portfolios and the Carry factor is the Portfolio 5 - Portfolio 1.
- Dollar factor: the average of all forward premium sorted portfolios.
- Momentum: at the beginning of month \(t\), sort currencies into five portfolios based on their excess return in month \(t-1\), then compute the equally-weighted excess currency return of each potfolio in \(t\). The Momentum factor will be the HML portfolio (Potfolio 5 - Portfolio 1).
- Value: at the beginning of month \(t\), sort currencies into five portfolios based on the 5-year log-return of their real exchange rate in month \(t-1\), then compute the equally-weighted excess currency return of each potfolio in \(t\). The Value factor will be the HML portfolio (Potfolio 5 - Portfolio 1).
3 Data
Spot and 1-month currency forward rates come from Barclays Bank (BBI), WM Refinitiv (WMR) and Refinitiv, via Datastream. We set the order of preference as WMR > Refinitiv > BBI, and change the source of the data whenever it becomes available. The data is in FCU/USD.
We use data for 36 countries: Australia, Austria, Belgium, Canada, Hong Kong, Czech Republic, Denmark, euro area, Finland, France, Germany, Greece, Hungary, India, Indonesia, Ireland, Italy, Japan, Kuwait, Malaysia, Mexico, Netherlands, New Zealand, Norway, Philippines, Poland, Portugal, Saudi Arabia, Singapore, South Africa, South Korea, Spain, Sweden, Switzerland, Taiwan, Thailand, and the United Kingdom. And we also use a smaller dataset of only 15 developed countries: Australia, Belgium, Canada, Denmark, euro area, France, Germany, Italy, Japan, Netherlands, New Zealand, Norway, Sweden, Switzerland, and the United Kingdom.
For the 15 developed countries, we also use data quoted in FCU/GBP, which starts earlier. With that, we use triangular no-arbitrage relations to retrieve the spot and forward exchange rate and obtain quotes agains USD. This data is from 1978 to 1983, and from that year onwards, we use the standard data against USD. Therefore, this spot and 1-month futures dataset goes from 1978 to 2024.
Besides that, we also get data on real effective exchange rates from the IMF International Financial Statistics, which starts in 1979.
4 Results
Below, we show the cummulative log-return of each of these currency factors.
5 Data download
Data is available for download below: