Downside Market Risk of Carry Trades

Abstract

Carry trades consistently generate high excess returns with high Sharpe ratios. I propose a new factor -the downside market factor -to explain the high currency returns. I show that carry trades crash systematically in the worst states of the world, when the stock market plunges or a disaster happens. High-interest currencies have significantly high downside market betas and significantly negative coskewness with the stock market, while low-interest currencies can serve as a hedging instrument against the downside market risk. The downside market factor can explain the returns to currency portfolios, sorted by the forward discount, better than other factors previously proposed in the literature. GMM estimates of the downside beta premium are similar in the currency and stock markets, statistically significant and close to its theoretical value. Therefore, the high returns to carry trades are a fair compensation for their high downside market risk. JEL classification: G12, G15, F31

The tail risk premia of the carry trades

“…They suggest that when the return to the equity market portfolio (CAPM) lies at the left of certain pre-defined quantiles, it might price the return to the carry trade. Especially Dobrynskaya (2014) and Atanasov and Nitschka (2014) focus on the subsample of observations conditioned by the negativity of the return to the equity portfolio while Lettau, Maggiori and Weber (2014) consider only the episodes when the market return is below one standard-deviation. Our work is actually closer to Lustig et al (2011), Menkhoff et al (2012), Rafferty (2011) and Burnside (2012) as we propose a factor mimicking the time varying tail risk in the currency market which we think is a good candidate to price the full distribution of the returns to the carry trade.…”

Yield curve risks in currency carry forwards

“…Besides the FM test, the GMM method has been alternatively used in that it is relatively simple and efficient method as to estimate linear asset pricing models and to make correct inference under weaker conditions. We jointly estimate betas and risk premiums as in Dobrynskaya (2014): where r jt is the excess return on currency pair j , f t a vector of factors, b j a vector of factor betas, γ a constant (pricing error), and λ a vector of factor risk premiums. The two moments in Equation (14) estimate the factor betas for each currency and the third moment estimates the risk premium.…”

“…A GMM estimator even with a small sample has a lower bias and higher efficiency than all the other estimators (see, Soto, 2009). On this account, numerous studies use the GMM method in examining currency risk proxies (e.g., Dobrynskaya, 2014; McCracken, Menkhoff, Sarno, Schmeling, & Schrimpf, 2012).…”

Asymmetric Volatility, Skewness, and Downside Risk in Different Asset Classes: Evidence from Futures Markets

“…I allow time‐varying betas of unsorted futures in an out‐of‐sample test as Dobrynskaya () finds variation in betas and downside betas. In the first step, Equations and are estimated in a five‐year rolling window, and in the second step, Equation is estimated with betas and downside betas in the preceding five years.…”

“…Menkhoff, Sarno, Schmaling and Schrimf () find that high returns to carry trades negatively relate to global currency volatility shocks and, therefore, come with a risk premium for their high currency volatility risk. Dobrynskaya () reports that carry trade returns are required for their high downside global stock market risk. As reported in Brunnermeier, Nagel and Pedersen (), Christiansen, Ranaldo and Söderlind (), and others, the popular funding currencies are the Japanese yen and Swiss franc and the investment currencies are the Australian dollar and New Zealand dollar.…”