This study investigates an efficient parametric portfolio policy model to improve the return distribution of the well-known currency carry trade investment strategy. This carry trade strategy invests into high-yielding currencies that are subsequently funded by low-yielding currencies. Following this investment procedure has led to significantly excess returns for the investors, at least over the past four decades. However, these returns were subject to a high crash risk, which hit its peak during the US subprime crisis in 2008/2009 with portfolio losses of up to one third of the investment value. The constructed model overcomes these bad portfolio properties through computing the optimal carry trade portfolio weight for any monthly revolving investment period. This is done by modeling the optimal weight as a function of the carry trade’s risk characteristics. Especially, when using global FX option-implied variance risk, as well as global consumer price inflation and commodity prices as background risk factors, the model delivers extremely-efficient out-of-sample results with annualized mean returns of up to 8.4% over an eight-year period, accompanied with a low standard deviation, positively skewed returns and leading to Sharpe ratios around unity, including transaction costs. These promising statistics are largely maintained when allowing for higher leveraged portfolios.
Keywords: portfolio policy, carry trade, implied variance, currency portfolio, crash risk
Continue reading here.