Carry Trade:Easier Said Than Done

Although 2017 has been lauded as the year for the carry trade we find that the weak USD has been the main driver of the strong EM FX performance in H1. In fact, we find that the carry trade per se (defined by long high- vs. short low-carry FX) has barely been in positive territory this year. Also, high-yielders have often lagged across hard-currency credit markets.

    If USD weakness is mostly behind us (as the Fed seems determined to continue to normalize monetary policy) then this tail wind would wane in 2H17. In this case, rather than hinging on direction (vs. the USD) as in H1, performance would depend on portfolio selection. Under our benign backdrop of gradual tightening of liquidity conditions, carry baskets could then outperform.

    Carry trades have naturally been plagued by large drawdowns during crises. USD strength, US yields surges and commodities sell-offs have been the main causes of such large drawdowns post-GFC.

    Our analysis of the underlying changes in commodities and UST that caused these drawdowns suggest that - while these seem the most important risks to monitor in H2 - the magnitude of the shocks needed to wipe H1 returns is outside our range of our expectations.

    Optimized dynamic carry portfolios show BRL, TRY, RUB, ZAR, INR, and IDR (and only recently MXN) to be the most frequent currencies among the high-yielders and TWD, CHF, JPY, CZK, SGD, and SEK the most frequent funding pairs. Rebalancing can improve performance vs. fixed-composition baskets. Also, we find that a subset of 4 currencies per basket produces the best risk/reward and lower drawdowns.