The carry factor is the tendency for higher-yielding assets to provide higher returns than lower-yielding assets. A simplified description of the carry trade is the return an investor receives (net of financing) if an asset’s price remains the same.
The classic application is in currencies—the currency carry trade—which calls specifically for going long currencies of countries with the highest interest rates and shorting those with the lowest. Currency carry has been both a well-known and a profitable strategy for several decades. However, the carry trade is a general phenomenon, having been profitable across asset classes.
Ralph Koijen, Tobias Moskowitz, Lasse Pedersen and Evert Vrugt, authors of the 2013 study “Carry,” found that a carry trade going long high-carry assets and short low-carry assets earns significant returns in various asset classes with an annualized Sharpe ratio, on average, of 0.7.
For the period beginning in 1983 and ending in 2012, the authors found the currency carry trade produced an annual return of 5.3% with a Sharpe ratio of 0.68. It was also highly persistent, with the 1-, 3-, 5-, 10- and 20-year odds of producing a negative return being 25%, 12%, 6%, 2% and 0%, respectively.
However, they also found that individual carry strategies have excess kurtosis (fat tails) and exhibit sizable declines for extended periods of time coinciding with bad economic states, such as during recessions and in liquidity crises. This provides support for the theory that the excess return of the carry trade is compensation for bearing the risk that assets will perform poorly in bad times.
The carry trade is also investable, as the markets in which the carry trade invests are among the most liquid in the world. Thus, implementation costs are low. And carry has a simple, intuitive rationale arising from the long-established concept that prices balance out the supply and demand for capital across markets. High interest rates can signal an excess demand for capital not met by local savings while low rates suggest an excess supply.
The Puzzle Of UIP
According to traditional economic theory, in what is known as uncovered interest parity (UIP), there should be an equality of expected returns on otherwise-comparable financial assets that are denominated in two different currencies.
Rate differentials would be offset by currency appreciation or depreciation such that investor returns would be the same across markets. There is an overwhelming amount of empirical evidence, however, contradicting UIP theory, resulting in the UIP puzzle.
The UIP anomaly may be due to the presence of nonprofit-seeking market participants, such as central banks and corporate hedgers (companies who must convert currencies in order to conduct business abroad), introducing inefficiencies to currency markets and interest rates. The carry strategy is not without risk, as there can be instances when capital flees to low-yielding “safe havens.” As mentioned earlier, this provides a simple risk-based explanation for the carry premium in which positive performance over the long term is compensation for potential losses in bad economic environments.
In other words, currencies that appreciate when the stock market falls might be a good investment because they provide valuable insurance against unfavorable fluctuations in equity markets. On the other hand, currencies that depreciate in times of poor stock market performance tend to further destabilize investors’ positions, and should therefore offer a premium for that risk. With these concepts in mind, we’ll review the literature on the currency carry trade.
Cash Flow’s Effect
Victoria Atanasov and Thomas Nitschka, authors of the 2015 study “Foreign Currency Returns and Systematic Risks,” found “a strong relation between currencies’ average returns and their sensitivities to cash-flow shocks in equity markets. High forward-discount currencies (currencies in which the futures trade at a large discount to the spot rate) react strongly to stock-market cash flows while low forward-discount currencies are much more resilient in this regard.”
They explain: “Basic finance theory suggests that high forward-discount currencies depreciate when the ‘home’ stock market receives bad cash-flow news that is associated with capital losses, whereas low forward-discount currencies appreciate under the same conditions. Thus, holding high forward-discount currencies is risky for a stockholder, while investing in low forward-discount currencies can provide him a hedge.”
The authors found their model “can explain between 81% and 87% in total variation in average returns on foreign-currency portfolios.” Atanasov and Nitschka concluded: “The free-lunch hypothesis on foreign-exchange markets is strongly rejected by the data. We argue that making money on currency investments is tightly linked to bad news about future dividend payments on stock markets: high forward-discount currencies load more on cash-flow risk than their low forward-discount counterparts.”
Martin Lettau, Matteo Maggiori and Michael Weber, authors of the 2014 study “Conditional Risk Premia in Currency Markets and Other Asset Classes,” also provide a risk-based explanation for the success of the carry trade. Their study covered the period January 1974 through March 2010 and more than 50 currencies.
The authors found that “while high yield currencies have higher betas (exposure to equity market risk) than lower yield currencies, the difference in betas is too small to account for the observed spread in currency returns.”
Debt And Currency Returns
Pasquale Della Corte, Steven Riddiough and Lucio Sarno contribute to the literature with the study “Currency Premia and Global Imbalances,” which appears in the August 2016 issue of The Review of Financial Studies.
Their paper offers investors empirical evidence that exposure to countries’ external imbalances (trade and capital accounts) is a key component to understanding currency risk premiums—countries run trade imbalances and financiers absorb the resultant currency risk. In other words, financiers are long the debtor country and short the creditor country.
The authors’ hypothesis was that currency excess returns are higher when the funding (investment) country is a net foreign creditor (debtor) and has a higher propensity to issue liabilities denominated in domestic (foreign) currency. The relationship between currency excess returns and net foreign assets captures the link between external imbalances and currency risk premiums.
Della Corte, Riddiough and Sarno also hypothesized that the currency denomination of external debt matters for currency risk premiums. Countries that cannot issue debt in their own currency are riskier. Thus, currency risk premiums are driven by the evolution and currency denomination of net foreign assets.
Specifically, in the presence of a financial disruption (i.e., risk-bearing capacity is low and global risk aversion is high), net-debtor countries experience a currency depreciation, unlike net-creditor countries. This risk generates currency risk premiums: Investors demand a risk premium for holding net-debtor countries’ currencies because these currencies perform poorly in bad times, which are times of large shocks to global risk aversion.
Their study covered a broad sample of 55 currencies and a subsample of 15 developed-market currencies for the period October 1983 through June 2014. Their global imbalance risk factor (IMB) is the return from a high-minus-low strategy that buys the currencies of debtor nations with mainly foreign-currency-denominated external liabilities (the riskiest currencies) and sells the currencies of creditor nations with mainly domestic-currency-denominated external liabilities (the safest currencies). Following is a summary of their findings:
- IMB explains a large fraction of the cross-sectional variation in currency excess returns, supporting a risk-based view of exchange rate determination based on macroeconomic fundamentals and, specifically, on net foreign asset positions.
- Investors demand a risk premium to hold the currency of net-debtor countries, especially if the debt is principally funded in foreign currency. The currencies of net-debtor countries with a relatively higher propensity to issue external liabilities in foreign currency have higher (risk-adjusted) returns than the currencies of net-creditor countries with higher propensity to issue liabilities in domestic currency.
- High-interest-rate currencies positively load on the global imbalance factor, thus delivering low returns in bad times during spikes in global risk aversion. The process of international financial adjustment requires their depreciation.
- Low-interest-rate currencies are negatively related to the global imbalance factor and thus provide a hedge by yielding positive returns in bad times.
- Net foreign asset positions contain information related but not identical to interest rate differentials in the cross-section of currencies. The main difference between sorting on interest rate differentials (a carry trade strategy) and sorting on global imbalances (a global imbalance strategy) is in the long portfolios of the two strategies. The riskiest countries in terms of net foreign asset positions are not necessarily the countries with the highest interest rates.
- A currency strategy that buys the extreme net-debtor countries with the highest propensity to issue external liabilities in foreign currency and sells the extreme creditor countries with the lowest propensity to issue liabilities in foreign currency (the global imbalance strategy) generates Sharpe ratios of 0.59 for a universe of major countries and 0.68 for the broader set of 55 countries. The excess return is greater than 5% per year.
- The global IMB risk factor has pricing power in the cross section of currency excess returns even when conditioning on the carry risk factor. The Sharpe ratio of the IMB factor also compares favorably to the Sharpe ratio of the carry trade.
- The global IMB risk factor has explanatory power for the cross sections of returns in other markets, including not just equities but bonds and commodities.
The authors concluded their results suggest “that returns to carry trades are compensation for time-varying fundamental risk, and thus carry traders can be viewed as taking on global imbalance risk.” They add: “Global imbalances are a key driver of currency risk premia: net debtor currencies are predicted to warrant an excess currency return in equilibrium and to depreciate at times when risk-bearing capacity falls.”
Thus, the IMB factor is related to, but different from, the currency carry trade. The bottom line is that “currency investors require a premium to hold the currency of debtor nations relative to creditor nations.” The authors’ findings are consistent with the financial theory that assets that perform poorly in bad times should have large premiums.
Currency And Credit Risk
Before closing, there’s one other study we need to cover that provides evidence investors should consider before adding a currency-carry trade allocation to their portfolio. Klaus Grobys and Jari-Pekka Heinonen, authors of the study “Is There a Credit Risk Anomaly in FX Markets?”, which was published in the August 2016 issue of Finance Research Letters, examined whether a link exists between sovereign credit ratings and currency returns. The availability of credit rating data dictated the sample period, which was the relatively short time frame from January 1998 through December 2010.
The authors divided a sample of 39 currencies into three portfolios by sorting on the previous month’s Oxford Economics sovereign credit rating. Portfolios were formed by going long the one-third of currencies with the lowest credit rating and going short the one-third of currencies with the highest credit rating. Following is a summary of their surprising findings:
- While premiums were found for the carry trade, volatility and momentum, there was a negative premium of 0.30% per month for the credit strategy. And importantly, the data was statistically significant at the 1% level.
- Average portfolio returns sorted by credit risk decrease linearly as they move from the low-credit-risk portfolio to the high-credit-risk portfolio. This suggests higher credit risk is associated with lower returns. In addition to negative returns, the long low-credit-quality and short high-credit-quality portfolio has a non-normal distribution. There is negative skewness (-0.5) and excess kurtosis (2.9), or a fat tail.
The authors concluded: “Even though risk-based asset pricing theory suggests that riskier assets should generate higher payoffs than less risky assets, our results suggest that currencies of countries with a high credit risk tend to generate lower returns than currencies of less risky countries.”
An important portfolio implication is that investors should account for credit risk when implementing a carry strategy. For example, investors can pursue carry only in the currencies of countries with high-quality sovereign debt, or avoid going long low-quality sovereign debt.
In conclusion, the academic evidence demonstrates there is strong empirical support for the existence of a link between exchange rate returns and macroeconomic fluctuations, providing a fundamental and theoretically motivated source of risk driving currency returns. Clearly, there are logical, risk-based explanations for the carry premium. These explanations also provide a logical resolution to the UIP puzzle.
Larry Swedroe is the director of research for The BAM Alliance, a community of more than 140 independent registered investment advisors throughout the country.