To paraphrase the father of macroeconomics, John Maynard Keynes: When the market environment changes, my portfolio currency allocation changes … what do you do?
In recent years, the U.S. dollar experienced one of its longest and largest rallies since floating exchange rates began in the early 1970s. Given the uncertainty in the outlook for U.S. interest rates and the path of the U.S. economy after the presidential election, the outlook for the U.S. dollar is far less clear than in the prior three years. In this environment, how should pension fund chief investment officers think about the issue of managing foreign currency exposure?
Historically, there exist two schools of thought about how to deal with currency risk in international portfolios. The first is to do nothing and leave all currency exposures unhedged. The second is to hedge it away and eliminate currency risk from the portfolio entirely. These two competing approaches are founded on several core beliefs that have questionable foundations.
School of thought No. 1: Do nothing
This approach is based upon two beliefs: i) Currencies revert to the mean over time and hence, the impact of short- to medium-term currency movements can be ignored; ii) If you like the return potential of a foreign asset, you must also like the currency, as the drivers of asset appreciation will also lead to currency appreciation. If these two assumptions are true, currency exposure can be ignored and never hedged.
i) It is dangerous to assume that currency markets mean-revert as exchange rates in both real and nominal terms can move by very large amounts over multiyear time frames and have a significant impact on portfolio risk and return. The reason this myth is commonly believed is that currency markets can exhibit cyclical behavior. However, these cyclical movements revolve around the long-term fair value of a currency, which does change materially over time such that currencies rarely return to the same value in either real or nominal terms.
For example, since the U.S. came off the gold standard in 1971 and the U.S. dollar became a freely floating currency, the dollar has fallen in nominal terms vs. the Japanese yen — from 357 yen per U.S. dollar to 76 yen at its 2012 low, a fall of more than 78%. Intervening periods have also experienced large moves.
Indeed, the Japanese yen is by no means an isolated example, and large currency moves can have a significant impact on the U.S. dollar value of international exposure. In fact, when applied to the currencies in general as held by the U.S. pension fund industry, the most recent bout of U.S. dollar strength that ended in mid-2015 is estimated to have cost the U.S. pension fund industry $1 trillion in lost value by not being hedged. Unmanaged currency exposure can have a large impact on your portfolio in either direction.
ii) The correlations between asset markets and their underlying currencies are highly variable depending upon the prevailing economic and financial conditions. Therefore, it is a myth to claim that if you buy a foreign asset you should always have a positive view on the underlying currency. In a vivid recent example, the Japanese Nikkei 225 index rose 83.7% between June 30, 2012, and Dec. 31, 2013, in local currency terms but only rose 37.1% in U.S. dollar terms given the collapse of the Japanese yen over the same period. A currency hedge would have precisely bridged that gap.
It is clear from historical evidence that these two beliefs in the “Do Nothing” school are myths, and therefore it can be extremely dangerous for a portfolio to leave currency exposure unmanaged at all times.
School of thought No. 2: Hedge all currency risk back to the base currency
This approach also is based upon two assumptions: i) Currency exposure adds risk to a portfolio with no expected return and therefore is uncompensated risk. As a result, it should be completely hedged away; ii) It is impossible to forecast currency markets and so they should not be invested in with the objective of generating an excess return.
i) Currency markets are mostly driven by different factors than equity and fixed-income markets, and as a consequence, currency returns have a very low and sometimes a negative correlation to these asset markets. As such, retaining some currency exposure in a portfolio can improve its risk characteristics due to the diversifying nature of the exposure. In the example below, hedging the full currency exposure is riskier than hedging only 80%.
Furthermore, from a practical point of view, a fully hedged foreign currency position can lead to large losses in periods of U.S. dollar depreciation, which can interfere with the asset allocation of the portfolio as asset holdings will need to be liquidated to fund the realized currency hedge losses.
ii) Many claim that currency markets are impossible to forecast because they are the most efficient markets in the world based on their deep liquidity, transparency and low transaction costs. This is a myth because while they are transactionally efficient (low cost to execute) they are not efficient in the sense that no edge can be gained by investors analyzing the key drivers of exchange rate changes. This lack of efficiency stems from the fact that not all currency markets participants are seeking to make a profit explicitly from their participation in the currency market. Central banks, for example, have monetary policy objectives, and corporate treasurers are focused on their currency payables and receivables.
This inefficiency is backed up by empirical evidence that professional investors in currency markets have been able to extract positive excess returns over time. A recent CFA Institute paper co-authored by New York University professor Richard M. Levich, explained that “empirical evidence suggests that some managers have been able to deliver statistically and economically significant alpha.”
For both of these reasons, it is not advisable or optimal to maintain a fully hedged static currency position in an international portfolio.
A more pragmatic approach
If neither school of thought can be relied upon to inform the best way to think about managing currency exposure, and there are significant problems with either static approach, what then is the preferable framework?
A more pragmatic and effective approach is to recognize that an active approach to managing currency exposure within an international portfolio is advisable in exactly the same way that an active allocation approach to the equity, fixed-income and alternative asset allocation is commonplace. The specific approach depends on the investment objective.
Should the primary investment objective be to reduce the risk inherent from having currency exposure and mitigating potential drawdowns, a dynamic hedging approach is an attractive option that seeks to adjust the hedge ratio on foreign currency exposures vs. the base currency to anticipate periods of base currency strength and/or weakness.
By contrast, should the investment objective be to add incremental and uncorrelated returns from changing the currency mix vs. the underlying asset mix, then an active currency overlay approach should be considered as it turns the problem of managing currency exposure into a virtue by improving the return per unit of risk of the entire portfolio.
Either way, the overarching conclusion is that a flexible approach is required while a straitjacketed, static approach is to be avoided.
Mark Astley is CEO of London-based Millennium Global Investments.