(Bloomberg) -- The dollar is staging a comeback. Pundits are out in full force, suggesting this might finally be the end of its recent weakness and the start of a strengthening trend they have been predicting for more than a year. They may be right, but in my career spanning more than three decades in finance, I’ve learned that nobody really knows what moves currencies.
The Federal Reserve began increasing interest rates in earnest early in 2017, while the European Central Bank and Bank of Japan kept their accommodative pedals firmly to the metal. But instead of appreciating, as the experts predicted, the dollar plunged almost 20 percent against the euro and 10 percent against the yen in the ensuing 12 months. Little wonder most currency prognosticators were left scratching their heads.
My first real job in markets in the mid-1980s was as a proprietary currency trader for a New York money center bank, as they were then called. I lasted about three years before I concluded I couldn’t make heads or tails of currencies. A decade and Ph.D. later, I still didn’t know what drove currencies. By then I was working in an investment bank, and one of my tasks was to assess the accuracy of forecasts. It turned out that the bank’s economists and strategists weren’t bad at predicting things like growth, inflation, policy rates, bond yields and stock market moves. But they were pretty hopeless when it came to currencies. I was relieved I wasn’t alone.
In my academic studies and practical experience, I was never convinced by textbook models of exchange-rate determination, in particular interest-rate differentials. It always seemed peculiar to believe that differences in the cost of borrowing or lending in two currencies would — alone — determine the rate of exchange. After all, the demand for currency reflects other capital and current-account transactions. Perhaps in the early 1960s, when interest differential models were the workhorse of exchange-rate theory and capital markets were relatively closed, they were the straw that stirred the currency drink. But the opening of the capital account and the rapid increase of world trade in the ensuing half century created enormous other sources of cross-border money demand. Equity flows, direct investment, valuations and investor sentiment, among others, became increasingly important determinants of currency moves.
For example, by early 2017 the dollar was becoming progressively overvalued. According to the Organization for Economic Cooperation and Development, the real, trade-weighted value of the dollar surged 20 percent from 2011 to the end of 2016. That trend was unlikely to persist, irrespective of interest differentials, particularly as it also became clear by early 2017 that growth was accelerating in Europe, Japan and various emerging markets.
By this time last year global investors had also already piled into U.S. equities, a love affair they began around 2010. These investors were therefore sated on U.S. stocks just as other markets began to outperform, most notably global emerging markets and Japan. Unsurprisingly, equity portfolio flows shifted from the U.S. to other markets last year, offsetting interest differentials as drivers of currency values.
Trade and trade deficits also matter. The U.S. current-account deficit has begun to widen again, reaching $471 billion in 2017, or 2.4 percent of gross domestic product. Japan’s external surplus last year was 3.9 percent of GDP and the euro zone’s was 3.4 percent of GDP. Those imbalances create an excess supply of dollars relative to yen or euros.
In addition, the Trump administration’s willingness to thrice announce tariffs this year may have created the impression among investors that the U.S. government favors a weak dollar as a means to boost competitiveness and create jobs.
So what conclusions are we supposed to draw? Whether or not the dollar’s recent bounce becomes a trend is anyone’s guess. True, that is unhelpful. But the reality is that exchange rates are intrinsically more difficult to forecast than most other economic or market variables because they are driven by a myriad of factors that can’t be simplified into a simple rule of thumb.
In strategies we run, we have recently removed some hedges against dollar weakness and even taken some small positions to take advantage of dollar strength. But I suspect that currency will never be a major source of active risk in our portfolios. Most of the time, after all, it is the stuff of heads and tails.
©2018 Bloomberg L.P.