Is the Gargantuan Currency Market Getting Overcrowded?

|
 |  Includes: FXE, UDN, UUP
by: Christopher Holt

The daily turnover in the world’s biggest market, currencies, hit $4 trillion last year, according to the latest triennial survey from the Bank for International Settlements (“BIS”), published in December. Who would ever think to ask if traders could get claustrophobic in such a huge market?

Yet that is exactly what currency manager Momtchil Pojarliev and NYU Professor Richard M. Levich ask in their latest of several seminal research articles on currency funds. The question of capacity is germane to one of BIS’s primary objectives: Promoting international financial stability, given how regulators see crowded trades as a “systemic risk” factor (along with bankers’ bonuses, housing bubbles, over the counter derivatives, etc.).

[Click all to enlarge]

Although there are 170 currencies in the world – and thousands of possible currency pairs – the two authors don’t investigate position overlap between the 107 managers in their study. Instead, they looked for three common “style factors” – carry, trend following and value – which are, in turn, inferred from managers’ weekly return patterns. “Crowdedness” in a particular strategy was inferred from more managers being positively (rather than negatively) correlated with that strategy. A surfeit of negative over positive correlations indicated that the cohort was, in aggregate, taking a contrarian approach to one of the three factors.

The researchers inferred that none of the three style factors explained more than one third of the return patterns (after netting off the contrarians) -- and even then crowdedness only touched such levels very briefly. The popularity of the carry trade return profile peaked at 32% in April 2008, but was much lower for most of the period. The inferred prevalence of trend following moved in the opposite direction, starting at 35% in 2005 and steadily declining towards 0 by May 2008, before briefly turning negative (indicating return patterns consistent with counter-trend, or mean reversion, positioning).

The value trade varied the most as a presumed strategy: Staying around 10% for most of the time and “peaking” at -25 percent in 2010. This means managers were believed to be taking the opposite side of the trade based on their returns.

So the inhabitants of the “DB Select” platform used for the study were a pretty diverse bunch, displaying both positive and negative exposure to the style factors, as well as varying this over time. The authors pointed out that systematic managers with a fixed and stated strategy, such as trend following, lack the discretion to switch into a carry trade or value trade, which limits scope for crowding – and conversely think that discretionary managers, who can dynamically vary style exposures, could increase crowding risks in the future.

Yet the evidence found so far for an overcrowding of styles is weak – and the study didn’t even try to gauge position level crowding. Managers could, of course, use very different positions to arrive at the same “style factor” exposures – owning Australian dollars versus Japanese yen, going long Brazilian reals and short Swiss francs or buying Turkish lira against euros, all presently carry trades.

A closer look at the BIS survey suggests it may be worth considering crowding in a broader sense: Are investors starting to dominate currency markets, which were once overwhelmed by non-profit-making participants? The 2010 survey does show turnover from “Other Financial Institutions” (i.e. insurance companies, mutual funds, pension funds, hedge funds, central banks) surpassed trading between dealers for the first time ever, as shown on the pie chart below.

Origin of Daily Global Currency Turnover

Moreover, the percentage of daily global turnover accounted for by this category of traders has been growing steadily over the past five BIS surveys, more than doubling since 1998, as the bar chart below shows.

Percentage of Global Daily Turnover from “Other Financial Institutions”

What we do not know is what proportion of these traders are hedgers or liquidity seekers, and how many are profit seekers. However, if currency trading is becoming a more popular investment strategy, it remains a profitable and uncorrelated one, according to an earlier 2010 paper from Pojarliev and Levich. In particular, the two authors highlighted how alpha-generating currency managers have provided excellent diversification benefits for equity investors. For now, it seems there is plenty of room for currency managers to pursue a range of strategies, many of which act as powerful diversifiers.