Quant boom leading to diminishing returns

For quant investors, fast money’s a hard addiction to break.

A mob of trend-followers is eroding the advantage that quantitative analysis usually provides to short-term equity trading strategies more quickly than normal, according to research from Bank of America Corp.

In particular, the Bank of America team lead by U.S. Head of Equity and Quantitative Strategy Savita Subramanian pointed to managed futures funds, which use faster-acting algorithms to spot trends in asset prices and volatility as trading signals. They’ve grown to represent about 10 per cent of the hedge fund universe with more than $250 billion assets under management, according to BarclayHedge.

At the heart of the issue is a phenomenon where a once-reliable strategy is destroyed when enough traders discover its potency. In quant speak, the alpha is arbitraged away. It “decays.”

Bank of America contends that this is occurring more frequently as new technology and data make get-rich-quick algorithms easier to build. The bank’s quant clients use three times as many pieces of predictive code, called factors, than they did 20 years ago.

However, with more math whizzes digging for more factors, the decaying of alpha means quant investors are bound to see quickly evaporating returns, they said.

“Good quantitative signals perform well in the shortterm, but the decay rate is extreme,” the researchers wrote in a note last week. “New alpha signals tend to be exploited and then quickly arbitraged away.”

Looking at managed futures funds, over the past 10 years their average yearly return was 3.1 per cent, according to a Credit Suisse Group AG Group basket that tracks more than 5,000 funds. The problem is over the preceding 10 years their average yearly return was more than double that. In 2016, managers tracked by Credit Suisse posted their worst year since 1995, falling 6.8 per cent.

Plenty of quants take issue with the bank’s line of thinking.

They note that the “managed futures” label doesn’t encapsulate all quant strategies, since not all quants trade in futures markets. What’s more, they say dataheavy funds use different time horizons, and not all quick-acting quants mine for the type of exotic data that Bank of America labels as quickly decaying.

Take Fort LP, a quantitative hedge fund that manages over $3 billion in Chevy Chase, Maryland, which since its founding in 1993 has used the same proprietary model for its managed futures funds, with little to no decay. From inception up to the financial crisis, the firm returned 16.3 per cent net of fees. Since then, it’s 13.1 per cent.

The decay some managed futures portfolios are experiencing likely has more to do with the trading environment than the crowded space, said Alan Marantz, a partner at Fort.

“For trend following to achieve high returns you need to have strong trends,” Marantz said by phone. “The actual trends that have been in the market over the last number of years haven’t been as strong as they’ve been in previous periods of times, like in bonds and currencies, for instance.”

Another issue comes in measuring the size of the space. Databases like BarclaysHedge show a growth in assets, compared to eVestment which shows a decline in assets under management from a peak in 2009.

Still, Subramanian and her team note that reliable quant strategies pay off over a longer time horizon through a style of investing that marries quantitative inputs and fundamental reasoning. And now, they say, the focus on short-term gains over long-term reliability is having an impact on the market.

“One of today’s greatest market inefficiencies may stem from the scarcity of capital devoted toward long-term, fundamental investing,” the researchers wrote. “Our analysis shows that fundamental signals significantly improve in efficacy over longer time horizons.”