Swedroe: Beware Lure Of Market Timing

Lowering volatility with market-timing strategies is no free lunch, Swedroe cautions.

Market-timing strategies attempt to outperform a buy-and-hold strategy by anticipating the future direction of a market. They can work, but mostly they don’t.

First, such strategies are based on the belief that future security prices are predictable, typically through the use of technical indicators, such as trend following or momentum, that are computed from the past prices.

There are numerous papers that confirm the superior performance of momentum strategies across markets and around the globe. Studies have also found that trend-following strategies have enhanced risk-adjusted returns, producing higher Sharpe ratios.

Also, since some trend-following trading rules would have helped investors avoid some of the massive losses experienced during the two severe bear markets that occurred during the first decade of 2000s, the popularity of such strategies has increased.

Valeriy Zakamulin, author of the November 2013 study “The Real-Life Performance of Market Timing with Moving Average and Time-Series Momentum Rules,” examined the performance of these strategies, with particular concern for the potential for data mining.

To address the issue of “data snooping,” he used a bootstrapping method to examine out-of-sample data. A bootstrap is a computer-intensive method of estimation of the sampling distribution of a test statistics by resampling—that is, randomizing—the historical data. His study also accounted for trading costs, though he didn’t examine taxes.

Using the S&P 500 Composite Stock Price Index, the Dow Jones industrial average, and both a long-term and intermediate-term U.S government bond index, Zakamulin examined two technical trading rules: the simple moving average rule; and the momentum rule. Every technical trading rule invests in stocks/bonds when a “buy” signal is generated, and then moves to the risk-free asset when a “sell” signal is generated.

Perhaps the most interesting of Zakamulin’s findings was that these strategies failed most of the time. He defined failure as a period in which the riskless investment, Treasury bills, failed to deliver a higher return than did stocks during a “sell” period. He found that the average failure rate was about 80 percent.

This result demonstrates that, when the passive benchmark is a stock price index, the superior performance of the market-timing strategy is confined to some relatively short particular episodes. Thus, compared to a buy-and-hold strategy, market-timing investors frequently have to wait a very long time and experience painful emotions—referred to as tracking-error regret—because their active portfolios consistently lag the benchmark.

In fact, over the total sample from 1926 to 2012, there were only four relatively short historical periods that contributed most to the superior performance of the market-timing strategies—the severe bear markets of 1930-31, 1973-74, 2001-02 and 2007-08. Of course, that doesn’t detract from the veracity of those data.

However, Zakamulin also highlighted the fact that the performance of market timing was inferior during a 25-year period from 1975 to 1999 regardless of the choice of passive stock-market index. How many investors would be willing to wait 25 years to be rewarded? How many would have stuck with the strategy? For some, it would have been longer than their investment horizon.

Zakamulin did find that the standard deviation of returns of the market-timing strategy is, on average, about 30 percent less than that of the buy-and-hold strategy regardless of the choice of the passive benchmark. Of course, a lower level of volatility should be expected because, about 30 percent of time, the money is allocated to risk-free investment. However, the reduction in risk wasn’t a free lunch. It was accompanied by reduced returns.

Depending on the passive benchmark, the market-timing strategy delivers from 9 to 26 percentage points lower mean returns as compared to that of the buy-and-hold strategy.

Zakamulin concluded that the real-life performance of the market-timing strategies suggest that over a long run, one can expect that a market-timing strategy delivers only marginally better risk-adjusted returns than the buy-and-hold strategy, yet with a substantially lower long-term capital growth.

“Over a long run the buy-and-hold strategy is more risky, but at the same time, it’s more rewarding than the market-timing strategy.” Moreover, he also cautioned that: “The fact that the market timing strategy showed a better risk-adjusted tradeoff in the past does not automatically mean that the same timing strategy will show a better risk-adjusted tradeoff in the future as well.”

Larry Swedroe is director of Research for the BAM Alliance, a community of more than 130 independent registered investment advisors throughout the country.