*This article was originally published by The Fair Observer.
Yash Pratap Singh | Market timing has long been the holy grail for investors, and evidence suggests that it works but might be a tricky game to play.
Homo sapiens like predictability and look for patterns. Sometimes these patterns hold, but at other times they fail. This apparent randomness of markets has long puzzled economists, mathematicians, physicians and, most importantly, investors. Clustering illusion, or what Kahneman and Tversky call representative heuristics, highlight the tendency of investors to look for patterns and fall prey to market bias.
Almost all investors predict and time the market before putting their money in. Their decisions are based both on biases and on trends that they can observe and predict. The key question in the age of big data is whether the availability of large computational capability and statistical techniques allow us to observe and predict trends better. Simply put, can markets be timed?
Well, the answer would partially depend on whether one believes that the markets are Markovian in nature and non-random, or whether they have martingale properties where past data doesn’t affect the future prices.
Are Markets Efficient?
Both institutional and retail investors have to make asset allocation decisions. The typical advice usually conferred is to keep a constant 60% of one’s wealth in equities. A constant exposure is sensible if expected returns on the aggregate stock market do not change over time. However, if expected returns vary over time, a constant exposure would not be optimal. Investors should dynamically adjust their exposure to the equities market based on the changing level of expected returns.
Two problems arise with adjusting equity exposure. First, one must be able to reliably forecast returns. Second, the investor must constantly monitor his/her forecast and change their positions accordingly. These two issues make managing any equity market allocation a difficult proposition.
Forecasting market returns is not an easy task. The Efficient Market Hypothesis (EMH) states that it is difficult, but not impossible, to forecast time-varying market returns. Robert Merton and Paul Samuelson, two Nobel laureates, think it is nearly impossible to forecast returns. On the other hand, Nobel laureates Eugene Fama and Robert Shiller have proposed various predictor variables. In the midst of this debate, the investor is left confused about how to forecast returns.
Efficient or not, markets might be predictable
Let us assume Fama and Shiller are right. It is possible to forecast returns. In such a situation, investors must constantly monitor forecasts and constantly adjust their equity exposure. While professional money managers may have the capacity to do so, retail investors would have to spend an exorbitant amount of time to do so. Translating return forecasts into suitable equity market positions is not a trivial task. In fact, research suggests that retail investors tend to buy stocks that are attention-grabbing. This implies that timing the market is likely to benefit large institutional investors instead of small retail ones.
Blair Hull, a highly successful proprietary trader, shares the beliefs of Fama and Shiller. He believes that it is possible to forecast returns and time the market. He certainly has the pedigree to suggest he might be onto something. Hull founded Hull Trading Company, ran it for 15 years and sold it for a princely $531 million to Goldman Sachs in 1999. He has been working on a market-timing strategy that would provide superior equity market exposure compared to buy-and-hold since 2013.
Hull’s idea centers around managing equity exposures based on statistical evidence of return predictability. There is a plethora of academic work on this subject, and a large set of forecasting variables have been proposed. Hull combines these variables using modern statistical techniques to produce better forecasts and construct market-timing portfolios. Those who take advantage of market timing can achieve higher returns with less risk, while receiving the same diversification benefits of having equity in one’s portfolio.
Trading Return Predictability
There has been a long-standing debate on forecasting market returns by both academics and practitioners. It has been traditionally considered nearly impossible for a fund manager to implement a market-timing strategy. However, four significant recent changes make market-timing a reality. First, many academic research papers have demonstrated the predictive power of carefully chosen variables, strengthening the statistical evidence in favor of return predictability. Second, there has been an explosion of financial and non-financial data. Third, modern statistical techniques have been developed to take advantage of big data and to enhance the reliability of statistical results. Finally, the rise in the number of foreign investors has added more synchrony to global markets despite increasing diversity.
Blair Hull and Xiao Qiao, a PhD in finance from the University of Chicago, Booth School of Business, take advantage of these developments to illustrate how to implement a market-timing strategy in their paper “A Practitioner’s Defense of Return Predictability.” Hull and Qiao enlist 20 variables from the academic literature and combine them to form a trading strategy. As per their paper, annual returns from a simulated strategy earned 12.11% per year from 2001 to 2015. S&P 500 offered a return of 5.79%. More importantly, Hull and Qiao report achieving more than twice the returns of S&P 500 with lower volatility. The market-timing strategy has a Sharpe ratio of 0.85, four times that of the S&P 500 in the same period. The Sharpe ratio is the industry standard for calculating risk-adjusted returns and was developed by Nobel laureate William F. Sharpe. Hull and Qiao also take into consideration transaction costs and taxes to make their setting as realistic as possible. They make the claim that it is possible to implement a successful market-timing strategy.
Possible But How Doable?
Hull and Qiao’s research suggests that market timing might not just be a myth. So, is it now time to rethink market timing, and perhaps absolve it from its poor reputation?
Hull believes so. “Just as it was considered irresponsible to time the market in the last 30 years, it will be considered irresponsible NOT to time the market in the next 30 years,” says Hull.
There is a wrinkle though. As per Rick Ferri, accurate market timing is hard because it requires an individual to understand when to seize the opportunity and when to leave the market. Ferri divides market timing into intentional and unintentional timing. “Intentional timing is based on fundamental and technical factors to determine when asset classes are attractive and when they are not, and then bets are placed accordingly. Unintentional timing is behavioral and it’s rooted in a natural fear and greed mechanism that we must learn to control.”
Investors have always been emotional. From the days of the South Sea Bubble to the Wall Street Crash of 1929, investors have let their hopes or fears define their decisions. Apparently, investors are again becoming fearful and stashing cash under their mattress instead of investing in a bear market.
This makes us ponder and wonder if market timing, though tempting and possible, might be risky and difficult to implement. Paul Merriman, a finance analyst, points out that market timing relies primarily on documented trends and is a short-term investing strategy. It is not value investing of the Warren Buffet kind. Merriman is suspicious of market timing and warns investors about market predicting pundits.
Clearly, market timing can be both beneficial and dangerous. Investors like Hull and Qiao, with the right knowledge and expertise, might be able to make a killing, but simpler souls might lose the shirts off their back. Still, market timing is here to stay and investors might have no alternative but to learn its techniques.
*This article was originally published by The Fair Observer.