Equity Trend Following Strategies are Shams

Posted on: 09/20/2016

iSectors Chuck Self Chief Investment Officer

By Chuck Self iSectors Chief Investment Officer

Equity trend following or momentum strategies and funds have become very popular in the past decade.  A number of providers, large and small, have created these funds to attract investors that moved away from actively managed strategies. These strategy providers tout their quantitative and objective methods utilizing simple algebraic processes.  Many of these strategies performed very well in the 2007-2009 US equity bear market and have been sold based on back tests or actual performance during this period.

Unfortunately, the performance of these strategies has been very disappointing in the seven year bull market we’ve recently experienced. One only has to research the five year results of equity momentum funds to see that they have lagged behind their benchmarks before management fees.  Adding management fees subtract from these poor results.  The statistics from these strategies have contributed to the recent concerns about the effectiveness of liquid alternative funds.

Most investors do not have access to the academic research indicating that trend following does not work in the equity market over time, including:

  • Studies by authors such as Fung and Hsieh indicating that trend following works best in the commodity, currency and fixed income markets but not in the equity market. The non-equity markets are subject to government intervention that prolongs trends.  Equity markets are relatively free and do not have one force driving the direction of the asset class.
  • Studies by authors such as Hutchinson and O’Brien indicating that equity market trend following works best in extreme market moves (positive and negative). As we know the 2007-2009 bear market was the worst stock market decline since the Great Depression.  To base an investment strategy mainly on this market move is folly.  Over time, the US equity market rises as evidenced by the new price highs standard indices have achieved this year. Many trend following strategies reduce downside capture but, unfortunately, reduce upside capture also. An effective strategy will reduce downside risk while taking advantage of rising stock markets.
  • Studies by authors such as ap Gwilym, Clare, Seaton, and Thomas indicating that whatever advantage trend following processes captured in the equity market has diminished over time. The discovery and exploitation of equity market trend following by researchers and market professionals as computer power increased has led to a poorly performing strategy in the 21st
  • If costs that are not always included in academic research such as management fees, transaction expenses and capital gain taxes are subtracted from returns, simple mathematically-based trend following or momentum strategies are found wanting. Authors such as Frazzini, Israel, and Moskowitz (FIM) have created methods to minimize trading costs for their momentum studies. It is not clear that most trend following strategies utilize these methods.  But the resulting Sharpe Ratios (excess returns adjusted for volatility) for FIM’s US equity screening of momentum are in an unimpressive range of 0.11 to 0.21 given that the raw momentum Sharpe Ratio is 0.08.

Recommendation: We sympathize with investors that have become disenchanted with actively managed equity funds but yet want some portion of their equity portfolio exposed to attractive areas of the market in an objective manner.  We propose that investors look for strategies that have the following characteristics:

  • The utilization of 21st century mathematical and statistical processes and computational power to systematically assess data provided to a model.
  • A reasoned and quantitatively studied schema of investments utilized by the model. In order to be successful, it is likely that there needs to be the possibility for allocation to assets beyond those found in standard equity indices such as the Standard & Poor’s 500 or the Russell 3000.
  • An objective of the quantitative model needs to be risk control, not benchmark outperformance. If investment results are too volatile (especially on the downside), unsophisticated investors and, even, financial advisors will not stay with the strategy.  This is a cause of “buy high, sell low” behavior.
  • All expenses of the strategy are accounted for. These include management expenses, fees accessed by underlying investments, transaction costs, and potential tax implications.

iSectors Implementation: The iSectors Post-MPT Growth & Moderate Separately Managed Account strategies meet the requirements of the recommendation above. Post-MPT utilizes a sophisticated, proprietary computational model with objective inputs to create a series of optimal asset allocations at different levels of risk (an efficient frontier.)

  • Post-MPT allocations invest in low-correlated sectors comprised of a subset of equity market sectors plus real estate investment trusts, gold stocks, and Treasury bonds.
  • The objective of the model behind Post-MPT is to reduce downside risk, not outperform a benchmark. This results in a likely return pattern that reduces risk when equity sectors are less attractive and increases risk when equity sectors are more attractive.
  • The underlying assets are low cost index funds, which increase diversification to the portfolio.

 

References

ap Gwilym, Owain, Andrew Clare, James Seaton, and Stephen Thomas. “Price and Momentum as Robust Tactical Approaches to Global Equity Investing.” Journal of Investing, 19, 80-92.

Frazzini, Andrea, Ronen Israel, and Tobias Moskowitz. “Trading Costs of Asset Pricing Anomalies,” Working Paper, AQR Capital Management.

Fung, William and David A. Hsieh. “Empirical Characteristics of Dynamic Trading Strategies: The Case of Hedge Funds.” Review of Financial Studies, 10, 275-302.

Hutchinson, Mark C. and John J. O’Brien. “Is This Time Different? Trend-Following and Financial Crises.” Journal of Alternative Investments, 17, 82-102.



Leave a Reply

%d bloggers like this: