Are stock returns predictable at different points in time?

The question of whether stock returns are predictable is of long-standing interest to both academics and investment practitioners. Commonly accepted investment strategies, for example, will behave quite differently in the presence of stock return predictability. The research literature is unclear on the answer and suggests that return predictability, if it exists, will be difficult to exploit on an out-of-sample basis under any scenario. Other studies find that return predictability is instead, highly unstable and time-varying. However, academic tests are hampered by estimation methods that are linear and produce a constant coefficient using data covering long time horizons. Linear regressions of this sort produce an “average” analysis. That is, they are only able to answer the question of whether returns are predictable on average.

Pockets of Predictability

• Leland E. Farmer, Lawrence Schmidt, and Allan Timmerman
• Journal of Finance
• A version of this paper can be found here
• Want to read our summaries of academic finance papers? Check out our Academic Research Insight category.

What are the research questions?

1. Are stock returns predictable?

What are the Academic Insights?

1. YES, but they occur in pockets across time. For the most part, the authors report that stock returns are unpredictable. However, there do exist points of pockets in time when returns can be predicted. Fortunately, the predictability that does occur is found to be exploitable and economically significant. The authors also develop a theoretical model of valuation that uses expectations about growth in future cash flows. In the model, a gap is allowed to form between expectations and forecasts of actual cash flows. Those gaps give rise to pockets of predictability confined to short-time horizons.   The model uses dividend to price, the risk-free rate, and actual volatility to demonstrate how this happens. See Table II for a statistical description of pockets that demonstrate return predictability.

Why does it matter?

This research presented here moves the narrative on return predictability by using an estimation method capable of identifying shifts in return predictability. Using regressions with time-varying features allows the estimated coefficient to follow the predictability as a function of calendar time.

The results are hypothetical results and are NOT an indicator of future results and do NOT represent returns that any investor actually attained.  Indexes are unmanaged and do not reflect management or trading fees, and one cannot invest directly in an index.

Abstract

For many benchmark predictor variables, short-horizon return predictability in the U.S. stock market is local in time as short periods with significant predictability (“pockets”) are interspersed with long periods with no predictability. We document this result empirically using a flexible time-varying parameter model that estimates predictive coefficients as a nonparametric function of time and explore possible explanations of this finding, including time-varying risk premia for which we find limited support. Conversely, pockets of return predictability are consistent with a sticky expectations model in which investors slowly update their beliefs about a persistent component in the cash flow process.