To Follow or not to Follow – An Analysis of the Profitability of Portfolio Strategies Based on Analyst Consensus EPS Forecasts
University of Hagen
ValueWalk's Raul Panganiban interviews Kirk Du Plessis, Founder and CEO of Option Alpha, and discuss Option Alpha and his general approach to investing. Q1 2021 hedge fund letters, conferences and more The following is a computer generated transcript and may contain some errors. Interview with Option Alpha's Kirk Du Plessis
University of Hagen
We measure the profitability of investment strategies relying on analyst forecasts. We introduce two measures of stock undervaluation, which relate analyst forecast momentum to the contemporaneous stock returns. Employing a 35-year window and a highly liquid US stock universe, we analyze portfolio strategies based on these two measures and show that self-financing trading strategies yield annualized Carhart alphas of up to 8.8%. These strategies clearly outperform existing pure earnings forecast momentum strategies.
To Follow or not to Follow – An Analysis of the Profitability of Portfolio Strategies Based on Analyst Consensus EPS Forecasts – Introduction
Financial analysts working in sell-side research departments of banks and related institutions act as important information intermediaries on capital markets, transforming private
and public information into a broad variety of forecasts, including buy-sell-hold recommendations, target prices, or more fundamental forecasts of a company’s sales, revenues or
earnings, etc. Theoretically, analysts should reduce agency conflicts between a company’s stockholders and management, while speeding up the price discovery process and finally increasing informational market efficiency. However, the existence of analysts conflicts with the efficient market hypothesis (EMH) as suggested by Fama (1970), which requires
that stock prices always fully reflect all available information. This implies that analysts are obsolete on informationally efficient markets, as all relevant information is already incorporated in stock prices at any time.
Since Fama’s seminal work, an extensive strand of empirical literature discovered various anomalies suggesting that financial markets are not always strictly efficient. The existence of price and earnings momentum effects indicates that price discovery on financial markets does not happen instantaneously as suggested by the EMH, but requires a reasonable amount of time. The earnings forecast momentum effect brought up by Givoly and Lakonishok (1980) describes the phenomenon that stock prices tend to drift in the direction of foregoing earnings forecast revisions. To prove the existence of earnings forecast momentum, research in this field applies a multitude of different approaches, such as percentage revisions of the earnings estimate, earnings revisions standardized by their foregoing standard deviation or a variety of ratios as the number of up revisions versus the number of down revisions making up the consensus earnings estimate. However, all these approaches are solely based on the earnings forecast itself.
In this paper, we analyze a measure related to earnings forecast momentum, which is however also related to the stock price development, which makes it actually a measure of undervaluation. Basically, the proposed measure relates the consensus EPS forecast revision to the market valuation revision (that is, the stock return). The basic approach is therefore similar to earnings forecast momentum strategies, but the measure does not actually capture forecast momentum. Instead, the approach quantities the degree to which a stock is undervalued by the market relative to the analyst valuation, based on forecast revisions and stock returns. Identified is therefore primarily that part of analyst expectations that has not yet been priced by the market and that is consequently not yet reflected in market’s stock prices. If analyst EPS forecasts are informative, trading strategies based on the measure of undervaluation should be protable and earn significant excess returns after adjusting for well-known risk factors.
We analyze the profitability of trading strategies relying on the new measures within the Carhart (1997) 4-factor framework, based on monthly portfolio re-formation. To reduce trading costs, we use a highly liquid US stock universe consisting of the S&P 100 index members within a long 35-year window starting in February 1978 and ending in December 2013.
See full PDF below.