Fundamental momentum : a new approach to investment analysis

Abstract

The study examined the momentum in the fundamentals of companies over time, and whether the information content in the momentum of the fundamentals improved the understanding of the long-standing price momentum and earnings momentum anomalies on the Johannesburg Stock Exchange (JSE). Fundamental momentum is defined as the difference between the change of a fundamental variable over consecutive time periods. The study included all industrial companies that were listed on the JSE between the period January 1990 and December 2013. The purpose of the study was to investigate whether price momentum or earnings momentum was subsumed by fundamental momentum. Price momentum and earnings momentum are long-standing anomalies that have been widely researched, yet no definitive explanation has been provided in the literature. The objective of the study was to improve the understanding of price momentum and earnings momentum through the analysis of fundamental momentum. The study also provided insight into the persistence of fundamental momentum of earnings. The study tested the profitability of the price momentum, earnings momentum and the fundamental momentum of earnings trading strategies. The research hypotheses were formulated and tested using equal-weighted sort analysis. The sustainability of fundamental momentum of earnings was also analysed. The size and value risk factors were taken into account to ensure that the results were not influenced by such risk factors. The Fama and French three-factor model was employed to test whether the results captured one of these risk effects. The fourth research question investigated whether the fundamental momentum of an underlying component of earnings increased the persistence of the fundamental momentum of future earnings. Earnings were shown to be mean reverting over time, and therefore, the expectation was that positive or negative fundamental momentum of earnings was not sustainable over a prolonged period of time. However, by decomposing earnings into the accrual and cash flow components, and their respective sub-components, the study undertook regression analysis to see whether a specific component of earnings could improve the sustainability of fundamental momentum. The final research question tested whether price momentum and/or earnings momentum was subsumed by fundamental momentum. Two-way analysis was conducted to test whether the strategies captured similar effects. Sort analysis was used by first constructing equal-weighted portfolios based on either price momentum or earnings momentum. Each portfolio was then further subdivided based on the fundamental momentum of earnings. The profitability of the resultant portfolios was then compared with the initial portfolio. The results confirmed that the price momentum and earnings momentum anomalies were present on the JSE for the sample selected for the study. The fundamental momentum of earnings trading strategy was also shown to be a profitable trading strategy for the extreme quintile portfolios. Using the Fama-MacBeth regression methodology, size and value effects were not found to impact the results across all three momentum strategies. A behavioural overreaction or underreaction hypothesis was argued to explain the profitability of the fundamental momentum of earnings strategy. The market was shown to anticipate the earnings surprise that resulted in earnings momentum up to 12 months prior to portfolio formation. Similarly, the market anticipated fundamental momentum of earnings 12 months prior to the earnings announcement. The fundamental momentum of future earnings was shown to be more sustainable when the fundamental momentum of the cash flow component of prior earnings was higher than the fundamental momentum of the accrual component of prior earnings. This result did not give insight into the nominal size effect of the underlying earnings components, rather, it only gave insight into the rates of change of the earnings components. The final result of the study showed that price momentum and fundamental momentum captured different effects. However, the earnings momentum and fundamental momentum results were not as clear cut. Both strategies used a variant of earnings to construct the quintile portfolios and thus it was very plausible that they captured a similar effect. The study contributed to the current literature in a number of ways. A new trading strategy based on the fundamental momentum of earnings was tested. Fundamental momentum of earnings as a trading strategy has yet to be defined; as a result, it has not been researched prior to this study. Given the results, it may be seen as a derivative of earnings momentum. Understanding the sustainability of fundamental momentum of future earnings was also researched. The final contribution of the study was the two-way analysis of price momentum and fundamental momentum, and earnings momentum and fundamental momentum.