This thesis studies the sources of the returns from the momentum strategy and attempts to find some hints for the heated debate on the market efficiency hypothesis over the past twenty years. By decomposing the momentum returns from a mathematical model, we investigate directly the contributors and their relative importance in generating these momentum returns. Our empirical results support that autocorrelation of own stock returns is one of the driving forces for the momentum expected returns. The magnitude of the autocorrelation decreases as the ranking period becomes more remote. The second important source comes from the cross-sectional variation of the expected returns in the winner and loser portfolios at a given time. The third important source is the difference of the expected returns between the winner and loser portfolios. To our surprise, the cross-autocovariance does not contribute much to the momentum expected returns. Thus, the lead-lag effect can cause momentum returns, but its impact is not as significant as we had anticipated. More importantly, by changing the weights of the winner and loser portfolios, we find that the own-autocovariance of the winner portfolio is almost negligible, compared to that of the loser portfolio. The returns of the winners are much more random than those of the losers. This asymmetric own-autocovariance found in the return decomposition provides another underlying explanation to the recent finding that the contribution of the winner and loser portfolios to the momentum returns is asymmetric, and it is the losers, rather than the winners, that drive the momentum returns. Therefore, the market may not be as efficient as we believed before.

This thesis consists of three empirical essays studying the capital market implications of the accounting for R&D costs. The first empirical study (Chapter 2) re-visits the debate over the positive R&D-returns relation. The second empirical study (Chapter 3) examines the risk relevance of current R&D accounting. The third empirical study (Chapter 4) explores the joint impact of R&D intensity and competition on the relative relevance of the idiosyncratic part of earnings. Prior research argues that the positive relation between current R&D activity and future returns is evidence of mispricing, a compensation for risk inherent in R&D or a transformation of the value/growth anomaly. The first empirical study contributes to this debate by taking into account the link between R&D activity, equity duration and systematic risk. This link motivates us to employ Campbell and Vuolteenaho (2004)'s intertemporal asset pricing model (ICAPM) which accommodates stochastic discount rates and investors' intertemporal preferences. The results support a risk based explanation; R&D intensive firms are exposed to higher discount rate risk. Hedge portfolio strategies show that the mispricing explanations is not economically significant. The second empirical study contributes to prior research on the value relevance of financial reporting information on R&D, by proposing an alternative approach which relies on a return variance decomposition model. We find that R&D intensity has a significant influence on market participants' revisions of expectations regarding future discount rates (or, discount rate news) and future cash flows (or, cash flow news), thereby driving returns variance. We extend this investigation to assess the risk relevance of this information by means of its influence on the sensitivity of cash flow and discount rate news to the market news. Our findings suggest R&D intensity is associated with significant variation in the sensitivity of cash flow news to the market news which implies that financial reporting information on R&D is risk relevant. Interestingly, we do not establish a similar pattern with respect to the sensitivity of discount news to the market news which may dismiss the impact of sentiment in stock returns of R&D intensive firms. The third empirical study examines the effect of financial reporting information on R&D to the value relevance of common and idiosyncratic earnings. More specifically, we investigate the value relevance of common and idiosyncratic earnings through an extension of the Vuolteenaho (2002) model which decomposes return variance into its discount rate, idiosyncratic and common cash flow news. We demonstrate that the relative importance of idiosyncratic over common cash flow news in explaining return variance increases with firm-level R&D intensity. Extending this analysis, we find that this relation varies with the level of R&D investment concentration in the industry. Those results indicate that the market perceives that more pronounced R&D activity leads to outcomes that enable the firm to differentiate itself from its rivals. However, our results also suggest that the market perceives that this relation depends upon the underlying economics of the industry where the firm operates.