Ball and Brown (1968) After Fifty Years

Abstract

The Editor commissioned this replication of Ball and Brown (1968) for a special issue of the Pacific-Basin Finance Journal commemorating the 50th anniversary of its publication. We also describe the background to the original paper and its research design, and offer observations on its interpretation, its impact on the literature and practice, some negative consequences, and its relation to papers published around the same time. One of the pleasing attributes of our original paper is that its results consistently replicate, the implication being that the research design and its implementation uncovered a universal relation between accounting earnings and changes in firm values. The current replication is in two dimensions: time; and geography, with an emphasis on Pacific Basin countries. In the USA and in a selection of 16 other countries, annual accounting earnings continue to contain a large proportion of the information that investors trade into share prices over the year, though not in a timely fashion. Post earnings announcement drift, the first acknowledged share market anomaly, continues today despite being reported five decades ago. One change is that reporting lags internationally have shortened on average and their range has narrowed. A notable change in USA data is that the proportion of the information incorporated in share prices that is contemporaneously incorporated in annual accounting earnings has declined, though the conclusions we draw from this decline are more cautious than in Lev and Gu (2016).

Keywords: anomaly, Ball and Brown (1968), earnings, EPS, event study, market efficiency, PEAD, replication, research design, stock returns, timeliness, value relevance

JEL Classification: B26, B41, G12, G15, M41