4.1.1.6. Discriminating between competing hypotheses.Rese archers have
used many different research designs to discriminate between the above
four competing hypotheses to explain the weak return–earnings correlation
and why estimated earnings response coefficients are too small compared
to those predicted on the basis of a random walk time series property of
annual earnings.Pr ices leading earnings and the presence of transitory earnings
appear to be the dominant explanations for the modest contemporaneous
return–earnings association and for the observed magnitudes of earnings
response coefficients.A summary of this research will hopefully make this
apparent
Prices leading earnings and the presence of transitory earnings
appear to be the dominant explanations for the modest contemporaneous
return–earnings association and for the observed magnitudes of earnings
response coefficients.
summary of this research will hopefully make this
apparent.
There is another reason for summarizing the above research.In many
applications, researchers choose a research design from among many
alternatives available.To facilitate research design selection in the future, I
summarize the central features of and pros and cons of the research designs
using common notation.The modeling below extends Fama’s (1990) analysis
of the effect of expanding the measurement window for both returns and
earnings (industrial production) on the return–earnings correlation and
earnings response coefficient.
An important distinction between the analysis in Fama (1990) or similar
studies in finance and the return–earnings literature in accounting centers
around the maintained hypothesis and motivation for the studies.In the
finance literature, the maintained hypothesis is that explanatory variables like
industrial production are real, economic, fundamental variables that a
researcher has measured with a reasonable degree of accuracy
The motivation
for their tests is to examine whether time-series or cross-sectional variation in
stock returns is rational (efficient) in the sense that it is largely explained by
economic fundamentals.The alternative hypothesis is that pricing in the
market is not an outcome of the market participants’ rational economic
behavior
The objective of the accounting literature like Ball and Brown (1968)
or Easton et al.(1992) is to assess whether the accounting earnings
determination process captures the factors that affect security prices, with
the maintained hypothesis that capital markets are informationally efficient
So, market efficiency is a maintained hypothesis and whether accounting
captures underlying economic reality that moves the market is tested in the
research (see Patell (1979) for a mathematically elegant treatment of these
issues).