Accrual anomaly and accounting standards - Evidence from the adoption of IFRS by publicly listed companies in Finland

Abstract

Accrual anomaly was introduced to the financial market and accounting research by Sloan (1996). The anomaly consists of two empirical regularities. (1) The current accrual component of earnings predicts future earnings less well than the current cash flow component, in other words, the "earnings persistency" of the accrual component is lower than that of the cash flow component. (2) Contrary to the efficient market hypothesis, stock prices fail to fully reflect this information contained in the current earnings components; financial markets treat the accrual component of earnings as more persistent and the cash flow component as less persistent than they truly are.

This thesis examines whether the accrual anomaly found mainly in the U.S. stock markets exists in Finland as well, and whether the adoption of IFRS has any positive or negative effects to this particular financial market anomaly. The empirical analysis employs an ordinary least squares regression analysis to discern any over- or underweighting of the earnings components by the financial markets. The sample consists of Finnish publicly listed companies included in the OMX Helsinki Index (HEX), spanning the years 1993-2013.

The preliminary results indicate accrual overweighting for the pre-IFRS sub-period, which vanishes by the introduction of IFRS. The introduction of IFRS to the Finnish institutional setting therefore increases the quality of financial statements, as evidenced by the elimination of accrual overweighting for the post-IFRS sub-period. The results after robustness testing however negate the preliminary results, as accrual overweighting vanishes for the pre-IFRS sub-period when running robust regressions. There is however disagreement among researchers on conducting robustness tests, and indeed most of the research on accrual anomaly does not conduct conventional robustness testing of the results. The interpretation of results and the conclusions to be drawn from them depend on the position taken towards robustness testing in accrual anomaly research. At the least it can be stated that the results of the empirical tests are contrary to establishing a positive connection between accrual anomaly and fair value accounting standards represented by IFRS.