Kazumasa Higashikawa

Osaka University, Japan
E-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.

This study examined the effect of earnings quality on the performance of earnings forecasts using a cross-sectional model. Earnings quality is measured by the residuals of a regression analysis that relates working capital accruals to operating cash flows for the current, previous, and next period (SRESID), the lower the earnings quality, the larger the forecasting errors. This result holds for both bias, defined by the difference between the actual and the forecasted values, and accuracy, measured as the absolute value of bias. The relationship between earnings quality and forecasting errors does not change after controlling for other potential earnings attributes. In addition, the basic conclusion remains the same when SRESID is estimated using a time-series model, and when the look-ahead bias inherent in SRESID is removed. These findings suggest that SRESID is useful for market participants in selecting a relevant earnings forecasting model.

Keywords: earnings quality, cross-sectional forecast model, forecast bias, forecast accuracy