The strategic use of prior-period benchmark disclosures in management earnings forecasts

Abstract

I investigate the way in which Australian managers issue their earnings forecasts, and the impact this has on the reaction of equity investors and security analysts. Using a sample of 233 management earnings forecasts issued from 1994 to 2001, I find that managers are more likely to issue earnings forecasts when they have bad earnings news than good earnings news. I find that a vast majority of forecasts are framed by the use of an accompanying earnings benchmark. Forecasts are issued with varying degrees of specificity (or precision) and also with variation in additional accompanying disclosures. Forecasts issued with negative framing (forecast earnings less than benchmark earnings) are more likely to be accompanied by statements about factors external to the firm in explaining performance, while forecasts issued with positive framing (forecast earnings greater than benchmark earnings) are more likely to be accompanied by additional verifiable forecasts of components of earnings. I find the market reaction to earnings forecasts released with positive framing is higher than for forecasts released with negative framing, after controlling for forecast news and other forecast properties. I also examine security analysts forecasts around the release of management earnings forecasts and find that after the release of a management earnings forecast, analyst activity increases, but that analysts forecasts become less accurate and more biased. Neither the extent of analyst activity nor changes in analysts forecast accuracy or bias is related to forecast framing.