No abstract

No abstract

No abstract

No abstract

We examine the 'disappearing dividends' era documented by Fama and French (2001) with respect to the traditional theory of signalling, wherein the positive signal is one of high future cash flows and continued payments. We report several new findings. First, during the disappearing dividends era, dividends vanished not only because they were less frequently initiated-the oft-cited reason-but also because, once initiated, they were less likely to be sustained. Second, we find that although future performance does increase with dividend sustainability, performance is merely average for permanent payers and poor for temporary payers. Third, we find that the market responded favourably to initiations but did not distinguish ex-ante between short-run and longrun payers. Fourth, we find that despite the market's similar treatment of shorter-and longer-term payers, dividend sustainability was in fact predictable out of sample, using information strictly available to investors at the time of the announcement. Fifth, we find that performance is predictable through sustainability; the firms we predict to become permanent payers significantly outperform their counterparts in subsequent years. Overall, our findings run counter to the traditional signalling theory of dividends in terms of both overall firm performance and the market's reaction to initiations.