Current vs. Permanent Earnings for Estimating Alternative Dividend Payment Behavioral Model: Theory, Methods, and Applications

Marsh and Merton (1987) and Garrett and Priestley (2000) have used aggregated permanent instead of current earnings to estimate aggregated dividend behavior models which were developed by Lintner (1956). Lee and Primeaux (1991) used permanent instead of current EPS to estimate Lintner’s dividend payment behavior model for individual companies. Most recently, Lambrecht and Myer (2012) have theoretically shown that permanent, instead of current, EPS should be used to estimate the dividend payment behavior model for individual companies to avoid measurement error and misspecification of the model.The main purposes of this paper are as follows: (1) theoretically explain why firms generally allocate permanent earnings and transitory earnings between dividend payments and retained earnings; (2) develop alternative methods for decomposing current earnings into permanent and transitory components; (3) empirically estimate alternative dividend payment behavior models by using two alternative permanent EPS estimates for both individual firms and pooled data; and (4) test Lambrecht and Myer’s (2012) theoretical results related to alternative dividend payment behavior models. We find that the average long-term payout ratio is downward biased and the average estimated intercept is generally upward biased when current instead of permanent EPS is used. We also find that the combined model performs well to deal with both measurement errors and specification errors in describing the dividend payment behavior model.