Working Paper: NBER ID: w0496
Authors: Jerry Green
Abstract: The behavior of stock prices around ex-dividend days has been suggested as evidence for tax-induced clientele effects and as a means to estimate the average effective tax rate faced by investors. In this paper these possibilities are examined theoretically and empirically. Theoretically it is shown that the measured price drop per dollar of dividend may provide a biased estimate of the effective tax rate. Looking at the volume of trade around ex-dividend days we show that the conditions under which it would be unbiased are unlikely to hold. Strong evidence, based on a broader database than that used by previous investigators, is presented for the presence of the clientele effect.
Keywords: taxation; ex-dividend day; stock prices; clientele effects; effective tax rate
JEL Codes: H24; G12; G14
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| measured price drop per dollar of dividend (G35) | biased estimate of the effective tax rate (H21) |
| higher dividend yields (G35) | perceived value of dividends among investors (G35) |
| trading volume patterns (C69) | divergence from expected behavior (D91) |
| average effective tax rates of owners (H32) | cannot be accurately estimated using price behavior around ex-dividend days (G14) |