As will become clear, it is very difficult to construct a coherent, non-catering explanation for why the propensity to initiate dividends is related to the dividend premium, the Citizens Utilities dividend premium, recent initiation announcement effects, and the future relative returns of payers and nonpayers. We consider three classes of explanations: time varying firm characteristics, time varying contracting problems, and catering.
A. Time varying firm characteristics
One possibility is that certain characteristics of the firms in our sample, important to dividend policy, are changing in the background in such a way as to explain the patterns we find. For example, investment opportunities or profitability may be varying over time. A time varying investment opportunities explanation, which we will consider first, goes as follows. If external finance is costly, such as in the environment of rational investors and asymmetric information studied by Myers (1984) and Myers and Majluf (1984), nonpayers with good investment opportunities may not want to initiate dividends. Alternatively, low investment opportunities could also spell free cash flow problems as in Jensen (1986), and firms with poor opportunities may initiate dividends as a reassurance to investors. Under either mechanism, nonpayers initiate dividends not because they are chasing the relative premium on payers but because their investment opportunities are low in an absolute sense.
Of course, the flip side of this explanation is that firms that are currently payers will be more likely to omit if their investment opportunities are high. This predicts a negative relationship between the dividend premium and the propensity to continue paying, not the positive relationship we found earlier. Therefore, the investment opportunities explanation is at most only relevant to the initiation results.
We evaluate the investment opportunities explanation in a few different ways. One test is to control for the level of investment opportunities and see if the dividend premium retains residual explanatory power for dividend policy choices. We consider two potential measures of investment opportunities, the average market-to-book of the set of firms in question and the overall CRSP value-weighted dividend yield. The first and fourth columns in Table 7 show the results. The investment opportunities proxies enter with the predicted signs – nonpayers are less likely to initiate when their average market-to-book is high, and when the overall dividend-price ratio is low. For dividend continuations and new lists, however, these variables enter with the wrong sign. More importantly, the dividend premium coefficient is not much affected.
The investment opportunities explanation also makes similar predictions for repurchases as for dividends, while the catering theory involves only dividends. Therefore we test whether or not the propensity to repurchase is also related to the dividend premium. We construct aggregate time series measures of the propensity to repurchase, defining a repurchaser as having nonzero purchase of common and preferred stock (Compustat item 115). The first useable year is 1972. Whether we measure aggregate repurchase activity as the propensity to repurchase among all firms, or as the propensity to “initiate” repurchases (new repurchasers in year t divided by surviving non-repurchasers), we find that repurchase activity has an insignificant negative correlation with the lagged dividend premium. The propensity to initiate dividends, by contrast, has a correlation of 0.73 over the same 29-year period.
A last test of the investment opportunities hypothesis is to examine the payout ratio and the dividend yield. Time varying investment opportunities lead to variation in the level of dividends, not necessarily the number of firms paying a dividend, as in our initiation and omission tests. We use updated data from Shiller (1989) on earnings and dividends for the S&P 500 over 1963 to 1998 and the CRSP value-weighted dividend yield over 1963 to 1999. Neither the payout ratio nor the dividend yield is significantly correlated to the lagged dividend premium. We also control for the dividend yield directly in the last three columns of Table 7. This actually increases the effect of the dividend premium on the propensity to initiate. The coefficient on the tax variable in Table 7 is discussed below.
These exercises cast doubt on the ability of time varying investment opportunities to explain the dividend premium results, and it is hard to construct a version of this explanation that could address the connection to future relative returns or the CU dividend premium. A more general possibility is that our results arise because our dividend demand measures are somehow related to the cross-sectional distribution of dividend-relevant characteristics within payer and nonpayer samples. As a contrived example along these lines, suppose the variance of investment opportunities among nonpayers increases – for some unspecified reason – whenever the dividend premium increases. Then an increasing propensity to initiate could reflect the fact that a relatively high fraction of nonpayers do not need to retain cash, not that nonpayers as a group are catering to the dividend premium. In this example, the average investment opportunities of nonpayers are being held constant, so the time series exercises in Table 7 would mistakenly attribute the effect to the dividend premium.
We evaluate this explanation by controlling directly for sample characteristics. In particular, we examine whether the dividend premium helps to explain the residual variation in firm-level dividend policy decisions left after controlling for the characteristics studied by Fama and French (2001). They model the propensity to pay as a function of four variables:
where size NYP is the NYSE market capitalization percentile, i.e. the percentage of firms on the NYSE having equal or smaller capitalization than the firm in question in that year. Market-to-book M/B is measured as defined previously, with the slight modification that here we use the fiscal year closing stock price (Compustat item 199) instead of the calendar year close. Growth dA/A in book assets (Compustat item 6) is self-explanatory. Profitability E/A is earnings before extraordinary items (18) plus interest expense (15) plus income statement deferred taxes (50) divided by book assets. The error term u is the residual propensity to pay dividends for a particular firm-year.
The tests proceed in two stages. In the first stage, we follow Fama and French in estimating firm-level logit regressions using these firm characteristics. As before, we examine the propensity to pay separately among surviving nonpayers, surviving payers, and new lists. We also follow Fama and French in estimating specifications that exclude market-to-book – they suggest that the degree to which this variable measures investment opportunities may change over time, and indeed we have been considering the hypothesis that market-to-book is affected by investor sentiment.
In the second stage, we regress the average annual prediction errors on the value-weighted dividend premium:
, where (13)
Explanatory power for the residual propensity to pay would mean that the dividend premium is not affecting dividend policy through the average or the cross-sectional distribution of these four characteristics.17 The regression in (13) is analogous to our earlier time series regressions, such as equation (11). Note that the two-stage approach gives deference to the characteristics variables by allowing the dividend premium to explain only residual variation. And in terms of statistical power, the dividend premium is using only 38 data points to fit, not thousands like the characteristics.
Table 8 shows the results of this exercise. The first stage results indicate that size and profitability have the most robust effects on the propensity to pay, as Fama and French find. The right column shows the second stage results. In general, controlling for characteristics directly, the dividend premium retains statistically significant explanatory power for most subsamples. Comparing these coefficients to our earlier time series results, one can see that controlling for firm characteristics barely affects the propensity to initiate coefficient. It is 3.90 in Table 5, and controlling for characteristics moves it only slightly, and does not affect its statistical significance. We view this as compelling evidence that the dividend premium is not working through a background correlation with the distribution of firm characteristics.
Controlling for characteristics does tend to reduce the effect of the dividend premium among the other samples, however. That characteristics would help to explain omissions might be expected given that omissions are often forced by characteristics such as low profitability. Nevertheless, the dividend premium approaches statistical significance even in this sample, and remains statistically significant in the new list sample.
We can also ask whether the average annual prediction errors predict the relative returns of payers and nonpayers. In other words, we ask whether the non-characteristics-related variation in dividend policy, which is presumably more closely related to catering, also predicts returns. In unreported results, we find that the average prediction errors indeed have comparable or greater predictive power than the raw dividend policy measures. This indicates that our earlier return predictability results also do not reflect a background correlation with firm characteristics.
B. Time varying contracting problems
Another class of alternative explanations involves time varying contracting problems, such as adverse selection or agency. In terms of adverse selection, one could imagine that when nonpayers trade at a low value, this is a particularly important time for them to signal their investment opportunities. Initiating dividends serves as a signal in the models of Bhattacharya (1979), Hakansson (1982), John and Williams (1985), and Miller and Rock (1985). Again, a natural way to evaluate this explanation is to control for the level of nonpayer market-to-book directly. The results in Table 7 indicate that doing so does not diminish the dividend premium effect. Moreover, it is difficult to imagine a rational expectations equilibrium model in which dividend policy choices predict future returns, or would have any natural reason to be correlated with the CU dividend premium.
Agency costs may also vary over time, with high agency costs requiring dividend payments. For example, La Porta, Lopez-de-Silanes, Shleifer, and Vishny (2000) find that dividend policy varies across countries according to the degree of investor protection. If the dividend premium were a simple time trend, this could be a more compelling explanation for our results. As it stands, this explanation requires governance to improve briefly in the late 1960s, deteriorate, and then improve again. Of course, it is possible that variation in investment opportunities and profits might affect agency costs, but we address this in Table 8. Here, one must imagine time varying agency problems that arise independent of firm characteristics.
Process of elimination leads to catering. This theory offers a natural explanation for the relationships between dividend premium measures and supply responses that we document. Here we go a bit deeper, asking what the data reveal about the precise sources of demand for payers. In turn, we consider the possibility that fluctuations in the dividend premium are driven by sharp changes in taxes, transaction costs, institutional investment constraints, and investor sentiment.
Black and Scholes (1974) suggest tax clienteles or transaction costs clienteles as potential drivers of uninformed demand for dividends. Of course, in taking an extreme view of competition among firms, they ruled out the catering theory’s suggestion that such sources of demand could induce significant variation in the dividend premium, as is suggested in Figure 1 and in future relative returns.18 We noted earlier that the 1986 Tax Reform Act had no visible effect on the dividend premium. Here we evaluate a catering-to-tax-clienteles explanation somewhat more thoroughly by using the difference between the top tax rates on personal income and capital gains as a proxy for the tax disadvantage of dividends. We take capital gains rates for 1962-1997 from Eichner and Sinai (2000) and capital gains rates for 1998-1999 and personal rates for 1962-1999 from www.taxplanet.com.19
Returning to Table 7, one can see the effect of the tax disadvantage of dividends. If anything, the propensity to pay dividends is positively related to this variable, not negatively related, and its inclusion does not much affect the dividend premium coefficient. Indeed, even in combination, taxes and the other variables add little explanatory power. (In Panel C, the large t-statistic on taxes disappears when year is included because of similar downward trends in the propensity to list as a payer and the tax disadvantage variable.)
A tax-based source of uninformed demand for dividends also implies a supply response in the level of dividends rather than the number of dividend-paying firms. Diversified investors will be satisfied with a certain amount of dividends in aggregate, regardless of the distribution across firms. In fact, Marsh and Merton (1987) point out that current dividend payers, with high financial slack and modest investment opportunities, are probably the lowest marginal cost source of dividends. So a tax-based explanation for the dividend premium would predict a closer relationship to the payout ratio and the dividend yield than on the number of payers. We find the opposite.
Transaction costs also vary over time, changing the cost of homemade dividends, and perhaps this induces significant changes in uninformed demand for payers. Black (1976) dismisses this argument, pointing out that there are simple institutional solutions to the problem of the small investor’s transaction costs. However, Jones (2001) shows that transaction costs have declined dramatically since the mid-1970s, which coincides with the reduction in the propensity to initiate that we document.20 Jones’s Figure 4 shows the average annual one-way transaction cost for the NYSE, or one half of the bid-ask spread plus commissions. This series is strongly positively correlated with the propensity to initiate dividends, though this comes mostly from a common time trend. The correlation between the detrended variables is not statistically significant. More importantly, in regressions that include both variables, the dividend premium has more statistical significance than transaction costs in explaining the propensity to initiate dividends. Transaction costs also have trouble accounting for the predictability of relative returns unless one allows transaction costs clienteles to vary sharply enough to induce substantial market inefficiency.
Firms could also cater to a mispricing induced by changes in institutional investment constraints. This would also fit naturally into the model. A potentially relevant institutional change was the 1974 ERISA, which may have increased the demand for payers among pension funds by creating a vague “prudent man” investment rule. At the time, investing in a nonpayer might have been considered imprudent. The law was revised in 1979 to allow pension funds to provide venture capital, thus erasing any doubt that nonpayers were acceptable investments. Figure 2 is broadly consistent with these institutional changes. But the dividend premium seems to anticipate the law, peaking two years before ERISA and starting to drop in 1977. ERISA may be part of the story in this period, but we are not aware of dramatic variation in institutional investment constraints that could conceivably explain the variation in the dividend premium in the 1960s and early 1970s.
A final possibility is that investor sentiment affects the demand for dividend-paying shares. Of course, economists are just beginning to understand investor sentiment. This means that the workings of sentiment are less refined by comparison to existing theories, and therefore sentiment explanations are less rejectable by construction, so we consider them after having established that traditional explanations are unable to fully account for the results.21
We outlined two sentiment stories earlier. One was based on the bird-in-the-hand fallacy, and the other on investor growth perceptions. To reiterate, the growth perceptions mechanism holds that a class of unsophisticated investors uses dividend policy to infer a firm’s investment plans. In particular, they infer from a zero-payout policy (controlling for profitability) that the firm wants to reinvest and grow. When the general investment outlook looks good to these unsophisticated investors, they favor nonpayers. If it looks bad, they favor payers.
As a simple test of the growth perceptions mechanism, we compare the closed-end fund discount with the dividend premium. If the closed-end fund discount is a measure of general investor expectations, as proposed by Zweig (1973) and Lee, Shleifer, and Thaler (1991), and if the dividend premium reflects sentiment about growth opportunities through the mechanism just described, then the two series should be positively correlated. The bird-in-the-hand interpretation of sentiment for dividends does not immediately suggest this prediction. Indeed, none of the alternative explanations suggest such a relationship. We gather value-weighted discounts on closed-end stock funds for 1962 through 1993 from Neal and Wheatley (1998), for 1994 through 1998 are from CDA/Wiesenberger, and for 1999 and 2000 from the discounts on stock funds reported in the Wall Street Journal in the turn-of-the-year issues.
Figure 3 shows the relationship between the dividend premium and the closed-end fund discount. They are not perfectly synchronous, but are related. Their correlation is 0.37 with a p-value of 0.02. Figure 3 seems most consistent with the growth perceptions mechanism for dividend sentiment, and more difficult to relate to the bird-in-the-hand story or other explanations. It makes a new connection between two phenomena that are hard to explain within traditional paradigms, closed-end fund discounts and dividends, and it provides some intriguing new evidence that sentiment plays a role in both.