Ex-dividend day stock price behaviour is a long-standing and unresolved topic that has
received considerable attention in the ﬁnancial literature. In perfect capital markets, one
expects the drop in stock price on the ex-day to be equal to the dividend. However,
empirical studies ﬁnd that the price drop is less than the dividend. There are two
competing explanations for this phenomenon a tax explanation and a market
Elton and Gruber (1970; hereafter E&G) theorise that the ex-day price drop is less than
the dividend because dividends and capital gains are taxed at different rates. Whereas
dividends are taxed at the ordinary income tax rate, capital gains are normally taxed at a
lower rate. The larger the difference between the ordinary income tax rate and the capital
gains tax rate, the smaller will be the ex-day premium (the ratio of the price drop to the
dividend). In special situations where the two tax rates are equal, the tax theory predicts
that the ex-day price drop will be equal to the dividend; and if the ordinary income tax
rate is less than the capital gains tax rate, the price drop will be greater than the dividend.
Cloyd et al. (2006), Karpoff and Walking (1990), Lakonishok and Vermaelen (1986),
Litzenberger and Ramaswamy (1979) and Michaely and Vila (1995) are some of the
studies that provide evidence in support of the tax theory.
A competing explanation for ex-day stock price behaviour is based on market
microstructure effects. Although there are many versions of the microstructure theory,
our study focuses on the tick size theory of Bali and Hite (1998). This theory argues that
the ex-day price drop is less than the dividend because of discreteness in stock pricing
and not because of differential taxation. Stocks trade in integral multiples of the tick size
(e.g., 1/8 and 1/16), whereas dividends are paid in continuous amounts. If a dividend is
not an integral multiple of the tick size, then the price drop will be rounded down to the
nearest tick multiple. This is because investors will not be willing to pay for the dividend
by accepting a price drop that is greater than the dividend. According to this theory, even
in the absence of differential taxation, the price drop will still be less than the dividend.
Among others, Hardin et al. (2002, 2007 and 2012) support this theory.
Using US REIT stock prices, ou r study shows that both tax and micr ostructure
effects are important and th at they jointly inﬂuence ex-day stock pricing. We present a
new theory that shows how thes e two effects interact with each other to produce the
observed ex-day behavio ur. This is our main contrib ution to the literature. This theory
combines the central eleme nts of the tax and microstruct ure theories. In essence, o ur
theory modiﬁes the central idea of the microstruct ure theory taking into accoun t the
tax effects. The microstruc ture theory of Bali and Hite (1998 ) says that investors will
not be willing to pay for a divide nd with an ex-day price drop tha t is greater than the
dividend, and the price drop gets rounded down to the tick jus t below the dividend.
Our theory modiﬁes this prediction. We posit th at differential taxation ef fects ﬁrst act
on the dividend to produce wha t we call the tax-predicte d price drop on the ex-day.
Then, the microstructure ef fects act on the tax-predic ted price drop to produce th e
actual ex-day price drop. Inst ead of the actual dividend being ro unded down to
the tick just below the divide nd as predicted by the microst ructure theory, it is the
tax-predicted price drop that ge ts rounded to the tick just bel ow. This is because
investors will not pay for a dividend with an ex-d ay price drop that is great er than
the tax-predicted price drop. The det ails of this theory are presented in secti on 3.3.
One of the interesting aspec ts of our theory is that it shows wh y empirical studies
© 2016 John Wiley & Sons, Ltd.
342 Kose John, Ravi S. Mateti, Duong Nguyen, Gopala Vasudevan