Do Price Discreteness and Transactions Costs Affect Stock Returns? Comparing
Ex-Dividend Pricing Before and After Decimalization (2003).
Graham, J. R., Michaely, R., Roberts, M. R.
, Main Idea
The authors test whether the ex-dividend day anomaly, where stock prices drop
by less than the dividend and ex-day abnormal returns appear, is driven by market
microstructure factors such as price discreteness, bid–ask spreads, and
transaction costs.
If this explanation is correct, then the anomaly should weaken or disappear when
the tick size becomes smaller, especially after decimalization.
Their empirical strategy exploits a natural experiment on the NYSE by comparing
ex-dividend price behavior before and after changes in the tick size regime (from
1/8 to 1/16 and then to decimals).
, Motivation
Ex-dividend day prices do not fall by the full dividend (“ex-day premium < 1”), and
ex-day abnormal returns appear.
The literature offers three main explanations: (1) taxes on dividends vs capital gains,
(2) taxes interacting with transaction costs, and (3) market microstructure - price
discreteness and bid–ask bounce.
This paper uses a natural experiment in U.S. markets:
From 1996–2001, tick size fell sharply (1/8 → 1/16 → decimals) and bid–ask spreads
declined, which provides a direct test of microstructure explanations emphasized by
Bali & Hite (1998) and Frank & Jagannathan (1998).
Ex-Dividend Pricing Before and After Decimalization (2003).
Graham, J. R., Michaely, R., Roberts, M. R.
, Main Idea
The authors test whether the ex-dividend day anomaly, where stock prices drop
by less than the dividend and ex-day abnormal returns appear, is driven by market
microstructure factors such as price discreteness, bid–ask spreads, and
transaction costs.
If this explanation is correct, then the anomaly should weaken or disappear when
the tick size becomes smaller, especially after decimalization.
Their empirical strategy exploits a natural experiment on the NYSE by comparing
ex-dividend price behavior before and after changes in the tick size regime (from
1/8 to 1/16 and then to decimals).
, Motivation
Ex-dividend day prices do not fall by the full dividend (“ex-day premium < 1”), and
ex-day abnormal returns appear.
The literature offers three main explanations: (1) taxes on dividends vs capital gains,
(2) taxes interacting with transaction costs, and (3) market microstructure - price
discreteness and bid–ask bounce.
This paper uses a natural experiment in U.S. markets:
From 1996–2001, tick size fell sharply (1/8 → 1/16 → decimals) and bid–ask spreads
declined, which provides a direct test of microstructure explanations emphasized by
Bali & Hite (1998) and Frank & Jagannathan (1998).
