Limit order adjustment mechanisms and ex-dividend day stock price behavior

Unlike the NYSE the Toronto Stock Exchange (TSX) does not adjust prices in the outstanding limit orders upon ex-dividend days. We find that TSX ex-day stock price behavior differs from that upon the NYSE in several lock opener aspects. In each case, the TSX ex-day behavior is consistent with the lack of a limit order adjustment mechanism. Our findings confirm that market microstructure is an important factor that contributes to the observ Canadian ex-day price behavior. Our findings also separate the puzzle of the relatively small ex-day price globule in Canada.

**********

In a completed capital market, the share price following a dividend should fall by means of exactly the amount of the dividend paid upon each share. However, empirical work upon this issue consistently finds that upon average, stock prices actually globule by less than the dividend amount upon the ex-dividend day.

Although there are several explanations for this phenomenon, single major puzzle of the ex-dividend day behavior remains. Lakonishok and Vermaelen (1983 1992) find that the ex-day price-drop-to-dividend ratio upon the Toronto Stock Exchange (TSX) is 40% or les (depending upon whether the price drop is measured from close-to-close or close-to-open). This ratio is a great quantity [i]or[/i] amount of smaller than the relative price globule of 70% to 80% reported in studies of US data. Lakonishok and Vermaelen note that they do not have an explanation for on what account the price drop is thus small in Canada, and consider this finding to be individual of the major unresolved issues in ex-dividend day stock price behavior.

In this article, we tender a resolution to this confound which we base on Dubofsky's (1992) market microstructure explanation of the ex-dividend day price globule behavior. Dubofsky argues that automated ex-day limit order adjustment mechanisms dominion government ex-dividend day stock price behavior. In the US, the NYSE AMEX, and Nasdaq all use the same automated limit order adjustment mechanism upon ex-dividend days.

Our lock opener observation is that the TSX does not have an ex-dividend day limit order adjustment mechanism like those of the NYSE AMEX, or Nasdaq. upon the TSX, limit orders are not at any time adjusted by the exchange upon ex-dividend days. If on the ex-day the transaction prices upon the TSX are influenced by dint of the existing limit order prices, then the size of the ex-day price globule will be biased toward naught We conjecture that this structural difference explains the a great deal of smaller Canadian price-drop-to-dividend ratio ground by Lakonishok and Vermaelen (1983 1992) We find evidence to support this conjecture

In addition to resolving the Canadian ex-day price globule puzzle, our work tests Dubofsky's (1992) and Bali and Hite's (1998) moulds using Canadian data. Although the one and the other models are market microstructure-based explanations of the ex-day price globule behavior, they differ in several lock opener aspects. The major factor in the Dubofsky mould is the limit order adjustment mechanism upon the ex-day, which is different in Canada than in the US. However, tick size, which is the single factor that determines ex-day price globule in Bali and Hite's mould is identical to the US for a large portion of the sample. Therefore, Bali and Hite make the same predictions about ex-day price behavior in Canada as in the US, on the contrary Dubofsky makes different predictions in the sum of two units countries. Our empirical results, that the ex-day price behavior in Canada is significantly different from that in the US in several important aspects (even during the portion of the sample when the tick sizes in the sum of two units countries are identical), support Dubofsky's model

Aside from the market microstructure explanation of the ex-day price globule behavior, two other explanations are the tax hypothesis (see Elton and Gruber, 1970 Robin, 1991 and Graham, Michaely, and Roberts 2003 among many others) and the interaction of taxes and transaction require to be paid [i]or[/i] undergones (Kalay, 1982, and Boyd and Jagannathan, 1994) Lakonishok and Vermaelen (1983 1992) find that the small price globule ratio in Canada is not consistent with the tax hypothesis. The relatively small price globule is also at odds with the transaction require to be paid [i]or[/i] undergones explanation, since that model predicts a larger price globule than the tax hypothesis does.

Therefore, among the three prevailing lines of research upon the ex-dividend day stock price behavior, the sole one that could explain the Canadian ex-day confound is a market microstructure argument. This is exactly the direction we track in this article.

Using the different exchange-mandated limit order adjustment mechanisms of the NYSE and the TSX we present and examine five hypotheses about the difference in the ex-day price behavior across the sum of two units exchanges. We test these hypotheses with all cash dividend distributions to belonging to all equities on the TSX from 1975 to 2002 We compare the TSX arises to findings for US data in Dubofsky (1992) Bali and Hite (1998) and Jakob and Ma (2004)

We report three lock opener findings. First, like Lakonishok and Vermaelen (1983) we find that the price-drop-to-dividend ratio in our larger Canadian sample is a great deal of smaller than that in the US. next to the first in Dubofsky (1992) and Jakob and Ma (2004) the NYSE dividends have average price-drop-to-dividend ratios that are larger than single for several dividend-size groupings, including the clump in which dividends are les than half the tick size. In contrast, we find that in Canada, average price globules are smaller than the dividend for all dividend sizes. Third, upon the NYSE, Dubofsky (1992), Bali and Hite (1998) and Jakob and Ma (2004) find a clear sawtooth-shaped relation between the dividend amount and ex-day price globule but we find no of that kind pattern in our TSX data.