# Who wants to trade around ex-dividend days?

We examine order flows around ex-dividend dates on the Taiwan Stock
Exchange. By using Taiwan's data, we can examine the heterogeneity of investors' behavior around ex-dividend dates to test different
hypotheses. We find that for both taxable and nontaxable samples, small
investors sell before the ex date and buy from the ex date, suggesting
that small investors prefer low-priced stocks. We find weaker evidence
that tax-disadvantaged foreign and large domestic investors avoid
participating in taxable dividends. We find strong evidence that
tax-neutral institutions play the role of short-term arbitrageurs around
ex-dividend dates.

In this article, we examine order flows around ex-dividend dates (ex date) on the Taiwan Stock Exchange. Not only does Taiwan's tax code allow us to separate the tax hypothesis from other explanations, but Taiwan's data also permits us to examine the heterogeneity of investors' behavior around ex dates.

Taiwanese companies pay both stock and cash dividends. There are two types of stock dividends, and they differ for both accounting and tax purposes. For accounting, the source of stock dividends can come from capital surplus or retained earnings. If the source is capital surplus, then the stock dividend is nontaxable. If the source is retained earnings, then the stock dividend will be taxable, just like cash dividends. However, since the accounting method has no real effect, the tax consequence is the only difference between the two types of stock dividends. Therefore, Taiwan's data allow us to separate the tax hypothesis from the nuisance hypothesis and the price effect hypothesis. If we examine a sample of stock dividends with capital surplus as the dividend source, then tax has no role to play. If we contrast the nontaxable sample with the taxable sample, then any differences should be due to taxes.

Taiwan's data allow us to examine the heterogeneity of investors' behavior around ex dates. Utilizing intraday order data, we can categorize investors into four groups: foreign, institutional, and large and small individual investors. Different investor groups have drastically different patterns of order submission under the nuisance, the price effect, and the tax hypotheses, even though these hypotheses have similar implications on returns.

Each of the three hypotheses identifies a reason why a certain group of investors would prefer the ex-dividend stock to the cum-dividend stock. The nuisance hypothesis considers the transaction cost associated with odd-lot trading, the price effect hypothesis considers the price drop on the ex date, and the tax hypothesis emphasizes the income tax liability of the dividend. Previous research tests some of these implications using trading volume data (Koski and Scruggs, 1998 and Kryzanowski and Chang, 1996). We go one step further and use order imbalance data. Studying the pattern of order imbalance across investor groups provides a powerful test for alternative hypotheses and improves our understanding of the ex-date phenomenon.

The article proceeds as follows. Section I introduces institutional details of stock dividends and stock trading in Taiwan. Section II develops testable implications. We discuss our sample in section III. Section IV and V report our findings and Section VI concludes.

I. Institutional Details

To provide a background for our sample data, we first introduce institutional details on the dividend tax and the trading mechanism in Taiwan.

A. Dividend Payments and Taxes

There are two types of stock dividends that Taiwanese companies can distribute. Both types are similar in the way the company distributes them, and both are proposed by the board of directors, approved in shareholder meetings, and publicly announced. The dividend amount determines the number of shares stockholders receive. Because stocks have a par value of NT$10 in Taiwan, shareholders receive D/10 shares for each share owned, where D is the announced dividend amount in NT$ and D/10 is the distribution rate.

However, the two types of stock dividends differ in terms of their accounting treatment and tax status. Given a distribution rate of D/10 and N outstanding shares, the amount of D*N is transferred to the paid-in capital item on the balance sheet. The source of the accounting transfer D*N defines the type of the stock dividend: it can be the capital surplus or the retained earnings item on the balance sheet. If its source is capital surplus, then the stock dividend is viewed legally as a distribution of shares and is nontaxable. If the source is retained earnings, then the stock dividend is viewed as a distribution of earnings and is taxable. We call these two types of stock dividends the "nontaxable sample" and the "taxable sample".

According to Article 241 of the Company Act, Taiwan companies can distribute new shares to existing shareholders from capital surplus if the balance of the capital surplus is sufficient for the accounting transfer. Capital surplus is a part of the equity that comes from items that increase the book value of assets, but not earnings.

Based on Article 238, the major sources of capital surplus are the additional paid-in capital, which is the premium of the issuing price of stock above its par value (NT$10), and the increase in asset values due to revaluations, etc. Since this type of dividend is considered a distribution of shares rather than a distribution of earnings, it is not subject to income tax.

According to Article 232 of the Company Act, companies can distribute earnings when their cumulative retained earnings are positive. To distribute earnings in the form of stock instead of cash requires the approval of a shareholder meeting (Article 240). When the source of the accounting transfer is retained earnings, in Taiwan the stock dividend is taxable. In contrast, in many countries a stock dividend is not subject to tax. For example, in the United States, following the Supreme Court decision in 1920 in Eisner v. Macomber, a stock dividend is not generally taxable as ordinary income (Andrews and Wilson, 1971). Retained earnings are taxed as capital gains when shareholders sell their stocks.

Capital gains in Taiwan are tax exempt. Without taxing retained earnings, government tax revenues are limited. As a result, earnings that are retained but distributed with stock (the second type of stock dividend) are subject to ordinary income tax (Article 14 of the Income Tax Act). Furthermore, Article 76-1 of the Income Tax Act stipulates that companies must keep undistributed earnings of less than half of their paid-in capital; otherwise, their shareholders will be taxed as if earnings have been distributed.

Starting in 1998, Taiwan adopted an imputation tax system that taxes corporate earnings and then rebates this tax to shareholders with a dividend tax credit. Similar systems are also used in Australia, Canada, Germany, Italy, Singapore, and Thailand. The purpose of this system is to eliminate the double taxation of corporate income, when it is earned and again when it is distributed to shareholders.

In Taiwan, the imputation tax system works as follows. Let us say a company has earnings before tax, E, and the corporate tax rate is tc. Suppose that the company distributes all the earnings after tax, (1- tc) E, to shareholders as a dividend. The dividend tax credit gives shareholders a credit for taxes already paid by the company. For a shareholder who receives a net dividend D, the dividend tax credit is tc/(1-tc) D. The sum of the net dividend and the tax credit is the gross dividend. The credit rate tc/(1-tc) equals total corporate taxes paid, tc E, divided by total dividend paid, (1-tc) E. Currently, the corporate tax rate is 25% in Taiwan, and so the tax credit rate is 33.33%.

The tax credit rate can be higher than 33.33%. To mitigate tax evasion caused by retained earnings, Article 66-9 of the Income Tax Act levies a 10% tax on retained earnings. Assume a company retains all the earnings after tax: 0.75E. Thus, it must pay, in addition to 0.25E, an extra 0.075E as income tax, and shareholders can only receive 0.675E as dividend D in the future. As a result, the credit rate will be 0.325E/0.675E, or 48.15%.

For shareholders, the stock dividends they receive that are distributed from earnings can be subject to income tax. The tax status of these dividends depends on whether shareholders are foreigners, domestic individuals, or domestic corporations. For domestic corporations, the net dividend received is not included in their taxable income, and the dividend tax credit received is passed onto their shareholders (Article 42 of the Income Tax Act).

If shareholders are domestic individuals, then the gross dividend received is included in their regular taxable income. The dividend tax credit is counted as prepaid tax and serves to reduce the tax liability of shareholders (Article 14 of the Income Tax Act). The marginal tax rate applied to regular taxable income can be 6%, 13%, 21%, 30%, or 40%.

When shareholders are foreigners, either individuals or corporations, they cannot use the tax credit accompanying the dividend. As for net dividends, foreigners will be subject to a withholding tax at the time of payment based on the prescribed tax rates. The withholding tax rates range from 5% to 30%.

B. Trading Mechanism

Because the tax status is different across investors, we use detailed order data to test the tax implications. To better understand the order data used, we first describe the trading mechanism of the exchange.

The Taiwan Stock Exchange (TSE) has no market makers. The exchange is fully computerized and is an order-driven market. All orders are limit orders and the detailed order book is not available to investors.

Despite having no market makers, the Exchange is very liquid. In our sample period 1999, the turnover rate is 288%, second only to the Korea Stock Exchange and Nasdaq.

During our sample period, trading occurs from 9 am to 12 pm Monday to Friday and from 9 am to 11 am every other Saturday. Orders accumulate starting from 8:30 am and unexecuted orders remain on the order book only until the end of the day, unless cancelled.

Trading on the TSE involves two mechanisms: a periodic call auction, which is used to open trading, and a batch call auction, which is used throughout the day. In either a periodic or a batch auction, orders accumulate and the computer sets a single market-clearing price at which all executed orders are transacted. The priority of the order execution depends first on the price and then on the arrival time of orders. Although structured as batch auctions, trading is almost continuous. For our sample firms in 1999, the median time interval between transactions is 62 seconds.

Figure 1 illustrates the determination of the market-clearing price. Suppose that at the instant before the matching t, the demand schedule D gives the number of shares investors are willing to buy, while the supply schedule S gives the number of shares investors are willing to sell at different prices. The market-clearing price, [P.sub.t] ([P.sub.3] in Figure 1) is the price that can maximize the trading volume [Q.sub.t] subject to demand and supply. After the matching, but before any new orders arrive, the best bid ([B.sub.t]) is the highest bid price from unfilled buy orders; the best ask ([A.sub.t]) is the lowest ask price from unfilled sell orders.

[FIGURE 1 OMITTED]

In Figure 1, all buy orders that are willing to pay [P.sub.3] have been filled, and the highest bid price is [P.sub.2]. The lowest ask price from unfilled sell orders is the same as the market-clearing price, [P.sub.3]. After each transaction, the exchange discloses to investors the clearing price [P.sub.t], the trading volume [Q.sub.t], the best bid price [B.sub.t], and the best ask price [A.sub.t].

In our empirical analysis, we use the best bid and ask prices to classify new orders as aggressive or conservative. We define aggressive orders as those that have the highest priority in trading. For orders that arrive between transactions t and t+1, aggressive buy orders have limit prices higher than the prevailing best ask price, [A.sub.t]; aggressive sell orders have prices lower than the prevailing best bid price, [B.sub.t].

We define conservative orders as those that have low priority in matching. The limit price of conservative buy (sell) orders is lower (higher) than the prevailing best bid price [B.sub.t] (best ask price [A.sub.t]).

II. Testable Hypotheses

Previous research has offered several explanations for the ex date phenomena. Most of them also have empirical implications for the pattern of order submission that we develop in this section.

A. Nuisance Hypothesis

Stock dividends can be a nuisance for some investors (Baker, 1958; Grinblatt, Masulis, and Titman, 1984). On the Taiwan Stock Exchange, one round lot is 1,000 shares. For investors who hold 1,000 shares, taking a stock dividend with a 20% distribution rate leaves them with an odd lot (less than 1,000 shares). Trading an odd lot is costly: the exchange requires that an odd lot order can only be submitted after the market closes, and the selling price is discounted 0.5% from the closing price. Therefore, small investors who will end up with an odd lot after the distribution have an incentive either to sell before the ex date or to buy afterwards.

A higher distribution rate can either increase or reduce the investor's incentive to avoid an odd lot. For example, investors who receive a dividend of a 10% distribution rate end up with an odd lot if their holdings are not in multiples often lots. Increasing the distribution rate to 20% can reduce the number of investors who try to avoid an odd lot, a situation that arises only if their holdings are not in multiples of five lots. If we increase the distribution rate further to 30%, the number of investors who will have an odd lot (anyone whose holding is not in a multiple often lots) will be higher rather than lower.

Hypothesis 1: For both nontaxable and taxable samples, the order imbalance for small investors will be negative before the ex date and/or be positive on the ex date and afterwards.

B. Price Drop Hypothesis

One attribute of a stock dividend is that the stock price on the ex date will drop significantly: a stock dividend with a 20% distribution rate prompts the price to drop 16.7%. Practitioners think that the price drop can attract investors with little means to buy stocks in round lots (Lakonishok and Lev, 1987). Black (1986) speculates that noise traders prefer low-priced stocks. Hence, the prediction from the price effect is that noise traders or investors who are subject to a wealth constraint will sell before and buy on or after the ex date. Further, they are more willing to buy when the distribution rate is higher.

Hypothesis 2: For both nontaxable and taxable samples, the order imbalance from noise traders or investors who are subject to a wealth constraint will be negative before the ex date, and positive on the ex date and afterwards. The order imbalance will be larger when the distribution rate is higher.

C. Tax Hypothesis

For the retained earnings sample, stock dividends are taxable. As discussed in the previous section, stockholders need to consider both the tax credit and tax liability (Elton and Gruber, 1970).

The tax credit received is [epsilon]kD, where D is the announced amount of net stock dividend in the local currency, k is the dividend tax credit rate, and e is the fraction of dividend tax credit that can be received by an investor. Based on Taiwan's Income Tax Law, [epsilon] is zero for corporations and foreign investors, and one for domestic individuals. The tax liability faced by an investor is [tau](1+[epsilon]k)D, where z is the tax rate applicable for dividends. For corporations, is zero, because net dividends are exempt from tax. For foreign investors, [tau] is the withholding tax rate for net dividends; for individuals, gross dividends are taxed at the personal income tax rate [tau].

Combining both tax credit and liability, the net dividend tax credit for an investor is [epsilon]kD - [tau] (1+[epsilon]k)D. Ignoring transaction costs, if [epsilon]kD - [tau](1+[epsilon]k)D>0, then investors have an incentive to receive the stock dividend, because it will increase their after-tax income. On the other hand, if [epsilon]kD-[tau](1+[epsilon]k)D<0, then investors who receive the dividend will have a lower after-tax income.

Whether the disposable income will be higher or lower depends on the magnitude of [epsilon], k, and [tau], as listed in Table I. Foreign investors will have a lower after-tax income by receiving stock dividends, because [epsilon] = 0 and [tau] > 0. For domestic individuals, the level of income depends on their personal tax rates. If the tax credit rate is 33.33% and the tax rate is higher (lower) than 25%, then receiving dividends will reduce (increase) investors' after-tax income. For corporations, receiving stock dividends does not affect their after-tax incomes, because both e and z equal zero.

When there are transaction costs and a nonzero expected return, investors' response to stock dividends will depend on the specific trading strategy they use. In the following analysis, we adopt the framework used in Boyd and Jagannathan (1994) and McDonald (2001) to develop specific hypotheses.

We assume a proportional transaction cost and discuss four trading strategies: long arbitrage, short arbitrage, delayed purchase, and delayed sale. Previous studies suggest that investors with a positive net dividend tax credit may prefer the long arbitrage and delayed sale strategies to receive dividends, but investors with a negative net dividend tax credit may prefer the short arbitrage and delayed purchase strategies to avoid dividends.

A long arbitrageur is someone who purchases shares cum dividend and sells them ex dividend. To purchase one share, the total cost is (1+c) [P.sub.-1], where [P.sub.-1] is the cum-dividend price on the day before the ex date and c is the transaction cost of the trading value in percentage terms. The net revenue from selling shares after distribution is (l-c) (1+d) E[[P.sub.0]], where E[[P.sub.0]] is the expected ex-dividend price on the ex date and d is the distribution rate of the stock dividend with the dollar amount D (d=D/10). (In reality, new shares distributed as a stock dividend are not available for sale immediately. To derive the expression we assume that short selling is costless.)

Including cost, revenue, and tax, the after-tax gain for a long arbitrageur is (1-c) (1+d) E[[P.sub.0]] - (1+c) [P.sub.-1] + [epsilon]kD - [tau](D+[epsilon]kD). Therefore, investors will use the long arbitrage strategy if the following conditions hold:

(1+d)E[[P.sub.0]]-[P.sub.-1]/[P.sub.-1] > 2c/(1 - c) - 1/(1 - c) D/[P.sub.-1] [[epsilon]]k - tau] (1+[epsilon]k)]. (1)

The left-hand side of the inequality is the stock return adjusted for the distribution effect.

As previously discussed, one direct implication of Equation (1) is that the incentive to pursue a long arbitrage strategy increases with the investors' net dividend tax credit [epsilon]kD - [tau] (l+[epsilon]k)D. Given the transaction costs, a long arbitrage will not be profitable unless the rate of return passes the threshold. The threshold is increasing to c to cover transaction costs, while the threshold is decreasing to the net dividend tax credit. When the net dividend tax credit is negative, the price increase on the ex date must be large enough to cover the tax disadvantage. When the net dividend tax credit is positive, a long arbitrage can still be profitable even if the price drops on the ex date.

A short arbitrageur is someone who sells shares cum dividend and buys them back ex dividend. Because the Exchange prohibits short selling over a five-day period starting from five days before the ex date, a short arbitrageur must actually own a stock to sell.

To sell one share, the net revenue is (1-c) [P.sub.-l]. To buy back the share and its dividend, the total cost is (1+c) (1+d) E[[P.sub.0]]. The after-tax gain for a short arbitrageur is (1-c) [P.sub.-1] - (1+c) (1+d) E[[P.sub.0].

Therefore, investors will use the short arbitrage strategy if the following condition holds:

(1+d)E[[P.sub.0]]-[P.sub.-1]/[P.sub.-1] < -2c/(1+c). (2)

Given the transaction costs, a short arbitrage will not be profitable unless the price drop passes the threshold. We note that the threshold for a short sale is not related to the net dividend tax credit [epsilon]kD - [tau] (1+[epsilon]k)D, because by selling shares in advance, investors are not subject to any dividend taxes.

McDonald (2001) suggests a different kind of short arbitrage that combines a stock loan with selling shares. Since foreign investors cannot receive the tax credit, they may be able to lend their shares to others and recover them ex-dividend. In return, the borrower may pay a fraction of the dividend tax credit. The borrower can then sell the stock to someone who can receive the dividend tax credit. However, the regulations in Taiwan do not allow this kind of short arbitrage. First, stock lending by foreign investors must go through financial institutions, because the law requires that their holdings be kept in the Taiwan Securities Central Depository. Second, lending stock through financial institutions is prohibited during a five-day period that starts from five trading days before the ex date. Therefore, there is no way foreign investors can receive the tax credit by lending to someone who can.

I classify the last two strategies as implicit arbitrage. These strategies are used by anyone who can decide the timing of trading. For either buyers or sellers, they can trade shares cum dividend or ex dividend. For a buyer who purchases shares ex dividend, the cost is (1+c) (1+d) E[[P.sub.0]]; the cost is (1+c) [P.sub.0] - [epsilon]kD + [tau] (D+[epsilon]kD) if the purchase occurs cum dividend. As a result, investors will delay a purchase if the inequality (3) holds; otherwise, they will purchase cum dividend.

(1+d)E[[P.sub.0]] - [P.sub.-1]/[P.sub.-1] < - 1/(1+c) D/[P.sub.-1] [[epsilon]k - [tau]](1+[epsilon]k]. (3)

A seller can choose to sell shares cure dividend or to sell ex dividend. To sell ex dividend, the income is (1-c) (1+d) E[[P.sub.0]] + [epsilon]kD - [tau](D+[epsilon]kD); to sell cure dividend, the income is (1-c) [P.sub.-1].

Therefore, investors will delay selling if:

(1+d)E[[P.sub.0]] - [P.sub.-1]/[P.sub.-1] > - 1/(1+c) D/[P.sub.-1] [[epsilon]k - [tau]](1+[epsilon]k]. (4)

Figure 2 illustrates these four strategies. In this figure we draw four straight lines to represent arbitrage conditions (1) to (4). The horizontal axis is the net dividend tax credit for each dividend in dollar amounts and the vertical axis is the expected stock return on the ex date.

[FIGURE 2 OMITTED]

From Figure 2, we see that the choice of each investor's strategy will

depend on the expected return on the ex date. When the expected return on the ex date is higher, more investors will use the long arbitrage and sell delay strategies, and fewer investors will find the buy delay strategy worthwhile. As a result, there will be more sell orders and fewer buy orders on the ex date, which puts an upper limit on the expected return on the ex date. The same reasoning suggests that the expected return on the ex date cannot be too low, either. The exact equilibrium price on the ex date will depend on the distribution of orders along the net dividend tax credit dimension as stated in the following lemma.

Lemma 1: Assuming that more wealth belongs to investors who have higher tax rates (negative net dividend tax credits), the expected return on the ex date is positive.

Proof: When the expected price change is zero, investors can choose to defer their purchase, defer their sale, or execute a long arbitrage strategy. Given that investors who choose to defer a purchase are those in high tax brackets, these investors are likely to be the majority in the stock market. Therefore, if the expected return is zero, there will be more buy orders than sell orders on the ex date. To clear the market, the expected price change must be positive so as to reduce buy orders on the ex date.

Given a positive expected return, we develop the following hypothesis for tax-related trading.

Hypothesis 3: For the taxable sample,

(1) High tax bracket investors will delay their purchase until the ex date. Therefore, their order imbalance is negative before the ex date and positive on the ex date or afterwards.

(2) Foreign investors will delay their purchase until the ex date. Therefore, their order imbalance is negative before the ex date and positive on the ex date or afterwards.

(3) Low tax bracket investors will adopt the long arbitrage strategy or delay their sales until the ex date. Therefore, their order imbalance is positive before the ex date and negative on the ex date or afterwards.

Since investors with wealth constraints tend to be in low tax brackets, the tax hypothesis makes exactly the opposite prediction to both the nuisance hypothesis and the price drop hypothesis.

Another factor that will affect investors' incentive to trade is the dividend distribution rate. From Equation (3), given a positive expected return on the ex date, it is evident that more investors will delay their purchase when the dividend becomes higher. Therefore, the expected return will change when the distribution rate rises, which we describe in lemma 2.

Lemma 2: When the dividend distribution rate increases, the expected return on the ex date will be higher. More high tax bracket investors will follow the delay purchase strategy. For low tax bracket investors, fewer investors will follow the delayed sale strategy, and there will be no clear-cut prediction on the number of investors following the long arbitrage strategy.

Proof: See Appendix.

Although we do not have clear predictions on trading by low tax bracket investors, the implications on trading by high tax bracket investors are clear. We list these predictions in the following hypothesis.

Hypothesis 4: For the taxable sample, when the dividend distribution rate is higher,

(1) More high tax bracket investors will delay their purchases until the ex date. Therefore, their order imbalance is more negative before the ex date and more positive on the ex date or afterwards.

(2) More foreign investors will delay their purchase until the ex date. Therefore, their order imbalance is more negative before the ex date and more positive on the ex date or afterwards.

D. Tax-Neutral Arbitrageurs

For institutions, receiving stock dividends does not affect their earnings and tax liabilities. They are tax-neutral arbitrageurs and trade for profits. For both taxable and nontaxable distributions, if the expected return is greater than 2c/(1-c), then institutions will buy cum dividend and sell ex dividend. If institutions are selling for other reasons, then if the expected return is positive, they will delay their sales until the ex date. The difference between taxable and nontaxable distributions is that Lemma 1 predicts a positive expected return for the former, but there is no such prediction for the latter.

Hypothesis 5: For the taxable sample,

(1) Institutional investors will delay their sales until the ex date. Therefore, their order imbalance is positive before the ex date and negative on the ex-date or afterwards.

(2) When the dividend distribution rate increases, the expected return on the ex date will be higher. Therefore, institutional investors' order imbalance will be more positive before the ex date and more negative on the ex date or afterwards.

III. Method and Sample

This section presents the method and sample used.

A. Method

To test our hypotheses, we construct daily relative order imbalances for different types of investors. We define the relative daily order imbalance as the difference between buy and sell values divided by the sum of buy and sell values. Using relative instead of absolute order imbalances can mitigate the influence from skewness and extreme values.

I categorize orders submitted by investors as either aggressive or conservative. We define aggressive orders as those that have high priority in matching, and conservative orders as those that have low priority. We define buy orders as aggressive if their limit prices are higher than the best ask, and as conservative if their prices are lower than the best bid. We define sell orders as aggressive if their limit prices are lower than the best bid, and as conservative if their prices are higher than the best ask.

To test for significance, we follow Lakonishok and Vermaelen (1986) to estimate standardized abnormal returns and standardized abnormal relative order imbalances during the event period -2 to +2, where zero is the ex date. Our estimation period is from day -50 to day -6.

Taking the abnormal order imbalance for example, for each sample stock, we first estimate a market model as in Equation (5) using OLS for the estimation period:

[O.sub.ijt] = [[alpha].sub.ij] + [[beta].sub.ij] [O.sub.mjt] + [u.sub.ijt] t = -50,...,-6 (5)

Here, [O.sub.ijt], is the relative order imbalance for firm i from the type-j investor on event day t, and [O.sub.mjt] is the market aggregate order imbalance.

To reduce the influence of extreme observations in estimating market models we delete influential observations using the DFFITS statistics, as suggested by Belsley, Kuh, and Welsch (1980). The DFFITS statistic for the ith observation is a scaled measure of the difference between the predicted value using all observations and the predicted value after deleting the ith observation.

I then define the abnormal order imbalance as follows:

A[O.sub.ijt] = [O.sub.ijt] - [[??].sub.ij] - [[??].sub.ij] [O.sub.mjt] (6)

where AO is the abnormal relative order imbalance. The variable we analyze is the standardized abnormal order imbalance as in Equation (7):

SA[O.sub.ijt] = SA[O.sub.ijt]/(A[O.sub.ijt]) (7)

where [sigma](AO) is the estimated standard error of residuals of the market model.

To test Hypotheses 1, 2, 3 and 5, we calculate the average of SAO and test its significance. To test Hypotheses 4 and 5, we regress SAO against the distribution rate and test the significance of the regression coefficient.

B. Sample Descriptions

Our sample period is 1999, because we have only obtained investors' order data for that year. To ensure a sharp contrast between taxable and nontaxable stock dividends, we include in our sample only those distributions that are either fully taxable or fully nontaxable. We delete any distributions that contain both taxable and nontaxable dividends. We also exclude distributions that are combined with cash dividends or rights issues. Our initial sample comprises 125 stocks.

To estimate abnormal returns and abnormal order imbalances, we require a minimum of 40 days of data in the estimation period from day -50 to -6. This requirement eliminates 13 stocks. We also delete one stock that has a very large distribution rate of 168% (the second largest distribution rate is 50%). The final sample comprises 111 stock dividends, 45 of which are nontaxable and 66 that are taxable.

Table II provides basic descriptions of the companies in our sample. The nontaxable capital surplus sample is quite different from the taxable retained earnings sample. Companies that pay out stock dividends from retained earnings are bigger (median market capitalization is NT$9.6 billion compared to NT$6.8 billion; the exchange rate at the end of 1999 was NT$31.4 = US$1); their stock prices are higher (median closing price on the day before the ex date is NT$30.4 compared to NT$14.4); their distribution rates are larger (median rate is 10% compared to 6%). Because the Company Act requires profitability as a prerequisite for companies to distribute earnings, companies in the retained earnings sample are also more profitable (median ROA is 7.7% compared to 1.5%).

The difference between the two samples in firm characteristics should not cause any systematic bias in our results. It is likely that differences in firm characteristics will show differences in the composition of investors and differences in their order strategies. However, it is highly unlikely that differences in firm characteristics will cause differences between the period before the ex date and the period around the ex date. Since we are testing the significance of an abnormal order imbalance (that is, the difference between the period around the ex date and the estimation period) rather than the significance of raw data, our results should not be biased.

We obtain order data from the Taiwan Stock Exchange. This file contains all orders submitted by investors to the TSE during 1999. The detailed information includes the time, investor type, order type (buy or sell), volume, and limit price. Due to availability, we can only examine order data rather than trading data. Nevertheless, using order data has the benefit of being better able to reflect investors' intentions. Using trading data requires the existence of the other side of the trade.

To test our hypotheses, we categorize investors by four groups of investors: foreign investors, domestic institutions, large individual investors, and small individual investors. We classify individual investors who submit a daily total order value greater than NT$200,000 as large investors; otherwise they are small investors. NT$200,000 is approximately half of the annual GNP per capita. Using this criterion causes more than half of all investors to be classified as small. (I note that in 1999, the median daily total order values submitted by individual investors was NT$114,000, and the GNP per capita was NT$427,097.)

We also report results based on critical daily total order values of NT$100,000 or NT$300,000. We use the order value rather than volume to classify individual investors, because order value is the dollar amount that needs to be invested, and thus should be a better indication of the wealth condition of investors.

I use the small individual investor group to test Hypotheses 1 and 2. Hypothesis 1 makes predictions about small investors who view stock dividends as a nuisance. Hypothesis 2 discusses investors with little means and noise traders who react to the price drop brought on by the ex date. We assume that individual investors whose order values are low are the ones who will view stock dividends as a nuisance or who will react to the price drop per se. Since we do not know the tax rate applicable to each individual investor, we also use the small individual investor group as a proxy for the low tax bracket investors, as discussed in Hypotheses 3 and 4.

IV. Empirical Results

We first examine returns and aggregate order imbalances around the ex date. Then we test our hypotheses by looking at order imbalances across different groups of investors.

A. Abnormal Returns

Table III reports the mean of standardized abnormal returns around the ex date. The average standardized abnormal return on the ex date for the nontaxable sample is 0.6, significant at a 0.05 level. Given that stock dividends paid out of capital surplus are not taxable, this result is similar to that of Eades, Hess, and Kim (1984) for a stock split sample in the US For the taxable sample, the average standardized abnormal return on the ex date is 0.5, also significant at a 0.05 level. Although the taxable sample has a slightly lower average return than the nontaxable sample, the difference is too small to be significant. Our results suggest that the tax is neither a necessary condition for the ex-date phenomenon, nor an important factor, perhaps even completely irrelevant, in explaining the ex-date phenomenon. Of course, our results are based on a small sample and future research is warranted.

When we look at the raw return on the ex date from the overall sample combining 111 nontaxable and taxable distributions, we find that 73 (66%) experience a positive abnormal return, where the average abnormal return is 1.21% and the median is 1.23% (both are significantly different from zero at a 0.01 level).

There are two reasons why arbitrage cannot eliminate abnormal returns. One is that arbitrage is costly. Arbitrageurs have to pay a 0.3% securities transaction tax when they sell, and a two-way commission rate of 0.1425% on the Exchange. After deducting both transaction costs, the average abnormal return shrinks to 0.62% and the median is 0.64%, barely significant at a 0.05 level. The second reason is that arbitrage around the ex date is risky. In our sample, 46 out of 111 stocks (41%) experience a negative abnormal return after cost on the ex date. The risk involved may deter some traders from doing more arbitrage.

We also note in Table III that returns around ex dates are positive and sometimes significant. For the nontaxable sample, the average abnormal return on day 1 is significant and positive, but for the taxable sample, the average is significant and positive on day -1.

Abnormal returns around the ex date are not necessarily associated with abnormal order imbalances. Table IV reports aggregate order imbalances around ex dates. Despite a significant return, there is no significant abnormal order imbalance from day-1 to day 1 for the nontaxable sample. In contrast, for the taxable sample, there are many more buy orders than sell orders from day -1 to day 1. To understand the source of order imbalances, we examine the order behavior of different investor types.

B. Abnormal Order Imbalances

Our hypotheses make specific predictions about order behaviors from foreign investors, domestic institutions, and large and small individual investors. Table V reports the average standardized abnormal order imbalance from day -2 to day 2. Panels A and B give the average order imbalances from aggressive orders for nontaxable and taxable samples; Panels C and D give the average from conservative orders.

For the nontaxable sample, our only predictions are for small individual investors: they want to sell before the ex date and buy afterwards. This finding reflects the nuisance hypothesis (Hypothesis 1) and the price drop hypothesis (Hypothesis 2). The evidence is consistent with our predictions. For small investors, we find that the imbalances from their aggressive orders on days -2 and -1 are both negative, although neither is significant at a 0.1 level (Column 4 in Panel A of Table V). Moreover, small investors turn into net buyers on the ex date. Their average order imbalance is a positive 0.48 on day 0, which is significant at a 0.05 level.

Corroborative evidence for the nuisance and price drop hypotheses comes from the taxable sample. Small investors sell more and sell aggressively before the ex date (Panel B in Table V). The averages of order imbalances are -0.41 and -0.56 for day -2 and -1, and both are significant at a 0.05 level. Then they become net buyers from the ex date. The averages are all positive from day 0 to day 2, and the average on day 0 is 0.56, which is statistically significant. Our evidence on small investors is related to findings in Jakob and Ma (2003). Jakob and Ma examine a sample of 106 distributions from companies listed on the NYSE. They find that order imbalances from small orders are significantly positive on the ex date. The difference is that they examine cash dividends and we look at stock dividends.

The evidence from small investors does not support the tax hypothesis. If small investors are in low tax brackets, they should buy before and sell after the ex date to capture the dividend tax credit. The evidence shows otherwise. It seems that small investors are very determined not to receive a stock dividend. Compared with aggressive orders, the evidence is much weaker for conservative orders submitted by small investors (Panels C and D in Table V). The only significant result occurs on the ex date for the taxable sample, where we find that, similar to aggressive orders, small investors submit a significantly higher number of conservative buy orders. The aggressiveness of small investors does not appear to be rational. The timing of the ex date and its effect on prices are matters of public knowledge. Rational investors should spread their orders to reduce the price impact rather than concentrate their orders.

Our tax hypothesis (Hypothesis 3) argues that foreign and large investors (if they are in high tax brackets) should sell before the ex date and buy afterwards. Table V presents some weak evidence to support the tax hypothesis. Both foreign and large investors submit significantly more conservative buy orders than sell orders on the ex date of the taxable sample. The average order imbalances are 0.37 for foreign investors and 0.45 for large investors (Columns 1 and 2 in Panel D, Table V). In contrast to aggressive trading by small investors, foreign and large investors do not trade aggressively around the ex date relative to the estimation period.

The lack of significance before the ex date may be a rational choice by these tax-motivated investors. Sophisticated investors know the exact timing of the ex date and its tax implications long before day 0 (the time interval between the ex date and the shareholder meeting that decides the dividend distribution rate ranges from 27 to 209 days, and the median is 59 in our sample). Harris (1998) argues that under such a circumstance, uninformed liquidity traders will submit conservative orders when deadlines are distant. Liquidity traders will also spread orders over the whole period to minimize the price impact. Therefore, it may be difficult to detect changes in behavior for tax-motivated investors.

In Table V, we define small and large investors using a daily order value of NT$200,000 as the cut-off point. NT$200,000 is approximately one half of the per capita gross national product in Taiwan in 1999. It also amounts to 14,000 shares for the nontaxable sample or 7,000 shares for the taxable sample based on the median cum-dividend prices of NT$14.4 or NT$30.4 for two samples. To examine the robustness of our choice, we also use NT$100,000 and NT$300,000 as cut-off points. Table VI reports order imbalance results for large and small individual investors based on aggressive orders and different cut-off points. When the cut-off point is NT$100,000, none of the order imbalances for small and large investors is significant. The results for a NT$300,000 cut-off point are similar to results for the NT$200,000 cut-off point. Therefore, the evidence for small and large investors is robust to the choice of the cut-off point.

Institutional investors behave very differently from other types of investors. For the taxable sample in Table V, institutions buy aggressively before the ex date (the average order imbalance is 0.4 on day -1) and start to sell aggressively from day 0 (averaged -0.34). Most abnormal order imbalances are significant. This result is consistent with Hypothesis 5, that institutions delay their sale or pursue a long arbitrage strategy, because dividend taxes are irrelevant for them and the expected return is positive. Lending credence to the tax story is the sharp contrast between the negative order imbalances in the taxable sample from day 0 to day 2 and the positive order imbalances in the nontaxable sample.

Institutions prefer to submit aggressive orders around the ex date. Although the sign of order imbalances is the same, numbers from conservative orders are not significant except on day 1. To be aggressive is a reasonable strategy if institutions are acting as short-term arbitrageurs. Arbitrage is risky and aggressive orders can reduce the uncertainty of failing to trade.

V. Further Results

We find that the pattern of the average order imbalance for the four groups of investors is consistent with our hypotheses. This section further examines the order imbalance from explicit arbitrage activities as well as the relation between order imbalances, distribution rates, and returns.

A. Order Imbalances from Explicit Arbitrage Activities

Table V shows that institutions buy more before the ex date and sell more afterwards, but small investors do just the opposite. These results can arise because investors choose the timing of their trade. They can also arise because investors take explicit arbitrage opportunities. Institutions do long arbitrage and small individual investors conduct short arbitrage. To provide further evidence to differentiate these two possibilities, we directly estimate the extent of arbitrage activities near the ex date.

To test for explicit arbitrage activities, we calculate the abnormal relative order imbalances from arbitrage activities. To estimate the extent of long arbitrage, we first identify those investors who submit both buy orders on day -1 (or -2) and sell orders on the ex date 0. We then calculate the total order value from buy orders to measure the extent of long arbitrage. To estimate the extent of short arbitrage, we identify investors who submit both sell orders on day -1 (or -2) and buy orders on the ex date, then calculate the total order values from sell orders. The relative order imbalances from arbitrage activities are the differences between long and short arbitrages divided by the sum of total buy and sell orders for that day.

To calculate the standardized abnormal arbitrage activities during the event period, we use the average and standard deviation of normal arbitrage activities during the estimation period. To estimate the normal relative order imbalances from arbitrageurs, we apply the same procedure over the estimation period of days -50 to -6. For each day t within the estimation period, we identify those investors who submit both buy (sell) orders on day t-1 and sell (buy) orders on day t, and use the total values from buy (sell) orders as the normal long (short) arbitrage volume. Table VII reports the results for abnormal arbitrage activities.

The table shows strong evidence of long arbitrage activities around the ex date for both the nontaxable and taxable samples. All types of investors do long arbitrage, and the only difference is their aggressiveness. Institutions and individual investors, both large and small, submit aggressive orders to do long arbitrage; foreign investors submit only conservative orders.

The results on explicit arbitrage from small individual investors are very different from results on total orders shown in Table V. When we look at the total order imbalances from small investors, we see that investors in both the nontaxable and taxable samples sell before the ex date and buy afterwards. When we examine only the order imbalances from explicit arbitrage activities, we see that they buy before the ex date and sell afterwards. These results suggest that individual investors are heterogeneous. Some will do the long arbitrage to capture the dividend tax credit and a positive expected return, as suggested in Hypothesis 3, but most will choose to avoid the dividend. The heterogeneity of small individual investors is unlikely to come from their unobserved marginal tax rates, because we can observe similar behaviors for both nontaxable and taxable distributions.

The results of explicit arbitrage from foreign and large individual investors suggest these investors are also heterogeneous. When we examine the order imbalances from explicit arbitrage activities for both nontaxable and taxable samples, we find that foreign and large investors do the long arbitrage. They buy before the ex date and sell afterwards. This behavior suggests that some of the large individual investors and foreign investors want to capture the positive expected return on the ex date. This behavior is not consistent with our tax Hypothesis 3. Both foreigners and high tax bracket investors should avoid receiving the tax-disadvantaged dividends. Although not all foreign and large investors avoid dividends, enough of them do want to avoid the dividend tax. Therefore, in Table V we observe that for the total order imbalances, foreign and large investors do not buy until the ex date.

B. Order Imbalances and Distribution Rates

In addition to average results, we also examine the effect of distribution rates on order imbalances. Hypothesis 5 predicts that for the taxable sample, order imbalances from institutions are more positive before the ex date and more negative afterwards, as the dividend increases. Table VIII provides evidence broadly consistent with this hypothesis.

Table VIII reports the coefficient on the distribution rate in a regression in which the dependent variable is the abnormal order imbalance. In Table VIII, Column 3, Panel B shows that the result is consistent with Hypothesis 5: as the distribution rate increases, institutions submit significantly more buy orders before the ex date and then submit significantly more sell orders from day 0.

Table VIII also examines the behavior of small investors. We find that they behave exactly opposite to institutions. As the distribution rate increases, small investors' order imbalances are more negative before the ex date and more positive from day 0. This result is more consistent with the price drop hypothesis (Hypothesis 2) than with the nuisance hypothesis (Hypothesis 1). The percentage of the price drop is a linear function of the distribution rate. Therefore, the magnitude of order imbalances should be increasing to the distribution rate.

The evidence in Table VIII supports the price drop hypothesis; investors like low-priced stocks, as Black (1986) suggests.

Similar to the average results in Table V, the supporting evidence for the tax hypothesis is weak. Contrary to Hypothesis 4, we find no significant changes in behavior by foreigners. However, when the distribution rate increases, large investors in the taxable sample submit more conservative buy orders on days 0 and 1, as predicted. In contrast, large investors' aggressive orders on day 0 drop significantly.

C. Order Imbalances and Returns

As noted earlier, on the ex date, the average abnormal return is significant and positive, but the average abnormal aggregate order imbalance is not. Therefore, the aggregate order imbalance cannot explain the return behavior. We want to find out if we can explain returns by using order imbalances from various types of investors.

Table IX reports regression results we obtain by using abnormal returns on the ex date as the dependent variable. We include both taxable and nontaxable samples in the regression. The independent variables are a dummy variable for the tax status and order imbalances from four groups of investors. We find that the only significant variable is the order imbalance of large individual investors. The regression shows that when large individuals buy more, the stock price goes higher. Large individual investors not only submit the largest number of shares (Table II reports that their median percentage is higher than 65%), they are also marginal traders.

Despite a significant relation between the abnormal return and the order imbalance of large investors, we still cannot explain the average abnormal return. Without including any order imbalances variables, the intercept is 0.514 (significant at a 0.05 level). The intercept becomes even larger, 0.58, when we include order imbalances from all the four groups of investors. However, we are not surprised by the result, since Table V shows that the average order imbalance of large individuals is not significantly different from zero.

VI. Conclusion

In this article, we examine order flows around ex-dividend dates on the Taiwan Stock Exchange. Not only does Taiwan's tax code allow us to separate the tax hypothesis from other explanations, but Taiwan's data also allows us to examine the heterogeneity of investor behavior around ex dates.

We find strong evidence that small investors sell before the ex date and start to buy from the ex date, which suggests that small investors prefer low prices. We find weaker evidence consistent with the tax hypothesis. Foreign and large domestic investors who are taxdisadvantaged avoid participating in dividends. We also find strong evidence that institutions play the role of short-term arbitrageurs around ex-dividend dates, buying before the ex date and selling afterwards.

Appendix

If the dividend amount is [D.sup.i], then let the associated equilibrium expected return on the ex date be [R.sup.i]. Given [D.sup.i] and [R.sup.i], investors choose a trading strategy based on their after-tax unit dividend income, [epsilon]k - [tau](l + [epsilon]k). We define the critical after-tax income, [x.sup.i,j], as the income that makes investors indifferent between adopting the strategy j and not adopting. Take the long arbitrage (LA) strategy as an example, investors are indifferent toward this strategy if their after-tax income equals [x.sup.i,LA] such that the following equality (Al) holds.

[R.sup.i] = 2c/(1-c) - 1/(1 - c) [D.sup.i]/[P.sub.-1] [x.sup.i,LA](A1)

Similarly, [X.sup.i,DP] and [x.sup.i,DS] are critical values that make investors indifferent toward the DP (delayed purchase) strategy and the DS (delay sale) strategy as in Equations (A2) and (A3).

[R.sup.i] = 1/(1-c) [D.sup.i]/[P.sub.-1][x.sup.i,DP] (A2)

[R.sup.i] = 1/(1-c) [D.sup.i]/[P.sub.-1][x.sup.i,DP] (A3)

To show that the expected return on the ex date will increase with the dividend, we first assume [D.sup.2] > [D.sup.1]. If the expected return stays at [R.sup.1] when the dividend increases from [D.sup.1] to [D.sup.2], then more investors will pursue the delayed purchase strategy and fewer investors will pursue the delayed sale strategy. Fewer investors will use the long arbitrage strategy if the expected return is higher than 2c/(1-c), which is approximately 0.59% in Taiwan, because the maximum commission rate is 0.1425% and the security transaction tax is 0.3%. Both these costs are levied on the seller. Hence, an order imbalance will be positive on the ex date if the expected return stays at [R.sup.1] and the expected return has to increase to a higher level [R.sup.2].

We want to show how the order imbalance for different investors changes when the dividend increases to [D.sup.2] and the equilibrium expected return becomes [R.sup.2]. To do so, we need to know how high [R.sup.2] can be. Can it be high enough to keep the order imbalance the same from investors following the delay purchase strategy?

To keep the order imbalance the same from investors following the delay purchase strategy, we must keep the critical x the same (that is, [x.sup.2,DP] = [X.sup.1,DP]), Hence, [R.sup.2] must satisfy the following equation (see Figure AI):

[R.sup.2] = [[??].sup.2] 1/(1-c) [D.sup.2]/[P.sub.-1][x.sup.1,DP] (A4)

Since [x.sup.1,DP]/(1 + c) = [X.sup.1,DS]/(1 - c) from (A1) and (A2), [R.sup.2] will also satisfy the following equation, and the order imbalance from investors following the delayed sale strategy will stay the same:

Figure Al. Investor's Trading Strategy

If the dividend amount is [D.sup.i], then let the associated equilibrium expected return on the ex date,

(1 + d) E[[P.sub.0]] - [P.sub.-1]/[P.sub.-1]

be [R.sup.i] Given [D.sup.i] and [R.sup.i], investors choose a trading strategy based on their after-tax unit dividend income, [epsilon]k - [tau](1 + [epsilon]k), where k is the dividend tax credit rate, e is the fraction of dividend tax credit that the investor can receive, and [tau] is the tax rate applicable for dividends. Line LA([D.sup.1]) represents expected return/tax credit pairs that investors are indifferent about long arbitrage that purchases on day -1 and sells at 0 given a dividend payment of [D.sup.1]. Line DS([D.sup.E]) represents pairs that investors are indifferent about delay selling from day -1 until date 0 given a dividend of [D.sup.1]. Line DP([D.sup.1]) represents pairs that investors are indifferent about delay purchasing from day -1 until date 0.

[R.sup.2] = 1/(1-c) [D.sup.2]/[P.sub.-1][x.sup.1,DS] (A5)

When [R.sup.2] satisfies (A4), the order imbalance from investors following the long arbitrage strategy will be larger than the order imbalance at [D.sup.1] and [R.sup.1], because (A1), (A2), and (A4) imply the following:

[R.sup.2] = 2c/(1-c) - 1/(1-c) [D.sup.2]/[P.sub.-1][x.sup.1,LA] (A6)

Therefore, when [R.sup.2] satisfies (A4), order imbalances on the ex date will be negative and thus will drive the equilibrium expected return [R.sup.2] lower. Given that:

[R.sup.2] < [[bar.r.].sup.2] (A7)

the critical value x will be higher ([x.sup.2,DP] > [x.sup.1,DP]), more investors will follow the delay purchase strategy, and the order imbalance on the ex date from these investors will be larger. Fewer investors will follow the delayed sale strategy. We can make no clear-cut prediction on the number of investors who will follow the long arbitrage strategy.

References

Andrews, A.W. and L.R. Wilson, 1971, "Stock Dividend Taxation under the Tax Reform Act of 1969: Expansion of an Ominous Past," Arizona Law Review 13, 751-796.

Baker, C.A., 1958, "Evaluation of Stock Dividends," Harvard Business Review July/August, 99-114.

Belsley, D.A., E. Kuh, and R.E. Welsch, 1980, Regression Diagnostics, New York, NY, John Wiley & Sons, Inc.

Black, F., 1986, "Noise," Journal of Finance 41, 529-543.

Boyd, J.H. and R. Jagannathan, 1994, "Ex-Dividend Price Behavior of Common Stocks," Review of Financial Studies 7, 711-741.

Eades, K.M., P.J. Hess, and E.H. Kim, 1984, "On Interpreting Security Returns During the Ex-Dividend Period," Journal of Financial Economics 13, 3-35.

Elton, E.J. and M.J. Gruber, 1970, "Marginal Stockholder Tax Rates and the Clientele Effect," Review of Economics and Statistics 52, 68-74.

Grinblatt, M., R. Masulis, and S. Titman, 1984, "The Valuation Effects of Stock Splits and Stock Dividends," Journal of Financial Economics 13, 461-490.

Harris, L., 1998, "Optimal Dynamic Order Submission Strategies in Some Stylized Trading Problems," Financial Markets, Institutions & Instruments 7, 1-75.

Jakob, K. and T. Ma, 2003, "Order Imbalance on Ex-Dividend Days," Journal of Financial Research 26, 65-75.

Koski, J.L. and J.T, Scruggs, 1998, "Who Trades Around the Ex-Dividend Day? Evidence from NYSE Audit File Data," Financial Management 27, 58-72.

Kryzanowski, L. and H. Zhang, 1996, "Trading Patterns of Small and Large Traders around Stock Split ExDates," Journal of Financial Research 19, 75-90.

Lakortishok, J. and B. Lev, 1987, "Stock Splits and Stock Dividends: Why, Who, and When," Journal of Finance 42, 913-932.

Lakonishok, J. and T. Vermaelen, 1986, "Tax-Induced Trading Around Ex-Dividend Days," Journal of Financial Economics 16, 287-319.

McDonald, R.L., 2001, "Cross-Border Investing with Tax Arbitrage: The Case of German Dividend Tax Credits," Review of Financial Studies 14, 617-657.

We thank Lemma Senbet and Alex Triantis (the former Editors), the anonymous referee, Yehning Chen, Robin Chou, Ravi Jagannathan, Jie-Haun Lee. Suming Lin, Yun Lin, and seminar participants at the National Taiwan University and National Central University for helpful comments. We also acknowledge the financial support from the National Science Council (92-2416-H-002-048-EF).

Shing-yang Hu and Yun-lan Tseng *

* Shing-yang Hu is an associate professor of finance at the National Taiwan University in Taipei, Taiwan. Yun-lan Tseng is an instructor of accounting at the National Pingtung Institute of Commerce in Pingtung, Taiwan.

In this article, we examine order flows around ex-dividend dates (ex date) on the Taiwan Stock Exchange. Not only does Taiwan's tax code allow us to separate the tax hypothesis from other explanations, but Taiwan's data also permits us to examine the heterogeneity of investors' behavior around ex dates.

Taiwanese companies pay both stock and cash dividends. There are two types of stock dividends, and they differ for both accounting and tax purposes. For accounting, the source of stock dividends can come from capital surplus or retained earnings. If the source is capital surplus, then the stock dividend is nontaxable. If the source is retained earnings, then the stock dividend will be taxable, just like cash dividends. However, since the accounting method has no real effect, the tax consequence is the only difference between the two types of stock dividends. Therefore, Taiwan's data allow us to separate the tax hypothesis from the nuisance hypothesis and the price effect hypothesis. If we examine a sample of stock dividends with capital surplus as the dividend source, then tax has no role to play. If we contrast the nontaxable sample with the taxable sample, then any differences should be due to taxes.

Taiwan's data allow us to examine the heterogeneity of investors' behavior around ex dates. Utilizing intraday order data, we can categorize investors into four groups: foreign, institutional, and large and small individual investors. Different investor groups have drastically different patterns of order submission under the nuisance, the price effect, and the tax hypotheses, even though these hypotheses have similar implications on returns.

Each of the three hypotheses identifies a reason why a certain group of investors would prefer the ex-dividend stock to the cum-dividend stock. The nuisance hypothesis considers the transaction cost associated with odd-lot trading, the price effect hypothesis considers the price drop on the ex date, and the tax hypothesis emphasizes the income tax liability of the dividend. Previous research tests some of these implications using trading volume data (Koski and Scruggs, 1998 and Kryzanowski and Chang, 1996). We go one step further and use order imbalance data. Studying the pattern of order imbalance across investor groups provides a powerful test for alternative hypotheses and improves our understanding of the ex-date phenomenon.

The article proceeds as follows. Section I introduces institutional details of stock dividends and stock trading in Taiwan. Section II develops testable implications. We discuss our sample in section III. Section IV and V report our findings and Section VI concludes.

I. Institutional Details

To provide a background for our sample data, we first introduce institutional details on the dividend tax and the trading mechanism in Taiwan.

A. Dividend Payments and Taxes

There are two types of stock dividends that Taiwanese companies can distribute. Both types are similar in the way the company distributes them, and both are proposed by the board of directors, approved in shareholder meetings, and publicly announced. The dividend amount determines the number of shares stockholders receive. Because stocks have a par value of NT$10 in Taiwan, shareholders receive D/10 shares for each share owned, where D is the announced dividend amount in NT$ and D/10 is the distribution rate.

However, the two types of stock dividends differ in terms of their accounting treatment and tax status. Given a distribution rate of D/10 and N outstanding shares, the amount of D*N is transferred to the paid-in capital item on the balance sheet. The source of the accounting transfer D*N defines the type of the stock dividend: it can be the capital surplus or the retained earnings item on the balance sheet. If its source is capital surplus, then the stock dividend is viewed legally as a distribution of shares and is nontaxable. If the source is retained earnings, then the stock dividend is viewed as a distribution of earnings and is taxable. We call these two types of stock dividends the "nontaxable sample" and the "taxable sample".

According to Article 241 of the Company Act, Taiwan companies can distribute new shares to existing shareholders from capital surplus if the balance of the capital surplus is sufficient for the accounting transfer. Capital surplus is a part of the equity that comes from items that increase the book value of assets, but not earnings.

Based on Article 238, the major sources of capital surplus are the additional paid-in capital, which is the premium of the issuing price of stock above its par value (NT$10), and the increase in asset values due to revaluations, etc. Since this type of dividend is considered a distribution of shares rather than a distribution of earnings, it is not subject to income tax.

According to Article 232 of the Company Act, companies can distribute earnings when their cumulative retained earnings are positive. To distribute earnings in the form of stock instead of cash requires the approval of a shareholder meeting (Article 240). When the source of the accounting transfer is retained earnings, in Taiwan the stock dividend is taxable. In contrast, in many countries a stock dividend is not subject to tax. For example, in the United States, following the Supreme Court decision in 1920 in Eisner v. Macomber, a stock dividend is not generally taxable as ordinary income (Andrews and Wilson, 1971). Retained earnings are taxed as capital gains when shareholders sell their stocks.

Capital gains in Taiwan are tax exempt. Without taxing retained earnings, government tax revenues are limited. As a result, earnings that are retained but distributed with stock (the second type of stock dividend) are subject to ordinary income tax (Article 14 of the Income Tax Act). Furthermore, Article 76-1 of the Income Tax Act stipulates that companies must keep undistributed earnings of less than half of their paid-in capital; otherwise, their shareholders will be taxed as if earnings have been distributed.

Starting in 1998, Taiwan adopted an imputation tax system that taxes corporate earnings and then rebates this tax to shareholders with a dividend tax credit. Similar systems are also used in Australia, Canada, Germany, Italy, Singapore, and Thailand. The purpose of this system is to eliminate the double taxation of corporate income, when it is earned and again when it is distributed to shareholders.

In Taiwan, the imputation tax system works as follows. Let us say a company has earnings before tax, E, and the corporate tax rate is tc. Suppose that the company distributes all the earnings after tax, (1- tc) E, to shareholders as a dividend. The dividend tax credit gives shareholders a credit for taxes already paid by the company. For a shareholder who receives a net dividend D, the dividend tax credit is tc/(1-tc) D. The sum of the net dividend and the tax credit is the gross dividend. The credit rate tc/(1-tc) equals total corporate taxes paid, tc E, divided by total dividend paid, (1-tc) E. Currently, the corporate tax rate is 25% in Taiwan, and so the tax credit rate is 33.33%.

The tax credit rate can be higher than 33.33%. To mitigate tax evasion caused by retained earnings, Article 66-9 of the Income Tax Act levies a 10% tax on retained earnings. Assume a company retains all the earnings after tax: 0.75E. Thus, it must pay, in addition to 0.25E, an extra 0.075E as income tax, and shareholders can only receive 0.675E as dividend D in the future. As a result, the credit rate will be 0.325E/0.675E, or 48.15%.

For shareholders, the stock dividends they receive that are distributed from earnings can be subject to income tax. The tax status of these dividends depends on whether shareholders are foreigners, domestic individuals, or domestic corporations. For domestic corporations, the net dividend received is not included in their taxable income, and the dividend tax credit received is passed onto their shareholders (Article 42 of the Income Tax Act).

If shareholders are domestic individuals, then the gross dividend received is included in their regular taxable income. The dividend tax credit is counted as prepaid tax and serves to reduce the tax liability of shareholders (Article 14 of the Income Tax Act). The marginal tax rate applied to regular taxable income can be 6%, 13%, 21%, 30%, or 40%.

When shareholders are foreigners, either individuals or corporations, they cannot use the tax credit accompanying the dividend. As for net dividends, foreigners will be subject to a withholding tax at the time of payment based on the prescribed tax rates. The withholding tax rates range from 5% to 30%.

B. Trading Mechanism

Because the tax status is different across investors, we use detailed order data to test the tax implications. To better understand the order data used, we first describe the trading mechanism of the exchange.

The Taiwan Stock Exchange (TSE) has no market makers. The exchange is fully computerized and is an order-driven market. All orders are limit orders and the detailed order book is not available to investors.

Despite having no market makers, the Exchange is very liquid. In our sample period 1999, the turnover rate is 288%, second only to the Korea Stock Exchange and Nasdaq.

During our sample period, trading occurs from 9 am to 12 pm Monday to Friday and from 9 am to 11 am every other Saturday. Orders accumulate starting from 8:30 am and unexecuted orders remain on the order book only until the end of the day, unless cancelled.

Trading on the TSE involves two mechanisms: a periodic call auction, which is used to open trading, and a batch call auction, which is used throughout the day. In either a periodic or a batch auction, orders accumulate and the computer sets a single market-clearing price at which all executed orders are transacted. The priority of the order execution depends first on the price and then on the arrival time of orders. Although structured as batch auctions, trading is almost continuous. For our sample firms in 1999, the median time interval between transactions is 62 seconds.

Figure 1 illustrates the determination of the market-clearing price. Suppose that at the instant before the matching t, the demand schedule D gives the number of shares investors are willing to buy, while the supply schedule S gives the number of shares investors are willing to sell at different prices. The market-clearing price, [P.sub.t] ([P.sub.3] in Figure 1) is the price that can maximize the trading volume [Q.sub.t] subject to demand and supply. After the matching, but before any new orders arrive, the best bid ([B.sub.t]) is the highest bid price from unfilled buy orders; the best ask ([A.sub.t]) is the lowest ask price from unfilled sell orders.

[FIGURE 1 OMITTED]

In Figure 1, all buy orders that are willing to pay [P.sub.3] have been filled, and the highest bid price is [P.sub.2]. The lowest ask price from unfilled sell orders is the same as the market-clearing price, [P.sub.3]. After each transaction, the exchange discloses to investors the clearing price [P.sub.t], the trading volume [Q.sub.t], the best bid price [B.sub.t], and the best ask price [A.sub.t].

In our empirical analysis, we use the best bid and ask prices to classify new orders as aggressive or conservative. We define aggressive orders as those that have the highest priority in trading. For orders that arrive between transactions t and t+1, aggressive buy orders have limit prices higher than the prevailing best ask price, [A.sub.t]; aggressive sell orders have prices lower than the prevailing best bid price, [B.sub.t].

We define conservative orders as those that have low priority in matching. The limit price of conservative buy (sell) orders is lower (higher) than the prevailing best bid price [B.sub.t] (best ask price [A.sub.t]).

II. Testable Hypotheses

Previous research has offered several explanations for the ex date phenomena. Most of them also have empirical implications for the pattern of order submission that we develop in this section.

A. Nuisance Hypothesis

Stock dividends can be a nuisance for some investors (Baker, 1958; Grinblatt, Masulis, and Titman, 1984). On the Taiwan Stock Exchange, one round lot is 1,000 shares. For investors who hold 1,000 shares, taking a stock dividend with a 20% distribution rate leaves them with an odd lot (less than 1,000 shares). Trading an odd lot is costly: the exchange requires that an odd lot order can only be submitted after the market closes, and the selling price is discounted 0.5% from the closing price. Therefore, small investors who will end up with an odd lot after the distribution have an incentive either to sell before the ex date or to buy afterwards.

A higher distribution rate can either increase or reduce the investor's incentive to avoid an odd lot. For example, investors who receive a dividend of a 10% distribution rate end up with an odd lot if their holdings are not in multiples often lots. Increasing the distribution rate to 20% can reduce the number of investors who try to avoid an odd lot, a situation that arises only if their holdings are not in multiples of five lots. If we increase the distribution rate further to 30%, the number of investors who will have an odd lot (anyone whose holding is not in a multiple often lots) will be higher rather than lower.

Hypothesis 1: For both nontaxable and taxable samples, the order imbalance for small investors will be negative before the ex date and/or be positive on the ex date and afterwards.

B. Price Drop Hypothesis

One attribute of a stock dividend is that the stock price on the ex date will drop significantly: a stock dividend with a 20% distribution rate prompts the price to drop 16.7%. Practitioners think that the price drop can attract investors with little means to buy stocks in round lots (Lakonishok and Lev, 1987). Black (1986) speculates that noise traders prefer low-priced stocks. Hence, the prediction from the price effect is that noise traders or investors who are subject to a wealth constraint will sell before and buy on or after the ex date. Further, they are more willing to buy when the distribution rate is higher.

Hypothesis 2: For both nontaxable and taxable samples, the order imbalance from noise traders or investors who are subject to a wealth constraint will be negative before the ex date, and positive on the ex date and afterwards. The order imbalance will be larger when the distribution rate is higher.

C. Tax Hypothesis

For the retained earnings sample, stock dividends are taxable. As discussed in the previous section, stockholders need to consider both the tax credit and tax liability (Elton and Gruber, 1970).

The tax credit received is [epsilon]kD, where D is the announced amount of net stock dividend in the local currency, k is the dividend tax credit rate, and e is the fraction of dividend tax credit that can be received by an investor. Based on Taiwan's Income Tax Law, [epsilon] is zero for corporations and foreign investors, and one for domestic individuals. The tax liability faced by an investor is [tau](1+[epsilon]k)D, where z is the tax rate applicable for dividends. For corporations, is zero, because net dividends are exempt from tax. For foreign investors, [tau] is the withholding tax rate for net dividends; for individuals, gross dividends are taxed at the personal income tax rate [tau].

Combining both tax credit and liability, the net dividend tax credit for an investor is [epsilon]kD - [tau] (1+[epsilon]k)D. Ignoring transaction costs, if [epsilon]kD - [tau](1+[epsilon]k)D>0, then investors have an incentive to receive the stock dividend, because it will increase their after-tax income. On the other hand, if [epsilon]kD-[tau](1+[epsilon]k)D<0, then investors who receive the dividend will have a lower after-tax income.

Whether the disposable income will be higher or lower depends on the magnitude of [epsilon], k, and [tau], as listed in Table I. Foreign investors will have a lower after-tax income by receiving stock dividends, because [epsilon] = 0 and [tau] > 0. For domestic individuals, the level of income depends on their personal tax rates. If the tax credit rate is 33.33% and the tax rate is higher (lower) than 25%, then receiving dividends will reduce (increase) investors' after-tax income. For corporations, receiving stock dividends does not affect their after-tax incomes, because both e and z equal zero.

When there are transaction costs and a nonzero expected return, investors' response to stock dividends will depend on the specific trading strategy they use. In the following analysis, we adopt the framework used in Boyd and Jagannathan (1994) and McDonald (2001) to develop specific hypotheses.

We assume a proportional transaction cost and discuss four trading strategies: long arbitrage, short arbitrage, delayed purchase, and delayed sale. Previous studies suggest that investors with a positive net dividend tax credit may prefer the long arbitrage and delayed sale strategies to receive dividends, but investors with a negative net dividend tax credit may prefer the short arbitrage and delayed purchase strategies to avoid dividends.

A long arbitrageur is someone who purchases shares cum dividend and sells them ex dividend. To purchase one share, the total cost is (1+c) [P.sub.-1], where [P.sub.-1] is the cum-dividend price on the day before the ex date and c is the transaction cost of the trading value in percentage terms. The net revenue from selling shares after distribution is (l-c) (1+d) E[[P.sub.0]], where E[[P.sub.0]] is the expected ex-dividend price on the ex date and d is the distribution rate of the stock dividend with the dollar amount D (d=D/10). (In reality, new shares distributed as a stock dividend are not available for sale immediately. To derive the expression we assume that short selling is costless.)

Including cost, revenue, and tax, the after-tax gain for a long arbitrageur is (1-c) (1+d) E[[P.sub.0]] - (1+c) [P.sub.-1] + [epsilon]kD - [tau](D+[epsilon]kD). Therefore, investors will use the long arbitrage strategy if the following conditions hold:

(1+d)E[[P.sub.0]]-[P.sub.-1]/[P.sub.-1] > 2c/(1 - c) - 1/(1 - c) D/[P.sub.-1] [[epsilon]]k - tau] (1+[epsilon]k)]. (1)

The left-hand side of the inequality is the stock return adjusted for the distribution effect.

As previously discussed, one direct implication of Equation (1) is that the incentive to pursue a long arbitrage strategy increases with the investors' net dividend tax credit [epsilon]kD - [tau] (l+[epsilon]k)D. Given the transaction costs, a long arbitrage will not be profitable unless the rate of return passes the threshold. The threshold is increasing to c to cover transaction costs, while the threshold is decreasing to the net dividend tax credit. When the net dividend tax credit is negative, the price increase on the ex date must be large enough to cover the tax disadvantage. When the net dividend tax credit is positive, a long arbitrage can still be profitable even if the price drops on the ex date.

A short arbitrageur is someone who sells shares cum dividend and buys them back ex dividend. Because the Exchange prohibits short selling over a five-day period starting from five days before the ex date, a short arbitrageur must actually own a stock to sell.

To sell one share, the net revenue is (1-c) [P.sub.-l]. To buy back the share and its dividend, the total cost is (1+c) (1+d) E[[P.sub.0]]. The after-tax gain for a short arbitrageur is (1-c) [P.sub.-1] - (1+c) (1+d) E[[P.sub.0].

Therefore, investors will use the short arbitrage strategy if the following condition holds:

(1+d)E[[P.sub.0]]-[P.sub.-1]/[P.sub.-1] < -2c/(1+c). (2)

Given the transaction costs, a short arbitrage will not be profitable unless the price drop passes the threshold. We note that the threshold for a short sale is not related to the net dividend tax credit [epsilon]kD - [tau] (1+[epsilon]k)D, because by selling shares in advance, investors are not subject to any dividend taxes.

McDonald (2001) suggests a different kind of short arbitrage that combines a stock loan with selling shares. Since foreign investors cannot receive the tax credit, they may be able to lend their shares to others and recover them ex-dividend. In return, the borrower may pay a fraction of the dividend tax credit. The borrower can then sell the stock to someone who can receive the dividend tax credit. However, the regulations in Taiwan do not allow this kind of short arbitrage. First, stock lending by foreign investors must go through financial institutions, because the law requires that their holdings be kept in the Taiwan Securities Central Depository. Second, lending stock through financial institutions is prohibited during a five-day period that starts from five trading days before the ex date. Therefore, there is no way foreign investors can receive the tax credit by lending to someone who can.

I classify the last two strategies as implicit arbitrage. These strategies are used by anyone who can decide the timing of trading. For either buyers or sellers, they can trade shares cum dividend or ex dividend. For a buyer who purchases shares ex dividend, the cost is (1+c) (1+d) E[[P.sub.0]]; the cost is (1+c) [P.sub.0] - [epsilon]kD + [tau] (D+[epsilon]kD) if the purchase occurs cum dividend. As a result, investors will delay a purchase if the inequality (3) holds; otherwise, they will purchase cum dividend.

(1+d)E[[P.sub.0]] - [P.sub.-1]/[P.sub.-1] < - 1/(1+c) D/[P.sub.-1] [[epsilon]k - [tau]](1+[epsilon]k]. (3)

A seller can choose to sell shares cure dividend or to sell ex dividend. To sell ex dividend, the income is (1-c) (1+d) E[[P.sub.0]] + [epsilon]kD - [tau](D+[epsilon]kD); to sell cure dividend, the income is (1-c) [P.sub.-1].

Therefore, investors will delay selling if:

(1+d)E[[P.sub.0]] - [P.sub.-1]/[P.sub.-1] > - 1/(1+c) D/[P.sub.-1] [[epsilon]k - [tau]](1+[epsilon]k]. (4)

Figure 2 illustrates these four strategies. In this figure we draw four straight lines to represent arbitrage conditions (1) to (4). The horizontal axis is the net dividend tax credit for each dividend in dollar amounts and the vertical axis is the expected stock return on the ex date.

[FIGURE 2 OMITTED]

From Figure 2, we see that the choice of each investor's strategy will

depend on the expected return on the ex date. When the expected return on the ex date is higher, more investors will use the long arbitrage and sell delay strategies, and fewer investors will find the buy delay strategy worthwhile. As a result, there will be more sell orders and fewer buy orders on the ex date, which puts an upper limit on the expected return on the ex date. The same reasoning suggests that the expected return on the ex date cannot be too low, either. The exact equilibrium price on the ex date will depend on the distribution of orders along the net dividend tax credit dimension as stated in the following lemma.

Lemma 1: Assuming that more wealth belongs to investors who have higher tax rates (negative net dividend tax credits), the expected return on the ex date is positive.

Proof: When the expected price change is zero, investors can choose to defer their purchase, defer their sale, or execute a long arbitrage strategy. Given that investors who choose to defer a purchase are those in high tax brackets, these investors are likely to be the majority in the stock market. Therefore, if the expected return is zero, there will be more buy orders than sell orders on the ex date. To clear the market, the expected price change must be positive so as to reduce buy orders on the ex date.

Given a positive expected return, we develop the following hypothesis for tax-related trading.

Hypothesis 3: For the taxable sample,

(1) High tax bracket investors will delay their purchase until the ex date. Therefore, their order imbalance is negative before the ex date and positive on the ex date or afterwards.

(2) Foreign investors will delay their purchase until the ex date. Therefore, their order imbalance is negative before the ex date and positive on the ex date or afterwards.

(3) Low tax bracket investors will adopt the long arbitrage strategy or delay their sales until the ex date. Therefore, their order imbalance is positive before the ex date and negative on the ex date or afterwards.

Since investors with wealth constraints tend to be in low tax brackets, the tax hypothesis makes exactly the opposite prediction to both the nuisance hypothesis and the price drop hypothesis.

Another factor that will affect investors' incentive to trade is the dividend distribution rate. From Equation (3), given a positive expected return on the ex date, it is evident that more investors will delay their purchase when the dividend becomes higher. Therefore, the expected return will change when the distribution rate rises, which we describe in lemma 2.

Lemma 2: When the dividend distribution rate increases, the expected return on the ex date will be higher. More high tax bracket investors will follow the delay purchase strategy. For low tax bracket investors, fewer investors will follow the delayed sale strategy, and there will be no clear-cut prediction on the number of investors following the long arbitrage strategy.

Proof: See Appendix.

Although we do not have clear predictions on trading by low tax bracket investors, the implications on trading by high tax bracket investors are clear. We list these predictions in the following hypothesis.

Hypothesis 4: For the taxable sample, when the dividend distribution rate is higher,

(1) More high tax bracket investors will delay their purchases until the ex date. Therefore, their order imbalance is more negative before the ex date and more positive on the ex date or afterwards.

(2) More foreign investors will delay their purchase until the ex date. Therefore, their order imbalance is more negative before the ex date and more positive on the ex date or afterwards.

D. Tax-Neutral Arbitrageurs

For institutions, receiving stock dividends does not affect their earnings and tax liabilities. They are tax-neutral arbitrageurs and trade for profits. For both taxable and nontaxable distributions, if the expected return is greater than 2c/(1-c), then institutions will buy cum dividend and sell ex dividend. If institutions are selling for other reasons, then if the expected return is positive, they will delay their sales until the ex date. The difference between taxable and nontaxable distributions is that Lemma 1 predicts a positive expected return for the former, but there is no such prediction for the latter.

Hypothesis 5: For the taxable sample,

(1) Institutional investors will delay their sales until the ex date. Therefore, their order imbalance is positive before the ex date and negative on the ex-date or afterwards.

(2) When the dividend distribution rate increases, the expected return on the ex date will be higher. Therefore, institutional investors' order imbalance will be more positive before the ex date and more negative on the ex date or afterwards.

III. Method and Sample

This section presents the method and sample used.

A. Method

To test our hypotheses, we construct daily relative order imbalances for different types of investors. We define the relative daily order imbalance as the difference between buy and sell values divided by the sum of buy and sell values. Using relative instead of absolute order imbalances can mitigate the influence from skewness and extreme values.

I categorize orders submitted by investors as either aggressive or conservative. We define aggressive orders as those that have high priority in matching, and conservative orders as those that have low priority. We define buy orders as aggressive if their limit prices are higher than the best ask, and as conservative if their prices are lower than the best bid. We define sell orders as aggressive if their limit prices are lower than the best bid, and as conservative if their prices are higher than the best ask.

To test for significance, we follow Lakonishok and Vermaelen (1986) to estimate standardized abnormal returns and standardized abnormal relative order imbalances during the event period -2 to +2, where zero is the ex date. Our estimation period is from day -50 to day -6.

Taking the abnormal order imbalance for example, for each sample stock, we first estimate a market model as in Equation (5) using OLS for the estimation period:

[O.sub.ijt] = [[alpha].sub.ij] + [[beta].sub.ij] [O.sub.mjt] + [u.sub.ijt] t = -50,...,-6 (5)

Here, [O.sub.ijt], is the relative order imbalance for firm i from the type-j investor on event day t, and [O.sub.mjt] is the market aggregate order imbalance.

To reduce the influence of extreme observations in estimating market models we delete influential observations using the DFFITS statistics, as suggested by Belsley, Kuh, and Welsch (1980). The DFFITS statistic for the ith observation is a scaled measure of the difference between the predicted value using all observations and the predicted value after deleting the ith observation.

I then define the abnormal order imbalance as follows:

A[O.sub.ijt] = [O.sub.ijt] - [[??].sub.ij] - [[??].sub.ij] [O.sub.mjt] (6)

where AO is the abnormal relative order imbalance. The variable we analyze is the standardized abnormal order imbalance as in Equation (7):

SA[O.sub.ijt] = SA[O.sub.ijt]/(A[O.sub.ijt]) (7)

where [sigma](AO) is the estimated standard error of residuals of the market model.

To test Hypotheses 1, 2, 3 and 5, we calculate the average of SAO and test its significance. To test Hypotheses 4 and 5, we regress SAO against the distribution rate and test the significance of the regression coefficient.

B. Sample Descriptions

Our sample period is 1999, because we have only obtained investors' order data for that year. To ensure a sharp contrast between taxable and nontaxable stock dividends, we include in our sample only those distributions that are either fully taxable or fully nontaxable. We delete any distributions that contain both taxable and nontaxable dividends. We also exclude distributions that are combined with cash dividends or rights issues. Our initial sample comprises 125 stocks.

To estimate abnormal returns and abnormal order imbalances, we require a minimum of 40 days of data in the estimation period from day -50 to -6. This requirement eliminates 13 stocks. We also delete one stock that has a very large distribution rate of 168% (the second largest distribution rate is 50%). The final sample comprises 111 stock dividends, 45 of which are nontaxable and 66 that are taxable.

Table II provides basic descriptions of the companies in our sample. The nontaxable capital surplus sample is quite different from the taxable retained earnings sample. Companies that pay out stock dividends from retained earnings are bigger (median market capitalization is NT$9.6 billion compared to NT$6.8 billion; the exchange rate at the end of 1999 was NT$31.4 = US$1); their stock prices are higher (median closing price on the day before the ex date is NT$30.4 compared to NT$14.4); their distribution rates are larger (median rate is 10% compared to 6%). Because the Company Act requires profitability as a prerequisite for companies to distribute earnings, companies in the retained earnings sample are also more profitable (median ROA is 7.7% compared to 1.5%).

The difference between the two samples in firm characteristics should not cause any systematic bias in our results. It is likely that differences in firm characteristics will show differences in the composition of investors and differences in their order strategies. However, it is highly unlikely that differences in firm characteristics will cause differences between the period before the ex date and the period around the ex date. Since we are testing the significance of an abnormal order imbalance (that is, the difference between the period around the ex date and the estimation period) rather than the significance of raw data, our results should not be biased.

We obtain order data from the Taiwan Stock Exchange. This file contains all orders submitted by investors to the TSE during 1999. The detailed information includes the time, investor type, order type (buy or sell), volume, and limit price. Due to availability, we can only examine order data rather than trading data. Nevertheless, using order data has the benefit of being better able to reflect investors' intentions. Using trading data requires the existence of the other side of the trade.

To test our hypotheses, we categorize investors by four groups of investors: foreign investors, domestic institutions, large individual investors, and small individual investors. We classify individual investors who submit a daily total order value greater than NT$200,000 as large investors; otherwise they are small investors. NT$200,000 is approximately half of the annual GNP per capita. Using this criterion causes more than half of all investors to be classified as small. (I note that in 1999, the median daily total order values submitted by individual investors was NT$114,000, and the GNP per capita was NT$427,097.)

We also report results based on critical daily total order values of NT$100,000 or NT$300,000. We use the order value rather than volume to classify individual investors, because order value is the dollar amount that needs to be invested, and thus should be a better indication of the wealth condition of investors.

I use the small individual investor group to test Hypotheses 1 and 2. Hypothesis 1 makes predictions about small investors who view stock dividends as a nuisance. Hypothesis 2 discusses investors with little means and noise traders who react to the price drop brought on by the ex date. We assume that individual investors whose order values are low are the ones who will view stock dividends as a nuisance or who will react to the price drop per se. Since we do not know the tax rate applicable to each individual investor, we also use the small individual investor group as a proxy for the low tax bracket investors, as discussed in Hypotheses 3 and 4.

IV. Empirical Results

We first examine returns and aggregate order imbalances around the ex date. Then we test our hypotheses by looking at order imbalances across different groups of investors.

A. Abnormal Returns

Table III reports the mean of standardized abnormal returns around the ex date. The average standardized abnormal return on the ex date for the nontaxable sample is 0.6, significant at a 0.05 level. Given that stock dividends paid out of capital surplus are not taxable, this result is similar to that of Eades, Hess, and Kim (1984) for a stock split sample in the US For the taxable sample, the average standardized abnormal return on the ex date is 0.5, also significant at a 0.05 level. Although the taxable sample has a slightly lower average return than the nontaxable sample, the difference is too small to be significant. Our results suggest that the tax is neither a necessary condition for the ex-date phenomenon, nor an important factor, perhaps even completely irrelevant, in explaining the ex-date phenomenon. Of course, our results are based on a small sample and future research is warranted.

When we look at the raw return on the ex date from the overall sample combining 111 nontaxable and taxable distributions, we find that 73 (66%) experience a positive abnormal return, where the average abnormal return is 1.21% and the median is 1.23% (both are significantly different from zero at a 0.01 level).

There are two reasons why arbitrage cannot eliminate abnormal returns. One is that arbitrage is costly. Arbitrageurs have to pay a 0.3% securities transaction tax when they sell, and a two-way commission rate of 0.1425% on the Exchange. After deducting both transaction costs, the average abnormal return shrinks to 0.62% and the median is 0.64%, barely significant at a 0.05 level. The second reason is that arbitrage around the ex date is risky. In our sample, 46 out of 111 stocks (41%) experience a negative abnormal return after cost on the ex date. The risk involved may deter some traders from doing more arbitrage.

We also note in Table III that returns around ex dates are positive and sometimes significant. For the nontaxable sample, the average abnormal return on day 1 is significant and positive, but for the taxable sample, the average is significant and positive on day -1.

Abnormal returns around the ex date are not necessarily associated with abnormal order imbalances. Table IV reports aggregate order imbalances around ex dates. Despite a significant return, there is no significant abnormal order imbalance from day-1 to day 1 for the nontaxable sample. In contrast, for the taxable sample, there are many more buy orders than sell orders from day -1 to day 1. To understand the source of order imbalances, we examine the order behavior of different investor types.

B. Abnormal Order Imbalances

Our hypotheses make specific predictions about order behaviors from foreign investors, domestic institutions, and large and small individual investors. Table V reports the average standardized abnormal order imbalance from day -2 to day 2. Panels A and B give the average order imbalances from aggressive orders for nontaxable and taxable samples; Panels C and D give the average from conservative orders.

For the nontaxable sample, our only predictions are for small individual investors: they want to sell before the ex date and buy afterwards. This finding reflects the nuisance hypothesis (Hypothesis 1) and the price drop hypothesis (Hypothesis 2). The evidence is consistent with our predictions. For small investors, we find that the imbalances from their aggressive orders on days -2 and -1 are both negative, although neither is significant at a 0.1 level (Column 4 in Panel A of Table V). Moreover, small investors turn into net buyers on the ex date. Their average order imbalance is a positive 0.48 on day 0, which is significant at a 0.05 level.

Corroborative evidence for the nuisance and price drop hypotheses comes from the taxable sample. Small investors sell more and sell aggressively before the ex date (Panel B in Table V). The averages of order imbalances are -0.41 and -0.56 for day -2 and -1, and both are significant at a 0.05 level. Then they become net buyers from the ex date. The averages are all positive from day 0 to day 2, and the average on day 0 is 0.56, which is statistically significant. Our evidence on small investors is related to findings in Jakob and Ma (2003). Jakob and Ma examine a sample of 106 distributions from companies listed on the NYSE. They find that order imbalances from small orders are significantly positive on the ex date. The difference is that they examine cash dividends and we look at stock dividends.

The evidence from small investors does not support the tax hypothesis. If small investors are in low tax brackets, they should buy before and sell after the ex date to capture the dividend tax credit. The evidence shows otherwise. It seems that small investors are very determined not to receive a stock dividend. Compared with aggressive orders, the evidence is much weaker for conservative orders submitted by small investors (Panels C and D in Table V). The only significant result occurs on the ex date for the taxable sample, where we find that, similar to aggressive orders, small investors submit a significantly higher number of conservative buy orders. The aggressiveness of small investors does not appear to be rational. The timing of the ex date and its effect on prices are matters of public knowledge. Rational investors should spread their orders to reduce the price impact rather than concentrate their orders.

Our tax hypothesis (Hypothesis 3) argues that foreign and large investors (if they are in high tax brackets) should sell before the ex date and buy afterwards. Table V presents some weak evidence to support the tax hypothesis. Both foreign and large investors submit significantly more conservative buy orders than sell orders on the ex date of the taxable sample. The average order imbalances are 0.37 for foreign investors and 0.45 for large investors (Columns 1 and 2 in Panel D, Table V). In contrast to aggressive trading by small investors, foreign and large investors do not trade aggressively around the ex date relative to the estimation period.

The lack of significance before the ex date may be a rational choice by these tax-motivated investors. Sophisticated investors know the exact timing of the ex date and its tax implications long before day 0 (the time interval between the ex date and the shareholder meeting that decides the dividend distribution rate ranges from 27 to 209 days, and the median is 59 in our sample). Harris (1998) argues that under such a circumstance, uninformed liquidity traders will submit conservative orders when deadlines are distant. Liquidity traders will also spread orders over the whole period to minimize the price impact. Therefore, it may be difficult to detect changes in behavior for tax-motivated investors.

In Table V, we define small and large investors using a daily order value of NT$200,000 as the cut-off point. NT$200,000 is approximately one half of the per capita gross national product in Taiwan in 1999. It also amounts to 14,000 shares for the nontaxable sample or 7,000 shares for the taxable sample based on the median cum-dividend prices of NT$14.4 or NT$30.4 for two samples. To examine the robustness of our choice, we also use NT$100,000 and NT$300,000 as cut-off points. Table VI reports order imbalance results for large and small individual investors based on aggressive orders and different cut-off points. When the cut-off point is NT$100,000, none of the order imbalances for small and large investors is significant. The results for a NT$300,000 cut-off point are similar to results for the NT$200,000 cut-off point. Therefore, the evidence for small and large investors is robust to the choice of the cut-off point.

Institutional investors behave very differently from other types of investors. For the taxable sample in Table V, institutions buy aggressively before the ex date (the average order imbalance is 0.4 on day -1) and start to sell aggressively from day 0 (averaged -0.34). Most abnormal order imbalances are significant. This result is consistent with Hypothesis 5, that institutions delay their sale or pursue a long arbitrage strategy, because dividend taxes are irrelevant for them and the expected return is positive. Lending credence to the tax story is the sharp contrast between the negative order imbalances in the taxable sample from day 0 to day 2 and the positive order imbalances in the nontaxable sample.

Institutions prefer to submit aggressive orders around the ex date. Although the sign of order imbalances is the same, numbers from conservative orders are not significant except on day 1. To be aggressive is a reasonable strategy if institutions are acting as short-term arbitrageurs. Arbitrage is risky and aggressive orders can reduce the uncertainty of failing to trade.

V. Further Results

We find that the pattern of the average order imbalance for the four groups of investors is consistent with our hypotheses. This section further examines the order imbalance from explicit arbitrage activities as well as the relation between order imbalances, distribution rates, and returns.

A. Order Imbalances from Explicit Arbitrage Activities

Table V shows that institutions buy more before the ex date and sell more afterwards, but small investors do just the opposite. These results can arise because investors choose the timing of their trade. They can also arise because investors take explicit arbitrage opportunities. Institutions do long arbitrage and small individual investors conduct short arbitrage. To provide further evidence to differentiate these two possibilities, we directly estimate the extent of arbitrage activities near the ex date.

To test for explicit arbitrage activities, we calculate the abnormal relative order imbalances from arbitrage activities. To estimate the extent of long arbitrage, we first identify those investors who submit both buy orders on day -1 (or -2) and sell orders on the ex date 0. We then calculate the total order value from buy orders to measure the extent of long arbitrage. To estimate the extent of short arbitrage, we identify investors who submit both sell orders on day -1 (or -2) and buy orders on the ex date, then calculate the total order values from sell orders. The relative order imbalances from arbitrage activities are the differences between long and short arbitrages divided by the sum of total buy and sell orders for that day.

To calculate the standardized abnormal arbitrage activities during the event period, we use the average and standard deviation of normal arbitrage activities during the estimation period. To estimate the normal relative order imbalances from arbitrageurs, we apply the same procedure over the estimation period of days -50 to -6. For each day t within the estimation period, we identify those investors who submit both buy (sell) orders on day t-1 and sell (buy) orders on day t, and use the total values from buy (sell) orders as the normal long (short) arbitrage volume. Table VII reports the results for abnormal arbitrage activities.

The table shows strong evidence of long arbitrage activities around the ex date for both the nontaxable and taxable samples. All types of investors do long arbitrage, and the only difference is their aggressiveness. Institutions and individual investors, both large and small, submit aggressive orders to do long arbitrage; foreign investors submit only conservative orders.

The results on explicit arbitrage from small individual investors are very different from results on total orders shown in Table V. When we look at the total order imbalances from small investors, we see that investors in both the nontaxable and taxable samples sell before the ex date and buy afterwards. When we examine only the order imbalances from explicit arbitrage activities, we see that they buy before the ex date and sell afterwards. These results suggest that individual investors are heterogeneous. Some will do the long arbitrage to capture the dividend tax credit and a positive expected return, as suggested in Hypothesis 3, but most will choose to avoid the dividend. The heterogeneity of small individual investors is unlikely to come from their unobserved marginal tax rates, because we can observe similar behaviors for both nontaxable and taxable distributions.

The results of explicit arbitrage from foreign and large individual investors suggest these investors are also heterogeneous. When we examine the order imbalances from explicit arbitrage activities for both nontaxable and taxable samples, we find that foreign and large investors do the long arbitrage. They buy before the ex date and sell afterwards. This behavior suggests that some of the large individual investors and foreign investors want to capture the positive expected return on the ex date. This behavior is not consistent with our tax Hypothesis 3. Both foreigners and high tax bracket investors should avoid receiving the tax-disadvantaged dividends. Although not all foreign and large investors avoid dividends, enough of them do want to avoid the dividend tax. Therefore, in Table V we observe that for the total order imbalances, foreign and large investors do not buy until the ex date.

B. Order Imbalances and Distribution Rates

In addition to average results, we also examine the effect of distribution rates on order imbalances. Hypothesis 5 predicts that for the taxable sample, order imbalances from institutions are more positive before the ex date and more negative afterwards, as the dividend increases. Table VIII provides evidence broadly consistent with this hypothesis.

Table VIII reports the coefficient on the distribution rate in a regression in which the dependent variable is the abnormal order imbalance. In Table VIII, Column 3, Panel B shows that the result is consistent with Hypothesis 5: as the distribution rate increases, institutions submit significantly more buy orders before the ex date and then submit significantly more sell orders from day 0.

Table VIII also examines the behavior of small investors. We find that they behave exactly opposite to institutions. As the distribution rate increases, small investors' order imbalances are more negative before the ex date and more positive from day 0. This result is more consistent with the price drop hypothesis (Hypothesis 2) than with the nuisance hypothesis (Hypothesis 1). The percentage of the price drop is a linear function of the distribution rate. Therefore, the magnitude of order imbalances should be increasing to the distribution rate.

The evidence in Table VIII supports the price drop hypothesis; investors like low-priced stocks, as Black (1986) suggests.

Similar to the average results in Table V, the supporting evidence for the tax hypothesis is weak. Contrary to Hypothesis 4, we find no significant changes in behavior by foreigners. However, when the distribution rate increases, large investors in the taxable sample submit more conservative buy orders on days 0 and 1, as predicted. In contrast, large investors' aggressive orders on day 0 drop significantly.

C. Order Imbalances and Returns

As noted earlier, on the ex date, the average abnormal return is significant and positive, but the average abnormal aggregate order imbalance is not. Therefore, the aggregate order imbalance cannot explain the return behavior. We want to find out if we can explain returns by using order imbalances from various types of investors.

Table IX reports regression results we obtain by using abnormal returns on the ex date as the dependent variable. We include both taxable and nontaxable samples in the regression. The independent variables are a dummy variable for the tax status and order imbalances from four groups of investors. We find that the only significant variable is the order imbalance of large individual investors. The regression shows that when large individuals buy more, the stock price goes higher. Large individual investors not only submit the largest number of shares (Table II reports that their median percentage is higher than 65%), they are also marginal traders.

Despite a significant relation between the abnormal return and the order imbalance of large investors, we still cannot explain the average abnormal return. Without including any order imbalances variables, the intercept is 0.514 (significant at a 0.05 level). The intercept becomes even larger, 0.58, when we include order imbalances from all the four groups of investors. However, we are not surprised by the result, since Table V shows that the average order imbalance of large individuals is not significantly different from zero.

VI. Conclusion

In this article, we examine order flows around ex-dividend dates on the Taiwan Stock Exchange. Not only does Taiwan's tax code allow us to separate the tax hypothesis from other explanations, but Taiwan's data also allows us to examine the heterogeneity of investor behavior around ex dates.

We find strong evidence that small investors sell before the ex date and start to buy from the ex date, which suggests that small investors prefer low prices. We find weaker evidence consistent with the tax hypothesis. Foreign and large domestic investors who are taxdisadvantaged avoid participating in dividends. We also find strong evidence that institutions play the role of short-term arbitrageurs around ex-dividend dates, buying before the ex date and selling afterwards.

Appendix

If the dividend amount is [D.sup.i], then let the associated equilibrium expected return on the ex date be [R.sup.i]. Given [D.sup.i] and [R.sup.i], investors choose a trading strategy based on their after-tax unit dividend income, [epsilon]k - [tau](l + [epsilon]k). We define the critical after-tax income, [x.sup.i,j], as the income that makes investors indifferent between adopting the strategy j and not adopting. Take the long arbitrage (LA) strategy as an example, investors are indifferent toward this strategy if their after-tax income equals [x.sup.i,LA] such that the following equality (Al) holds.

[R.sup.i] = 2c/(1-c) - 1/(1 - c) [D.sup.i]/[P.sub.-1] [x.sup.i,LA](A1)

Similarly, [X.sup.i,DP] and [x.sup.i,DS] are critical values that make investors indifferent toward the DP (delayed purchase) strategy and the DS (delay sale) strategy as in Equations (A2) and (A3).

[R.sup.i] = 1/(1-c) [D.sup.i]/[P.sub.-1][x.sup.i,DP] (A2)

[R.sup.i] = 1/(1-c) [D.sup.i]/[P.sub.-1][x.sup.i,DP] (A3)

To show that the expected return on the ex date will increase with the dividend, we first assume [D.sup.2] > [D.sup.1]. If the expected return stays at [R.sup.1] when the dividend increases from [D.sup.1] to [D.sup.2], then more investors will pursue the delayed purchase strategy and fewer investors will pursue the delayed sale strategy. Fewer investors will use the long arbitrage strategy if the expected return is higher than 2c/(1-c), which is approximately 0.59% in Taiwan, because the maximum commission rate is 0.1425% and the security transaction tax is 0.3%. Both these costs are levied on the seller. Hence, an order imbalance will be positive on the ex date if the expected return stays at [R.sup.1] and the expected return has to increase to a higher level [R.sup.2].

We want to show how the order imbalance for different investors changes when the dividend increases to [D.sup.2] and the equilibrium expected return becomes [R.sup.2]. To do so, we need to know how high [R.sup.2] can be. Can it be high enough to keep the order imbalance the same from investors following the delay purchase strategy?

To keep the order imbalance the same from investors following the delay purchase strategy, we must keep the critical x the same (that is, [x.sup.2,DP] = [X.sup.1,DP]), Hence, [R.sup.2] must satisfy the following equation (see Figure AI):

[R.sup.2] = [[??].sup.2] 1/(1-c) [D.sup.2]/[P.sub.-1][x.sup.1,DP] (A4)

Since [x.sup.1,DP]/(1 + c) = [X.sup.1,DS]/(1 - c) from (A1) and (A2), [R.sup.2] will also satisfy the following equation, and the order imbalance from investors following the delayed sale strategy will stay the same:

Figure Al. Investor's Trading Strategy

If the dividend amount is [D.sup.i], then let the associated equilibrium expected return on the ex date,

(1 + d) E[[P.sub.0]] - [P.sub.-1]/[P.sub.-1]

be [R.sup.i] Given [D.sup.i] and [R.sup.i], investors choose a trading strategy based on their after-tax unit dividend income, [epsilon]k - [tau](1 + [epsilon]k), where k is the dividend tax credit rate, e is the fraction of dividend tax credit that the investor can receive, and [tau] is the tax rate applicable for dividends. Line LA([D.sup.1]) represents expected return/tax credit pairs that investors are indifferent about long arbitrage that purchases on day -1 and sells at 0 given a dividend payment of [D.sup.1]. Line DS([D.sup.E]) represents pairs that investors are indifferent about delay selling from day -1 until date 0 given a dividend of [D.sup.1]. Line DP([D.sup.1]) represents pairs that investors are indifferent about delay purchasing from day -1 until date 0.

[R.sup.2] = 1/(1-c) [D.sup.2]/[P.sub.-1][x.sup.1,DS] (A5)

When [R.sup.2] satisfies (A4), the order imbalance from investors following the long arbitrage strategy will be larger than the order imbalance at [D.sup.1] and [R.sup.1], because (A1), (A2), and (A4) imply the following:

[R.sup.2] = 2c/(1-c) - 1/(1-c) [D.sup.2]/[P.sub.-1][x.sup.1,LA] (A6)

Therefore, when [R.sup.2] satisfies (A4), order imbalances on the ex date will be negative and thus will drive the equilibrium expected return [R.sup.2] lower. Given that:

[R.sup.2] < [[bar.r.].sup.2] (A7)

the critical value x will be higher ([x.sup.2,DP] > [x.sup.1,DP]), more investors will follow the delay purchase strategy, and the order imbalance on the ex date from these investors will be larger. Fewer investors will follow the delayed sale strategy. We can make no clear-cut prediction on the number of investors who will follow the long arbitrage strategy.

References

Andrews, A.W. and L.R. Wilson, 1971, "Stock Dividend Taxation under the Tax Reform Act of 1969: Expansion of an Ominous Past," Arizona Law Review 13, 751-796.

Baker, C.A., 1958, "Evaluation of Stock Dividends," Harvard Business Review July/August, 99-114.

Belsley, D.A., E. Kuh, and R.E. Welsch, 1980, Regression Diagnostics, New York, NY, John Wiley & Sons, Inc.

Black, F., 1986, "Noise," Journal of Finance 41, 529-543.

Boyd, J.H. and R. Jagannathan, 1994, "Ex-Dividend Price Behavior of Common Stocks," Review of Financial Studies 7, 711-741.

Eades, K.M., P.J. Hess, and E.H. Kim, 1984, "On Interpreting Security Returns During the Ex-Dividend Period," Journal of Financial Economics 13, 3-35.

Elton, E.J. and M.J. Gruber, 1970, "Marginal Stockholder Tax Rates and the Clientele Effect," Review of Economics and Statistics 52, 68-74.

Grinblatt, M., R. Masulis, and S. Titman, 1984, "The Valuation Effects of Stock Splits and Stock Dividends," Journal of Financial Economics 13, 461-490.

Harris, L., 1998, "Optimal Dynamic Order Submission Strategies in Some Stylized Trading Problems," Financial Markets, Institutions & Instruments 7, 1-75.

Jakob, K. and T. Ma, 2003, "Order Imbalance on Ex-Dividend Days," Journal of Financial Research 26, 65-75.

Koski, J.L. and J.T, Scruggs, 1998, "Who Trades Around the Ex-Dividend Day? Evidence from NYSE Audit File Data," Financial Management 27, 58-72.

Kryzanowski, L. and H. Zhang, 1996, "Trading Patterns of Small and Large Traders around Stock Split ExDates," Journal of Financial Research 19, 75-90.

Lakortishok, J. and B. Lev, 1987, "Stock Splits and Stock Dividends: Why, Who, and When," Journal of Finance 42, 913-932.

Lakonishok, J. and T. Vermaelen, 1986, "Tax-Induced Trading Around Ex-Dividend Days," Journal of Financial Economics 16, 287-319.

McDonald, R.L., 2001, "Cross-Border Investing with Tax Arbitrage: The Case of German Dividend Tax Credits," Review of Financial Studies 14, 617-657.

We thank Lemma Senbet and Alex Triantis (the former Editors), the anonymous referee, Yehning Chen, Robin Chou, Ravi Jagannathan, Jie-Haun Lee. Suming Lin, Yun Lin, and seminar participants at the National Taiwan University and National Central University for helpful comments. We also acknowledge the financial support from the National Science Council (92-2416-H-002-048-EF).

Shing-yang Hu and Yun-lan Tseng *

* Shing-yang Hu is an associate professor of finance at the National Taiwan University in Taipei, Taiwan. Yun-lan Tseng is an instructor of accounting at the National Pingtung Institute of Commerce in Pingtung, Taiwan.

Table I. Impact on After-Tax Income by Receiving Stock Dividends This table presents the net tax credit, [epsilon]kD - [tau] (1+[epsilon]k)D, received by different categories of investors. D is the announced amount of net stock dividend in the local currency, k is the dividend tax credit rate, [epsilon] is the fraction of dividend tax credit that the investor can receive, and [tau] is the tax rate applicable for dividends to the investor. Investor Category Parameters Foreign investors [epsilon]=0, [tau]>0 Domestic high tax bracket [epsilon]=1, [tau]>k/(l+k) individuals Domestic corporation [epsilon]=0, [tau]=0 Domestic low tax bracket [epsilon]=1, [tau]<k/(1+k) individuals Net Tax Credit [epsilon]kD-[tau] Changes in Investor Category (D+[epsilon]kD) After-Tax Incomes Foreign investors <0 Decrease Domestic high tax bracket <0 Decrease individuals Domestic corporation = 0 Unchanged Domestic low tax bracket >0 Increase individuals Table II. Descriptive Statistics of the Stock Dividend Sample This table presents descriptive statistics of related items for the ex-date sample during 1999. There are 45 and 66 ex dates in the nontaxable and taxable sample, respectively. We measure the value of market capitalization at the end of 1998. We measure return on equity, return on assets, daily trading value and daily turnover rate over the year 1998, and the percentage of total orders from different categories of investors over the year 1999. Nontaxable Sample Standard Item Mean Median Deviation Distribution rate of dividend (%) 8.06 6.00 5.00 Closing prices cum dividend 18.58 14.40 12.14 Market capitalization 13.18 6.83 18.46 NT$ billion) Return on equity (%) 1.59 1.13 6.63 Return on assets (%) 1.79 1.52 3.22 Daily trading value (NT$ million) 152.29 36.21 393.02 Daily turnover rate (%) 0.73 0.61 0.57 Percentage of orders from 1.06 0.09 1.82 foreign investors Percentage of orders from 64.69 65.60 10.65 large individual investors Percentage of orders from 8.16 5.99 8.70 institutional investors Percentage of orders from 26.09 25.83 9.44 small individual investors Taxable Sample Standard Item Mean Median Deviation Distribution rate of dividend (%) 15.82 21.64 10.00 Closing prices cum dividend 73.27 116.19 30.40 Market capitalization 30.39 65.43 9.56 NT$ billion) Return on equity (%) 14.62 12.47 11.82 Return on assets (%) 9.52 7.81 7.71 Daily trading value (NT$ million) 364.08 714.11 82.54 Daily turnover rate (%) 1.03 0.97 0.72 Percentage of orders from 2.18 5.79 0.26 foreign investors Percentage of orders from 68.73 12.89 72.10 large individual investors Percentage of orders from 10.94 7.02 8.76 institutional investors Percentage of orders from 18.16 11.45 16.26 small individual investors Table III. Average Abnormal Return Around the Ex Date This table reports average standardized daily abnormal returns. To calculate abnormal returns, we use estimates from a market model from day -50 to -6. There are 45 stocks for the nontaxable sample and 66 for the taxable sample. Day Nontaxable Sample Taxable Sample Difference -2 -0.031 0.074 0.105 -1 0.284 0.431 ** 0.147 0 0.514 *** 0.632 *** 0.118 1 0.408 ** 0.108 -0.300 2 0.165 0.124 -0.041 *** Significant at the 0.01 level. ** Significant at the 0.05 level. * Significant at the 0.10 level. Table IV. Average Abnormal Order Imbalance around the Ex Date This table reports the average daily standardized abnormal relative order imbalance. The relative order imbalance is the difference between daily buy and sell values divided by the sum of buy and sell values. To calculate abnormal order imbalances, we use estimates from a market model for order imbalances from day -50 to -6. We categorize orders submitted by investors as either aggressive or conservative. We define buy (sell) orders as aggressive if their limit prices are higher (lower) than the best ask (bid), and as conservative if their prices are lower (higher) than the best bid (ask). There are 45 stocks for the nontaxable sample and 66 for the taxable sample. Panel A. Aggressive Orders Day Nontaxable Sample Taxable Sample Difference -2 -0.050 0.002 0.052 -1 0.214 0.436 *** 0.222 0 0.200 0.070 -0.131 1 0.037 0.261 * 0.223 2 -0.152 0.009 0.161 Panel B. Conservative Orders Day Nontaxable Sample Taxable Sample Difference -2 0.164 0.176 0.012 -1 0.094 0.174 0.080 0 0.054 0.440 *** 0.386 * 1 0.002 0.170 0.168 2 0.323 * 0.114 -0.209 *** Significant at the 0.01 level. ** Significant at the 0.05 level. * Significant at the 0.10 level. Table V. Average Abnormal Order Imbalances Across Investor Categories This table reports the average daily standardized abnormal relative order imbalance. The relative order imbalance is the difference between daily buy and sell values divided by the sum of buy and sell values. To calculate abnormal order imbalances, we use estimates from a market model for order imbalances from day -50 to -6. W e define buy (sell) orders as aggressive if their limit prices are higher (lower) than the best ask (bid), and as conservative if their prices are lower (higher) than the best bid (ask). Large individual investors are individuals whose daily order value is greater than NT$200,000. There are 45 stocks for the nontaxable sample and 66 for the taxable sample. Panel A. Aggressive Orders for the Nontaxable Sample Large Small Foreign individual Institutional individual Day Investors Investors Investors Investors -2 0.173 -0.173 0.067 -0.325 -1 0.090 -0.084 0.059 -0.088 0 0.249 0.081 0.189 0.477 *** 1 -0.074 0.128 -0.027 0.072 2 -0.139 -0.356 ** 0.025 -0.210 Panel B. Aggressive Orders for the Taxable Sample -2 -0.117 -0.223 0.235 -0.405 *** -1 -0.085 0.093 0.400 *** -0.564 *** 0 0.040 0.038 -0.340 * 0.560 *** 1 -0.226 0.173 -0.018 0.129 2 0.001 0.213 -0.244 * 0.077 Panel C. Conservative Orders for the Nontaxable Sample -2 -0.181 0.057 0.217 0.176 -1 0.159 -0.084 0.420 * 0.328 0 -0.153 0.096 0.075 -0.051 1 0.064 -0.015 0.251 -0.112 2 -0.079 0.265 0.253 -0.062 Panel D. Conservative Orders for the Taxable Sample -2 0.133 -0.239 0.400 0.287 -1 0.175 -0.077 0.149 0.227 0 0.371 * 0.446 *** -0.095 0.388 *** 1 0.219 0.206 -0.271 * 0.054 2 0.219 0.010 -0.073 0.157 *** Significant at the 0.01 level. ** Significant at the 0.05 level. * Significant at the 0.10 level. Table VI. Average Abnormal Order Imbalances under Different Definitions of Large and Small Individual Investors This table reports the average daily standardized abnormal relative order imbalance from large and small individual investors using different definitions. The relative order imbalance is the difference between daily buy and sell values divided by the sum of buy and sell values. To calculate abnormal order imbalances, we use estimates from a market model for order imbalances from day -50 to -6. We define buy (sell) orders as aggressive if their limit prices are higher (lower) than the best ask (bid). Large (small) individual investors are individuals whose daily order value is greater than (not greater than) NT$100,000, $200,000, or $300,000. There are 45 stocks for the nontaxable sample and 66 for the taxable sample. Panel A. Nontaxable Sample Large Individual Investors Day Daily Order Value Greater Than $100,000 $200,000 $300,000 -2 -0.281 -0.173 -0.156 -1 -0.120 -0.084 -0.165 0 0.100 0.081 0.142 1 0.074 0.128 0.122 2 -0.256 -0.356 ** -0.330 * Small Individual Investors Day Daily Order Value Greater Than $100,000 $200,000 $300,000 -2 -0.115 -0.325 -0.257 -1 0.191 -0.088 -0.050 0 -0.023 0.477 *** 0.378 ** 1 -0.030 0.072 0.150 2 -0.214 -0.210 -0.098 Panel B. Taxable Sample Large Individual Investors Day Daily Order Value Greater Than $100,000 $200,000 $300,000 -2 -0.213 -0.223 -0.180 -1 0.041 0.093 0.201 0 0.059 0.038 0.033 1 0.167 0.173 0.228 * 2 0.086 0.213 0.297 ** Small Individual Investors Day Daily Order Value Greater Than $100,000 $200,000 $300,000 -2 -0.158 -0.405 *** -0.327 ** -1 -0.079 -0.564 *** -0.560 ** 0 0.255 0.560 *** 0.547 *** 1 -0.096 0.129 0.039 2 -0.017 0.077 -0.047 *** Significant at the 0.01 level. ** Significant at the 0.05 level. * Significant at the 0.10 level. Table VII. Average Abnormal Order Imbalances from Explicit Arbitrage Activities This table reports the average daily standardized abnormal relative order imbalance from explicit arbitrage activities. The relative order imbalances from arbitrage activities are the differences between long and short arbitrages divided by the sum of total buy and sell orders on that day. To estimate the extent of long (short) arbitrage on day -2 (or -1), we first identify investors who submit both buy (sell) orders on that day and sell (buy) orders on the ex date 0. Then, to measure the extent of long (short) arbitrage on day -2 (or -1), we calculate the total order value from buy (sell) orders on that day. We define buy (sell) orders as aggressive if their limit prices are higher (lower) than the best ask (bid), and as conservative if their prices are lower (higher) than the best bid (ask). To calculate abnormal arbitrage activity, we use estimates from a market model for arbitrage activities from day -50 to -6. Large individual investors are individuals whose daily order value is greater than NT$200,000. There are 45 stocks for the nontaxable sample and 66 for the taxable sample. Small Foreign Large Individual Institutional Individual Day Investors Investors Investors Investors Aggressive orders from the nontaxable sample: -2 0.320 0.241 0.243 -0.240 -1 0.013 0.649 *** 0.719 *** 0.258 Aggressive orders the taxable sample: -2 0.015 0.188 0.088 0.138 -1 0.237 0.928 *** 0.337 ** 0.319 ** Conservative orders from the nontaxable sample: -2 0.041 0.250 0.872 *** -0.004 -1 1.091 *** 0.494 ** 0.828 *** 0.324 Conservative orders from the taxable sample: -2 -1.402 ** -0.172 1.187 *** 0.040 -1 0.548 ** 0.127 0.147 0.144 *** Significant at the 0.01 level. ** Significant at the 0.05 level. * Significant at the 0.10 level. Table VIII. Regressions Explaining Abnormal Order Imbalances Based on the Distribution Rate This table reports the OLS estimates of the coefficient on the log of dividend distribution rate in a simple regression in which the dependent variable is the daily standardized abnormal relative order imbalance. The daily relative order imbalance is the difference between daily buy and sell values divided by the sum of buy and sell values. To calculate abnormal order imbalances, we use estimates from a market model for order imbalances from day -50 to -6. We define buy (sell) orders as aggressive if their limit prices are higher (lower) than the best ask (bid), and as conservative if their prices are lower (higher) than the best bid (ask). Large individual investors are individuals whose daily order value is greater than NT$200,000. There are 45 stocks for the nontaxable sample and 66 for the taxable sample. Panel A. Aggressive Orders for the Nontaxable Sample Large Individual Day Foreign Investors Investors -2 0.488 0.138 -1 0.220 0.268 0 -0.466 0.265 1 0.491 0.535 * 2 -0.150 0.022 Panel B. Aggressive Orders for the Taxable Sample -2 -0.065 -0.275 -1 -0.081 -0.084 0 0.062 -0.993 *** 1 0.126 0.280 * 2 0.099 0.101 Panel C. Conservative Orders for the Nontaxable Sample -2 -0.093 -0.136 -1 -0.470 0.236 0 -0.148 -0.086 1 0.012 0.104 2 -0.106 -0.211 Panel D. Conservative Orders for the Taxable Sample -2 -0.135 0.089 -1 -0.006 -0.258 0 0.196 0.578 *** 1 -0.185 0.334 * 2 0.087 0.070 Panel A. Aggressive Orders for the Nontaxable Sample Institutional Small Individual Day Investors Investors -2 0.061 0.098 -1 0.162 -0.337 0 -0.283 0.493 * 1 -0.108 0.231 2 -0.147 0.432 Panel B. Aggressive Orders for the Taxable Sample -2 0.187 -0.396 ** -1 0.350 * -0.700 *** 0 -0.430 * 0.454 * 1 -0.450 *** 0.319 ** 2 -0.109 0.132 Panel C. Conservative Orders for the Nontaxable Sample -2 0.175 -0.200 -1 0.599 * -0.394 0 -0.225 0.221 1 0.010 0.243 2 0.064 -0.027 Panel D. Conservative Orders for the Taxable Sample -2 0.163 -0.166 -1 0.354 * 0.010 0 0.029 0.473 *** 1 -0.201 0.488 *** 2 -0.209 0.020 *** Significant at the 0.01 level. ** Significant at the 0.05 level. * Significant at the 0.10 level. Table IX. Regressions Explaining Abnormal Returns Based on Abnormal Order Imbalances This table reports the OLS estimates of the coefficient on the standardized abnormal relative order imbalance on the ex date in a simple regression in which the dependent variable is the abnormal return on the ex date. The daily relative order imbalance is the difference between daily aggressive buy and aggressive sell values divided by the sum of buy and sell values. To calculate abnormal order imbalances (returns), we use estimates from a market model for order imbalances (returns) from day -50 to -6. We define buy (sell) orders as aggressive if their limit prices are higher (lower) than the best ask (bid), and as conservative if their prices are lower (higher) than the best bid (ask). Large individual investors are individuals whose daily order value is greater than NT$200,000. There are 111 stocks in the sample, 45 stocks is nontaxable and 66 is taxable. Order Order Imbalance Imbalance from Large Taxable From Foreign Individual Intercept Dummy Investors Investors (1) 0.514 ** 0.118 (2) 0.522 ** 0.111 -0.058 (3) 0.503 ** 0.124 0.143 * (4) 0.526 ** 0.085 (5) 0.545 ** 0.123 (6) 0.580 ** 0.072 -0.090 0.171 ** Order Order Imbalance Imbalance from from Small Institutional Individual Investors Investors (1) (2) (3) (4) -0.064 (5) -0.064 (6) -0.099 -0.101 ** Significant at the 0.05 level. * Significant at the 0.10 level.

Printer friendly Cite/link Email Feedback | |

Author: | Hu, Shing-yang; Tseng, Yun-Ian |
---|---|

Publication: | Financial Management |

Date: | Dec 22, 2006 |

Words: | 12554 |

Previous Article: | Does dividend policy relate to cross-sectional variation in information asymmetry? Evidence from returns to insider trades. |

Next Article: | Dividend policy inside the multinational firm. |

Topics: |