We study the impact of Toronto Stock Exchange (TSE) decimalization on the competition for order f... more We study the impact of Toronto Stock Exchange (TSE) decimalization on the competition for order flow. For TSE stocks cross-listed on the NYSE/AMEX, spreads decrease by 27% on the TSE and do not change on the NYSE/AMEX. For TSE stocks cross-listed on Nasdaq, spreads decline by 16% and 8% on the TSE and Nasdaq, respectively. However, order flow does not migrate from U.S. markets to the TSE. Our results indicate that the savings in TSE transaction costs do not offset the benefits of trading on the NYSE/AMEX, and that Nasdaq dealers might not operate as efficiently as perfect competition warrants.
This article tests for differences in execution costs among specialist firms for New York Stock E... more This article tests for differences in execution costs among specialist firms for New York Stock Exchange listed securities. Execution cost differences provide a measure of the relative performance of specialist firms. We find a substantial difference in effective spreads and order processing costs across specialist firms, controlling for stock characteristics. While economically significant, the differences in execution costs between specialist firms are much smaller than the cross-market differences reported by . Within a specialist firm, there is a positive relation between order processing costs and trading activity that is consistent with the hypothesis that active stocks subsidize inactive stocks. York Stock Exchange (NYSE) specialist firms and among stocks traded by the same specialist firm. Execution costs are paid by investors when completing a trade, and are separate from the commission costs paid t o a broker. Execution costs arise from the bid-ask spread and the price impact of an investor's order on the future bid-ask quote. shows that a transactionally efficient financial market which minimizes execution costs also minimizes the deviation in transaction prices from the true price of the underlying security. Evidence of differential execution costs for similar securities can be used to identify high cost and low cost liquidity providers on the NYSE. Furthermore, differential execution costs among stocks traded by the same specialist firm provide evidence consistent with a subsidy from actively traded to inactively traded stocks.
Using reported large short position data from Korea stock market, we studied stock price movement... more Using reported large short position data from Korea stock market, we studied stock price movements around the short covering trades. We found a significant negative relationship between the short covering trades and the stock return on the previous five trading days. trading days. It means that short sellers engage in contrarian trading when they cover short positions. Short coverings are associated with positive returns on the trading day, confirming that there is a positive transient price impact at the coverings of large short positions. We also found that stock prices decline steadily in the 60 trading days following short coverings. This indicates that some short sellers are forced to prematurely close out their positions, thus suffering opportunity losses. Particularly, when a sudden surge in stock prices causes involuntary large-scale short coverings to occur in the short squeeze, the price impact gets larger and short sellers bear greater opportunity losses. This study sheds...
This study analyzes whether a value averaging (VA) strategy, which adjusts the amount of investme... more This study analyzes whether a value averaging (VA) strategy, which adjusts the amount of investment each period to achieve the target amount of investment in risk assets, as a modified form of a dollar cost averaging (DCA) strategy, improves investment performance. Using 18.5 years of fund market data in Korea from 2001 to June 2019, we compare the investment performance of VA strategy relative to two alternatives: DCA strategy, which invests a certain amount in each period, and Buy-and-Hold (BH) strategy, which refers to half-and-half asset allocation between risky and risk-free assets and has an expected return which is the same as that of DCA in the ex-ante sense. Our historical performance analysis reveals that the VA strategy has lower average return and higher standard deviation compared to the BH strategy and has lower average return and lower standard deviation compared to the DCA strategy. These findings are in stark contrast to the claims made by advocates of VA strategy t...
This paper reports that prices of NASDAQ stocks are more volatile around the market opening than ... more This paper reports that prices of NASDAQ stocks are more volatile around the market opening than closing. Evidence fiom individual stocks indicates that there is a systematic relationship between the excess opening volatility and trading activity. Much of the excess opening volatility is related to two factors, bid-ask bounce and price formation. For inactively
This paper conducts a n empirical test of a market microstructure model using a new econometric a... more This paper conducts a n empirical test of a market microstructure model using a new econometric approach. I treat the direction of a trade a s a discrete latent variable following a stationary Markov chain. By overlaying a t h r e e-s t a t e Markov c h a i n o n a familiar m a r k e t microstructure model, I can extract information on the directions of trades efficiently from time-series data. An analysis of 100 large and 100 small firms for the year 1990 yields several important results: (1) Order types (sale, cross, purchase) are serially correlated, and the mean transition probability matrix is very similar for large and small firms. (2) Information asymmetry is greater for smaller firms. (3) The per share order processing cost is greater for larger firms. (4) When trades are classified by the bid-ask test supplemented by the tick test, t h e estimated misclassification probabilities are typically small for sales and purchases, but they are often fairly large for crosses. (5) Buy-sell classification error r e s u l t s i n systematic b i a s e s for regression coefficients.
This paper examines two competing views about the impact of program trading on stock returns. The... more This paper examines two competing views about the impact of program trading on stock returns. The first view is that program trading destabilizes the stock market by making market participants overreact to the program trading. This view, which is popular among practitioners and regulators, implies that program trading increases market volatility. The proponents of this view claim that the market
We study the impact of Toronto Stock Exchange (TSE) decimalization on the competition for order f... more We study the impact of Toronto Stock Exchange (TSE) decimalization on the competition for order flow. For TSE stocks cross-listed on the NYSE/AMEX, spreads decrease by 27% on the TSE and do not change on the NYSE/AMEX. For TSE stocks cross-listed on Nasdaq, spreads decline by 16% and 8% on the TSE and Nasdaq, respectively. However, order flow does not migrate from U.S. markets to the TSE. Our results indicate that the savings in TSE transaction costs do not offset the benefits of trading on the NYSE/AMEX, and that Nasdaq dealers might not operate as efficiently as perfect competition warrants.
This article tests for differences in execution costs among specialist firms for New York Stock E... more This article tests for differences in execution costs among specialist firms for New York Stock Exchange listed securities. Execution cost differences provide a measure of the relative performance of specialist firms. We find a substantial difference in effective spreads and order processing costs across specialist firms, controlling for stock characteristics. While economically significant, the differences in execution costs between specialist firms are much smaller than the cross-market differences reported by . Within a specialist firm, there is a positive relation between order processing costs and trading activity that is consistent with the hypothesis that active stocks subsidize inactive stocks. York Stock Exchange (NYSE) specialist firms and among stocks traded by the same specialist firm. Execution costs are paid by investors when completing a trade, and are separate from the commission costs paid t o a broker. Execution costs arise from the bid-ask spread and the price impact of an investor's order on the future bid-ask quote. shows that a transactionally efficient financial market which minimizes execution costs also minimizes the deviation in transaction prices from the true price of the underlying security. Evidence of differential execution costs for similar securities can be used to identify high cost and low cost liquidity providers on the NYSE. Furthermore, differential execution costs among stocks traded by the same specialist firm provide evidence consistent with a subsidy from actively traded to inactively traded stocks.
Using reported large short position data from Korea stock market, we studied stock price movement... more Using reported large short position data from Korea stock market, we studied stock price movements around the short covering trades. We found a significant negative relationship between the short covering trades and the stock return on the previous five trading days. trading days. It means that short sellers engage in contrarian trading when they cover short positions. Short coverings are associated with positive returns on the trading day, confirming that there is a positive transient price impact at the coverings of large short positions. We also found that stock prices decline steadily in the 60 trading days following short coverings. This indicates that some short sellers are forced to prematurely close out their positions, thus suffering opportunity losses. Particularly, when a sudden surge in stock prices causes involuntary large-scale short coverings to occur in the short squeeze, the price impact gets larger and short sellers bear greater opportunity losses. This study sheds...
This study analyzes whether a value averaging (VA) strategy, which adjusts the amount of investme... more This study analyzes whether a value averaging (VA) strategy, which adjusts the amount of investment each period to achieve the target amount of investment in risk assets, as a modified form of a dollar cost averaging (DCA) strategy, improves investment performance. Using 18.5 years of fund market data in Korea from 2001 to June 2019, we compare the investment performance of VA strategy relative to two alternatives: DCA strategy, which invests a certain amount in each period, and Buy-and-Hold (BH) strategy, which refers to half-and-half asset allocation between risky and risk-free assets and has an expected return which is the same as that of DCA in the ex-ante sense. Our historical performance analysis reveals that the VA strategy has lower average return and higher standard deviation compared to the BH strategy and has lower average return and lower standard deviation compared to the DCA strategy. These findings are in stark contrast to the claims made by advocates of VA strategy t...
This paper reports that prices of NASDAQ stocks are more volatile around the market opening than ... more This paper reports that prices of NASDAQ stocks are more volatile around the market opening than closing. Evidence fiom individual stocks indicates that there is a systematic relationship between the excess opening volatility and trading activity. Much of the excess opening volatility is related to two factors, bid-ask bounce and price formation. For inactively
This paper conducts a n empirical test of a market microstructure model using a new econometric a... more This paper conducts a n empirical test of a market microstructure model using a new econometric approach. I treat the direction of a trade a s a discrete latent variable following a stationary Markov chain. By overlaying a t h r e e-s t a t e Markov c h a i n o n a familiar m a r k e t microstructure model, I can extract information on the directions of trades efficiently from time-series data. An analysis of 100 large and 100 small firms for the year 1990 yields several important results: (1) Order types (sale, cross, purchase) are serially correlated, and the mean transition probability matrix is very similar for large and small firms. (2) Information asymmetry is greater for smaller firms. (3) The per share order processing cost is greater for larger firms. (4) When trades are classified by the bid-ask test supplemented by the tick test, t h e estimated misclassification probabilities are typically small for sales and purchases, but they are often fairly large for crosses. (5) Buy-sell classification error r e s u l t s i n systematic b i a s e s for regression coefficients.
This paper examines two competing views about the impact of program trading on stock returns. The... more This paper examines two competing views about the impact of program trading on stock returns. The first view is that program trading destabilizes the stock market by making market participants overreact to the program trading. This view, which is popular among practitioners and regulators, implies that program trading increases market volatility. The proponents of this view claim that the market
Uploads
Papers by Hyuk Choe