Sixteenths: Direct Evidence on Institutional Execution Costs

Abstract

In June 1997, the New York Stock Exchange (NYSE) lowered its minimum price increment on most stocks from eighths to sixteenths. Like other researchers investigating similar events, we find that quoted spreads and effective spreads decline with the introduction of sixteenths. However, spreads do not necessarily measure the cost of trading, particularly for market participants who execute larger orders over time. In this paper, we use a sample of institutional trades provided by the Plexus Group to directly measure the effect of the tick size reduction on execution costs. For these institutions, average execution costs actually increase post-sixteenths. More importantly, we find strong evidence that the costs of trading increase substantially for traders that demand liquidity. Specifically, while the cost of executing small orders (less than 1,000 shares) declines, the cost of executing large orders (greater than 100,000 shares) increases and the average execution cost for momentum traders increases. We also find that the cost of executing orders that are not worked by the trading desk increases, and there is an increase in the proportion of orders that are worked. These findings emphasize that spreads are not a sufficient statistic for measuring market quality; they also suggest that smaller price increments reduce market liquidity.

JEL Classification: G10, G19, G20, G23