The reduction of the minimum tick size to one penny may be the most controversial rule change in recent years. It is causing the most far-reaching overhaul in institutional trading practices since the introduction of the order handling rules.
These latest changes have had a profound effect on both the Nasdaq and New York Stock Exchange. On the latter, charges of penny jumping have been frequent. But how does penny trading actually improve the quality of the U.S. equity markets?
For the buyside, the biggest complaint is penny jumping on listed stocks. That's quite a change because five years ago the largest number of complaints came from Nasdaq traders. This may be because of the nature of Big Board trading. NYSE specialist firms are driven by profits. They buy and sell stocks that they make markets for. That means they often take long or short positions partly based on their unique knowledge of customer trades. In essence, they are trading against their customers.
The only way for the buyside to defend against penny jumping is to jump first. It can post limit orders, jumping to the head of the line while using ECNs to preserve anonymity in both listed and Nasdaq stocks. Many buyside firms are doing this.
To study the impact of the penny tick size I have examined the top twenty NYSE and Nasdaq stocks as measured by dollar trading volume for the months of August 1996, August 1998 and August 2001. That's when the tick sizes were eighths, teenies and pennies, respectively.
The data for both market centers were provided by the Securities Industry Automation Corporation, or SIAC, which means reporting differences do not allow intermarket comparisons. Quote frequency for the market centers for the top 20 stocks in these months are shown on Table 1. Nasdaq quote frequency in this sample, as shown in the table, grew at 17 times the rate for the NYSE. This suggests that the Nasdaq market has undergone radical structural changes – the order handling rules, the explosive growth of online trading and the growth of ECNs included – changes which have had a decided effect on quoted spreads (See Figure 1).
The distinctive tick size changes are obvious. In the 1/8 regime, about 40% to 60% of Nasdaq and NYSE quotes were at the 1/8. In the teenie category, those percentages were respectively 74% and 98%.
Dealer Market
The tightening of quotes has been far greater in the dealer market with tick size reductions. Yet the most striking revelation of the graph is that both eighth and teenie tick sizes were huge barriers to price competition. Only with a penny tick increase can one see a continuous relationship between spread width and the cumulative percentage of spreads. For example, on Figure 1, note the cumulative distribution for spreads on the Nasdaq market in August 1996 with the eighth tick size increase from 41% to 89% to 95% as spreads increased from 1/8 to 2/8 to 3/8, respectively.
In August 1998, as Nasdaq spreads increased from 1/16 to 2/16 to 3/16 to 4/16, the cumulative percentage increased from 59% to 84% to 91% to 96%, respectively. In each case, the jumps in the cumulative percentage are quite large, particularly for initial jumps.
By contrast, in August 2001, as spreads increase from one cent to two cents to three cents to four cents, the cumulative percentage rises from 51% to 68% to 78% to 84%, respectively. The incremental changes in the penny world are much smaller. This indicates that limit order submission preferences are more readily met in the penny tick incremental world than they were in the teenie or eighth environment. (Fifty one percent of spreads pressed down against one-penny intimates that even the penny tick size could be viewed as a problem.) In other words, a significant barrier to price competition has fallen.
The reduction in effective spreads is greater for the NYSE than the Nasdaq. But I also conclude that – while spreads are wider on the Big Board – the percentage of trading inside the spread is greater than on the Nasdaq. Still, once again the impact of tick size reductions is greater in the dealer market. So the removal of the barrier to price competition is again demonstrated. An economist generally supports the destruction of these barriers. Deregulation resulted in a big cut in long distance telephone rates and airline fares. So why is there such a fuss about pennies?
Perhaps it is because of volatility with smaller tick size, which tends to help the sellside and hurt the buyside. To study volatility, I formed portfolios of the top 20 stocks in each market. I created a minute-by-minute return index for each portfolio and figured their volatilities in each period. My view was that of a buyside manager who invested in two portfolios consisting of the top 20 NYSE and Nasdaq stocks. As Table 2 shows, volatility had increased somewhat for Nasdaq, but not appreciably for the NYSE. The differences in neither NYSE nor Nasdaq volatilities are not statistically significant.
What about liquidity?
Table 3 shows increased size for our Nasdaq sample quoted size, calculated from the NBBO. However, the NYSE sample decreased. I would expect that – given the shape of supply/demand curves – quoted volume as reflected in posted limit orders will decrease with tick size reduction. The closer price is to equilibrium the smaller the quantity offered for purchase or sale. The equilibrium price is, in theory, the spread midpoint.
Intuitively, the better price you give me, the more I am willing to buy/sell, where "better" is measured by distance from the equilibrium price. From another viewpoint, considering the fact that posting a limit order entails writing an option, and that the option premium is reduced as the spread decreases, smaller tick size should result in smaller quote size. Nevertheless, consider the trading volume for our sample shown in Table 4.
Share volume traded has increased 350% for the NYSE stocks and 600% for Nasdaq stocks from August 1996 to August 2001. Clearly, liquidity, as measured by share volume traded, has soared coincident with tick size reduction. I am not suggesting that the volume increases were caused solely by the tick size reduction. Yet trading volume increases are hardly consistent with the commonly heard complaint of a reduction in liquidity.
An additional piece of the puzzle is that in Europe and elsewhere small ticks have existed without complaints. Although the U.S. equity markets dominate others, one assumes that the more liquid a market, the more robust it would be able to handle irritations like decimals. That's if, in fact, decimal trading is an irritant.
Market Efficiency
It is understandable that the sellside would dislike the removal of the barrier to price competition as a result of penny ticks. After all, the sellside is the loser. No one is willing to throw himself on his sword in the interest of market efficiency. Still, why is the buyside yapping? They are the keepers of our wealth. They have a fiduciary obligation to pension and mutual funds. As our agents why wouldn't they welcome the fall of price barriers, which will, of course, improve our returns?
In retrospect, the 1/8 tick size seems big. The fat cat days are over for the suppliers of liquidity. And from all the reports, decimalization has increased the workload for the buyside that is required to execute buy and sell programs. Clearly, there has been a signal change in trading listed equities. And the change is a challenge to all trading participants.
Robert A. Wood is Distinguished Professor of Finance, University of Memphis.
