Market Shock and Trading Efficiency
A Comparison of Electronic and Nonelectronic Markets
About This Event

The central question in securities-market structure is whether investors are better served by human-mediated markets, such as the specialist system of the New York Stock Exchange, or by electronic markets, such as NASDAQ and the Electronic Communication Networks. The Securities and Exchange Commission's recent Regulation NMS, although it purports to be a reform measure, fails to address this basic issue. However, close analysis of how the different markets function under varying circumstances can shed light on the optimal structure for the securities markets in the United States. At this conference, as part of AEI's ongoing review of securities market structure, we will consider an important study of these markets by Kenneth M. Lehn and his two colleagues at the University of Pittsburgh.


12:45 p.m.



Introduction: Peter J. Wallison, AEI


Presenters: Kenneth M. Lehn, University of Pittsburgh

Sukesh Patro, University of Pittsburgh

Kuldeep Shastri, University of Pittsburgh


Paul Bennett, New York Stock Exchange

Lawrence Harris, Securities and Exchange Commission

Frank Hatheway, Nasdaq

Mike Plunkett, Instinet

Benn Steil, Council on Foreign Relations


Peter J. Wallison, AEI



Event Summary

June 2004

Market Shock and Trading Efficiency: A Comparison of Electronic and Nonelectronic Markets

On June 10, 2004, AEI continued its conference series on securities market structure.   Kenneth Lehn and his colleagues from the University of Pittsburgh presented a study to address the central question in securities market structure of whether investors are better served by human-mediated markets, such as the specialist system of the NYSE, or by electronic markets, such as NASDAQ. 

Kenneth M. Lehn
University of Pittsburgh

The SEC is grappling with the issue of whether investors are better served by competition among market centers that may fragment order flow and reduce liquidity, or by consolidation of order flow in a centralized market that may enhance liquidity but reduce competition among market centers. This paper compares the performance of the New York Stock Exchange (NYSE) and NASDAQ on two days in 2003 when securities markets received important new information that resulted in rapid stock price movements and heavy trading volume.  We refer to the periods surrounding the release of this information as periods of "stress."  The results document that liquidity is impaired for both NYSE and NASDAQ securities during the periods of stress.  However, we find that bid-ask spreads on NYSE stocks increase by significantly more than they do for NASDAQ stocks during stress periods, after controlling for various attributes of NYSE and NASDAQ stocks that are associated with spreads.  This difference is most significant for stocks with large market capitalizations.  We also find that within NASDAQ, spreads increase by more for traditional NASDAQ market-makers than they do for ECNs, suggesting that ECNs account for the relatively strong performance of NASDAQ during recent periods of market stress. 

The analysis in the paper suggests three avenues for future research: First, we plan to expand the sample of events from the two days in 2003 to a larger number of information shocks.  Since information shocks of the magnitude examined in this paper are relatively rare, expansion of the sample will involve collection of data for years prior to 2003.  This will be interesting in its own right, since it will allow us to examine whether the relative performance of the NYSE and NASDAQ during periods of stress has changed over time, and whether the change coincides with the growth of ECNs.  Second, in addition to examining transaction efficiency, as we have done in this paper, we plan to examine price discovery in the different marketplaces during periods of stress.  Finally, we examine how spreads on NYSE and NASDAQ securities changed when important new information was released during the trading day.  We also plan to examine the efficiency with which NYSE and NASDAQ securities open when important new information is released overnight.         

Lawrence Harris

These types of studies are extremely important in evaluating how market structure affects performance.  That said, I am not in full agreement with some of their conclusions.  The primary concern I have deals with quoted spreads as opposed to other measures of performance.  In particular, the NASDAQ data did not use the inside spread.  This small detail is important because we are trying to compare the same things for NASDAQ and the NYSE.  The method by which the authors calculated the spreads, in fact, does not use the same numbers.  I do not have enough confidence in the results, but I do credit the authors for their work.

Paul Bennett

The paper addresses three issues:  comparing execution costs, characteristics of market performance under stress, and super montage trading and other ECNs.  We reviewed our effective spreads and recently determined that we offer lower spreads and more cost-effective ones.  My understanding is that in the January time frame, the spreads were higher on the NYSE, but looking at the May data during that stress period, the difference is negligible.  The two markets show--with some exception--that while the NYSE spread starts off smaller than in the period of stress, NASDAQ widens, and the NYSE remains still smaller.  In general, the NYSE is performing quite well during the stress period. 

There are a few issues the authors need to address.  First, this paper does not address the problem that under the dynamics of a non-dealer, no specialist market can be bad.  It also does not address the pure liquidity stress situation where information does not arrive.  A related paper found that the NYSE dealt with this situation better.  In terms of market structure, there are changes in the elements that can explain some of the results cited in this paper. For example, auto quoting can improve the quality of the market on the NYSE.  This is also a positive aspect of an ECN.  The other point is that those quotations are increasingly auto-executed.  This helps to explain the difference between the January and May results in the paper. 

Frank Hatheway

We are talking about which markets do a better job trading stocks.  Neither the NYSE nor NASDAQ buys or sells a single stock in its name.  We have structures we set up for our participants to do this.  The NYSE uses the specialists, and rules are in place.  NASDAQ has market-makers, but we also have additional participants-so we are somewhat of a hybrid.  Other ECNs have a market-making desk, a program desk, and arbitrage desks.  Business models determine the venue in which to trade, whether it be on the NYSE or NASDAQ.  The authors have tried to separate the two types of participants NASDAQ has:  the ECN and non-ECN participants.  Averages are not representative of the bests of what these two groups offer.  I like the way the study is designed.  The analysis highlights the differences between markets, as well as the types of stocks offered in the markets.  This difference shows up in the volatility tests and can explain why NASDAQ's volatility is higher, since its stocks tend to be younger.  Participation differs from each market. 

Mike Plunkett

Institutions tend to be passive and prefer others to be more aggressive.  We decided to bring in the market-makers to fill that gap.  The market-makers created instant flow and used Instinet to reduce their risk.  In 1987 markets were manually driven, and this made Instinet more appealing.  Spreads are becoming more difficult to measure.  We initially resisted opening our markets, but we now realize that this was the best environment.  Spreads that we see now are not actually the effective spreads.  The issue we are facing now deals less with the value of the spreads, but where can you execute the trade most efficiently.  If I see a bid and an offer, can I get it-regardless of where it is offered?  I do think we are trying to make a valiant effort to figure out how things work.

Benn Steil
Council on Foreign Relations

The NASDAQ market structure is complicated.  There used to be a clearer distinction between a market-maker and an ECN.  A traditional market-maker in NASDAQ used private liquidity to price what they were willing to sell.  Before 1997, stock was purchased from the market-maker.  Then in 1997, the distinction was blurred due to rule changes like the order handling rules proposed by the SEC.  A market-maker and an ECN both began to provide public liquidity under the NASDAQ system.  Within the SEC and Congress there has been a belief that ECNs do have a role in the market structure.  However, there is something important in the NYSE's specialist system, especially during times of stress.  There should be some protection for the specialist system so that it remains at the NYSE. 

AEI research assistant Jessica Browning prepared this summary.

View complete summary.
Also Visit
AEIdeas Blog The American Magazine

What's new on AEI

AEI Election Watch 2014: What will happen and why it matters
image A nation divided by marriage
image Teaching reform
image Socialist party pushing $20 minimum wage defends $13-an-hour job listing
AEI Participants


Peter J.
AEI on Facebook