INTER-MARKET COMPETITION FOR EXCHANGE TRADED FUNDS

Journal of Economics and Finance, Summer 2007 by Nguyen, Vanthuan, Van Ness, Bonnie F, Van Ness, Robert R

Abstract

We examine how the different mix of informed and liquidity trading in the market for ETFs affects the nature of inter-market competition. We find that both the characteristics of the securities and the structures of the competing markets jointly determine the nature of inter-market competition. Given the superior execution quality on the ECNs and the low adverse selection costs in the ETF market, anonymous market such as the ECNs, attract both liquidity and informed traders. We also find that markets compete in a subset of ETFs. In addition, we find that quote-based competition is prevalent in the market for ETFs.

(JEL G 14, G 18)

Introduction

Exchange traded funds (ETFs) are some of the most actively traded securities in the financial markets. The purpose of this study is to examine competition for order flow among the various market centers that trade these securities. While there are many studies that look at competition between exchanges for NYSE-listed securities (Blume and Goldstein, 1997, and Bessembinder, 2003) and NASDAQ-listed securities (Goldstein et al. 2007), few studies examine the inter-market competition for ETFs. There is a consensus that market structure determines the nature of intermarket competition. Barclay et al. (2003) find that NASDAQ market makers preference or internalize less informed trades. Easley et al. (1996) and Bessembinder and Kaufman (1997b) find evidence that the regional exchanges skim less-informed order flow from the NYSE. A question naturally arises as to whether the results of studies on NYSE- and NASDAQ-listed securities can be generalized to ETFs.

Boehmer and Boehmer (2003) examine the change in liquidity of ETFs after the entry of the NYSE. They document a substantial increase in liquidity after the NYSE entry. Hendershott and Jones (2005) examine the impact of Island stopped display its limit order book in three actively traded ETFs. They find that overall market quality decreases. We add to the above studies by investigating the factors that determine the success of the ECNs in attracting ETF order flow. Studies on NASDAQ securities suggest that traders in more liquid securities favor ECNs while traders in less liquid securities rely on the market makers. Since ETFs are some of the most actively traded securities, ECNs should be dominant trading venues. While it may be true that liquidity plays an important role in the success of the ECNs in ETF markets, there is more to it. Many stocks on the NYSE are very liquid. If liquidity is the main determinant of competitive strategies, how can we explain the low market share of ECNs in NYSE listed securities? Blume and Goldstein (1997) and Bessembinder (2003) show that for NYSE listed stocks the primary market dominates trading while Goldstein et al. (2007) show that for NASDAQ listed stocks the primary market is not as dominant. As ETFs are diversified securities and therefore different from common stock, an analysis of the competition for order flow of ETFs should be of interest.

In this paper, we examine whether the different characteristic of ETFs with regard to adverse selection costs is a factor affecting tbe nature of inter-market competition. The ETFs track an index that constitutes a basket of securities. Subrahmanyam (1991) and Gorton and Pennacchi (1993) suggest that basket securities have lower adverse selection costs and low cost diversification possibilities. Hegde and McDermott (2004) provide empirical support for the view that adverse selection costs are lower for baskets of securities than for underlying stocks. Since basket securities provide diversification and have lower adverse selection costs for portfolio trading, we expect to find a higher proportion of liquidity traders in the market for ETFs. This different mix of informed trading and liquidity-motivated trading may cause the nature of competition among trading venues to differ between the market for ETFs and the market for underlying securities.

Barclay et al. (2003) suggest that electronic communication networks (ECNs) have a higher proportion of informed trading due to a high degree of trader anonymity. Since investors who trade ETFs are primarily liquidity traders (Hegde and McDermott, 2004) they will incur losses to informed traders when trading on ECNs. This leads to the conjecture that ECNs will not be able to attract a substantial proportion of order flow in ETFs. Contrary to this belief, ECNs have successfully captured a significant portion of ETF order flow (Hendershott and Jones, 2005). In this paper, we provide an in depth analysis of the factors that lead to this contradiction. While it is well known that ECNs dominate the ETF market, few studies investigate the factors that contribute to the success of ECNs in capturing ETF market share.

We suggest that the risk of facing informed traders in anonymous markets such as ECNs is less of an issue for ETFs, which have low adverse selection costs. Liquidity traders face the risk of loosing to informed traders on the ECNs. However, the losses to informed traders are small since ETFs have low adverse selection costs. Traders try to minimize trading costs but also consider another dimension of execution quality such as speed of execution. It is well known that ECNs provide fast execution speed. We suggest that ETF traders are drawn to the ECNs due to the low cost of trading with informed traders and the high execution speed of trades on the ECNs.