Do retail traders suffer from high frequency traders?

Using a change in regulatory fees in Canada in April 2012 that affected algorithmic quoting activities, we analyze the impact of high frequency quoting and trading on market quality, trader behavior, and trading costs and profits. Following the change, algorithmic message traffic, i.e. the number of orders, trades, and order cancellations, dropped by 30% and the bid-ask spread rose by 9%. Using trader-level data, we attribute this change to message-intensive algorithmic traders reducing their activity, and we show that their reduced activity had a negative impact on retail traders’ intraday returns, in particular on their returns from limit orders. We further find that institutional traders’ intraday returns from market orders increased. Financial support from the SSHRC (grant number 410101750) is gratefully acknowledged. We thank seminar participants at the 2013 WFA, the 2013 Central Bank Workshop on Microstructure (ECB), Erasmus University Rotterdam, the Free University of Amsterdam, and KU Leuven for comments. Special thanks to Mark Seasholes (WFA discussant), Bernt-Arne Odegaard (ECB discussant), Jonathan Brogaard, Terry Hendershott, Rob McMillan, David Panko, Elvira Sojli, and Mark Van Achter for detailed comments. The TMX Group kindly provided us with databases. The views expressed here are those of the authors and do not necessarily represent the views of the TMX Group. TSX Inc. holds copyright to its data, all rights reserved. It is not to be reproduced or redistributed. TSX Inc. disclaims all representations and warranties with respect to this information, and shall not be liable to any person for any use of this information. University of Toronto, [email protected] University of Toronto [email protected] (corresponding author) University of Ontario Institute of Technology [email protected] Although technological innovation has always played a critical role in financial markets, the unprecedented growth of automated, algorithmic trading in equity markets over the past decade has been the source of much controversy. Computer algorithms, capable of making and implementing trading decisions at speeds that are orders of magnitude faster than human reaction times, create, execute, modify, and cancel orders at microsecond speeds. To provide an example, in the U.S. during the Dotcom bull market in 2000, there were on average about 5 million trades and quotes per day ; in the fall of 2012, at peak times there were up to 5 million trades and quotes per second. The initial growth of algorithmic trading was associated with a decline in trading costs, and it was viewed as a positive development by market participants and academics. For instance, using the introduction of automated quotes on the New York Stock Exchange in 2003, Hendershott, Jones, and Menkveld (2011) documented that an increase in algorithmic trading causally improved liquidity. Yet market participants now frequently report that quotes change so frequently that the lower bid-ask spreads are only an illusion of liquidity and that quotes evaporate before traders are able to trade against them. Moreover, processing millions of orders, cancellations, and trades is costly and requires that dealers, exchanges, and regulators heavily invest in IT infrastructure. We analyze the empirical impact of intense quoting activity on market liquidity, on trader behavior, and on trading costs of market participants with different levels of sophistication. As a first step, we want to understand whether the decline in trading costs, documented by Hendershott, Jones, and Menkveld (2011), extends beyond the phase of initially modest use of automation and algorithmic trading. Second, equipped with trader-level data, we aim to understand who benefits from the change in trading costs. In See Larry Tabb’s testimony to U.S. Congress, available at http://www.banking.senate.gov. On August 8, 2011, the U.S. credit rating was downgraded, the number of trades and quotes was almost 2.3 billion; see http://www.nanex.net/aqck2/3528.html. See The Economist, February 25, 2012: “The fast and the furious”.