Market Microstructure Meeting
November 30, 2012
Tarun Chordia and Amit Goyal, Emory University; Charles Jones, Columbia Business School; Bruce Lehmann, University of California, San Diego and NBER; Gideon Saar, Cornell University; and Avanidhar Subrahmanyam, University of California, San Diego, Organizers
Carole Comerton-Forde, Australian National University, and Talis Putnins, University of Technology, Sydney
Dark Trading and Price Discovery
Regulators and stock exchanges around the world are concerned that growth in the share of equities volume executed without pre-trade transparency -- so called "dark trading" -- may harm price discovery. Comerton-Forde and Putnins empirically analyze the impact of dark trading on price discovery. They find that high levels of dark trading impede price discovery and cause prices to become less informationally efficient. Order flow that migrates to the dark is less informed than what is left behind, but it is not entirely uninformed. Therefore, the loss of pre-trade information on migrating order flow harms price discovery. It also increases adverse selection risk, bid-ask spreads, and price impact in the transparent exchange. This decreases the incentives to engage in costly information acquisition, thereby further reducing the informational efficiency of prices. The authors find no evidence that large block trades that occur in the dark will impede price discovery.
Jean-Edouard Colliard, European Central Bank
Catching Falling Knives: Speculating on Market Overreaction
Some agents on financial markets have private information not only about fundamentals, but also about the market itself. Particular funds or high-frequency traders are better at inferring information from past trades, and thus at identifying over- or undervalued assets. Colliard shows that the dynamic trading behavior of such agents has an ambiguous impact: they decrease the likelihood to observe large price deviations, but they slow down price discovery in the long run. They increase adverse selection, and eventually will be excluded by too high spreads if their informational advantage is too low. Such agents insure the market against short-run crashes by "catching falling knives" -- a service for which they earn a profit on average, with large gains when prices bounce back, but losses otherwise. Their profit distribution displays a high variance and fat tails; thus, it is likely that not enough agents acquire this type of information, and that insurance against crashes is under-provided.
Pete Kyle and Anna Obizhaeva, University of Maryland
Large Bets and Stock Market Crashes
Market microstructure invariance predicts a much greater price impact for market-wide selling pressure than does the conventional wisdom. For five stock market crashes that Kyle and Obizhaeva examine, invariance predicts price declines that are similar to observed price changes. Accurate predictions of price declines for the 1987 crash and for the 2008 sales of Societe Generale suggest that early warning systems are feasible. In two flash crashes, price declines temporarily overshot predictions from invariance, suggesting that rapid selling exacerbates transitory price impact. Smaller-than-predicted price declines for the 1929 crash suggest that less integrated markets are more resilient to market-wise selling pressure. Large quantities sold in three crashes suggest fatter tails or larger variance than the log-normal distribution estimated from portfolio transitions data.
David Easley and Maureen O'Hara, Cornell University, and Liyan Yang, University of Toronto
Opaque Trading, Disclosure, and Asset Prices: Implications for Hedge Fund Regulation
Easley, O'Hara, and Yang investigate the effect of ambiguity about hedge fund investment strategies on market efficiency and aggregate welfare. They model some traders (mutual funds) as facing ambiguity about the equilibrium trading strategies of other traders (hedge funds). This ambiguity limits the ability of mutual funds to infer information from prices and has negative effects on market performance. They use this analysis to investigate the implications of regulations that affect disclosure requirements of hedge funds or the cost of operating a hedge fund. Their analysis demonstrates how regulations affect asset prices and welfare through their influence on opaque trading.
Burton Hollifield and Artem Neklyudov, Carnegie Mellon University, and Chester Spatt, Carnegie Mellon University and NBER
Bid-Ask Spreads and the Pricing of Securitizations: 144a vs. Registered Securitizations
Traditionally, various types of securitization have been traded in opaque markets. As an initial step towards increasing transparency and enhancing its understanding of these markets, the Financial Industry Regulatory Authority (FINRA) began to collect transactions data from broker-dealers (without any public dissemination) in May 2011. Securitization markets are highly fragmented and require transaction matching methods to construct bid-ask spreads. Hollifield, Neklyudov, and Spatt study the relationship between those bid-ask spreads and transaction characteristics, such as the size of the underlying trade and the path by which trade execution and intermediation occurs. Retail-sized transactions lead to relatively wide spreads because of the absence of competition, but institutionally-sized transactions often result in much tighter spreads. The authors study the contrast between registered instruments that can be traded freely and Rule 144a instruments with much more limited disclosures that only can be purchased by sophisticated investors. They also study the structure of the dealer network and how it influences the nature of bid-ask spreads. Some dealers are relatively central in the network and trade with many other dealers, while others are more peripheral. Central dealers receive relatively lower spreads than peripheral dealers. This could reflect greater competition and reduced bargaining power of central dealers, or lower costs for the transactions which they intermediate. The order flow is more evenly divided among dealers and the customer spreads are relatively smaller for central dealers in Rule 144a than in registered instruments.
Matthew Baron, Princeton University; Jonathan Brogaard, University of Washington; and Andrei Kirilenko, Commodity Futures Trading Commission
The Trading Profits of High Frequency Traders
Baron, Brogaard, and Kirilenko examine the profitability of high frequency traders (HFTs). Using transaction-level data with user identifications, they find that high frequency trading (HFT) is highly profitable: HFTs collectively earn over $23 million in trading profits in the E-mini S&P 500 futures contract during the month of August 2010. The profits of HFTs are derived mainly from Opportunistic traders, but also from Fundamental (institutional) traders, Small (retail) traders, and Non-HFT Market Makers. While HFTs bear some risk, they generate unusually high average Sharpe ratios with a median of 4.5 across firms in August 2010. Finally, HFTs profits are persistent, new entrants have a higher propensity to underperform and exit, and the fastest firms (in absolute and in relative terms) earn the highest profits.