• Document: Identifying Informed and Liquidity Traders in Futures Markets
  • Size: 403.46 KB
  • Uploaded: 2019-05-16 23:30:56
  • Status: Successfully converted

Some snippets from your converted document:

Identifying Informed and Liquidity Traders in Futures Markets Raymond P. H. Fishe Aaron D. Smith Abstract We use data on the daily positions of futures market participants to identify informed traders, suppliers and demanders of liquidity, momentum traders, and contrarian traders. These data cover the period from 2000 to mid-2009 and contain 8,921 unique traders. We identify between 94 and 333 traders as informed about next day’s prices and 91 as informed intraday, with little overlap between these two groups. We also analyze the reported line of business (e.g., producer, hedge fund, swap dealer, etc.) and trading behavior of those identified by our methods. Floor brokers/traders are over-represented in the next-day informed group, which suggests the ability to process order flow information creates success at this horizon. The intraday informed group is dominated by managed money traders/hedge funds and swap dealers, with commercial hedgers significantly under-represented in this group. We find no evidence that commercial hedgers pay a risk premium to speculators. In fact, our results contradict hedging pressure theory as we find that liquidity demanders tend to be managed money/hedge fund traders and liquidity suppliers tend to be firms with a natural hedging motive; that is, those with a commercial interest in the underlying commodity. We find that trader characteristics— experience, size and activity—offer stronger predictive power for who is informed than the business line information. Keywords: Commodities, False Discovery Rate, Forecasting Ability, Informed Traders JEL classification: G10, G13 First Draft: April 2010 This version: September 2010 Fishe: CFTC and Department of Finance, Robins School of Business, University of Richmond, Richmond, VA 23173. Tel: (+1) 804-287-1269. Email: pfishe@richmond.edu. Smith: CFTC and Department of Agricultural and Resource Economics, University of California - Davis, Davis, CA 95616, Tel: (+1) 530-752-2138. Email: adsmith@ucdavis.edu. We thank seminar participants at the University of Illinois – Urbana-Champaign, University of California – Davis, Commodity Futures Trading Commission, and especially Brad Barber, Roger Edelen, Steve Kane and David Reiffen for detailed thoughts. All errors are our own responsibility. This paper reflects the opinion of its authors only, and not that of the CFTC, the Commissioners, or any other staff at the CFTC. Identifying Informed and Liquidity Traders in Futures Markets I. INTRODUCTION Commodity futures markets are populated by heterogeneous traders who vary by information, motive, and skill. Theoretical models recognize this heterogeneity in various ways. Classic risk transfer models tend to focus on heterogeneity of motives and thereby assume homogeneous information and skill across traders. Such models underlie the theory of normal backwardation, which posits that short hedgers pay a premium to transfer price risk to speculators (Keynes (1930), Hicks (1939) and Cootner (1960)). Market microstructure models focus on information heterogeneity by defining some traders as informed about the fundamental value of the asset, some as market makers, and assigning others a liquidity motive or treating them as uninformed (e.g., Madhavan (2000) and Easley and O’Hara (2003)). Both the risk transfer and microstructure approaches make predictions about the price-discovery mechanism and how trader’s positions correlate with prices. A common way to test predictions from these models is to classify traders ex ante and investigate whether each trader type behaves and earns payoffs that are consistent with the theory (e.g., Houthakker (1957), Chang (1985), Hartzmark (1987), Lyons (1995), Manaster and Mann (1996), Biais, Glosten, and Spatt (2005)). Thus, given the classification, the theory is supported or not by the data. In this article, we apply the opposite approach. Using a comprehensive daily dataset of individual trader positions, we first group participants as informed traders, suppliers or demanders of liquidity, momentum traders, or contrarian traders by how their positions correlate with prices. From these results, we then classify traders in each group by their reported line of business (e.g., producers, hedge funds, swap dealers, etc.) and by trading variables such as average position size, frequency of trading, and experience. Finally, we assess the extent to which our results confirm or contradict the predictions of various theoretical models. Our approach has two immediate advanta

Recently converted files (publicly available):