FIN-01-013

NYU Stern School of Business


Risk Management with Derivatives by Dealers and Market Quality in Government Bond Markets

September 21, 2001

Narayan Y. Naik and Pradeep K. Yadav

ABSTRACT


This paper examines how bond dealers use futures markets to manage the hedgeable market risk component of their core business risk exposure, and whether market quality is adversely affected by their selective risk taking activity. It also investigates the efficiency of market risk sharing within a decentralized semi-transparent market structure. We find that dealers engage in duration targeting, behaving as if they have a comparative advantage in bearing interest rate risk. They make significant directional bets often by holding futures that are in the same direction as the spot. They actively use futures to hedge changes in the spot exposure. They hedge changes in their spot exposure more when the potential costs of regulatory distress are high, when the cost of such hedging is low, and during periods of greater uncertainty. We find that duration targeting by dealers has adverse price effects due to capital constraints as predicted by Froot and Stein (1998). Finally, we find that trades in the spot market are not executed by dealers with extreme exposures. In this context, we recommend market reforms such as introduction of central quote posting or limit order book that will enable more efficient matching of liquidity demanders and suppliers, reduce trading costs, and improve the quality of risk sharing.

Classification: G10, G20, G24

Narayan Y. Naik
Institution: London Business School, UK
Telephone: +44 207 2625050
Fax: 212-995-4233
Email: nnaik@london.edu
Home Page: http://www.london.edu/faculty_research/

Pradeep K. Yadav
Institution: Leonard N. Stern School of Business, New York University
Telephone: 212-998-0305
Fax: 212-995-4233
Email: pyadav@stern.nyu.edu
Home Page: http://www.stern.nyu.edu/~pyadav


To download a copy of this paper click here
To request a copy of this paper click here

The Finance Department Working Paper Series has been generously sponsored by