Skip to main content

Please enter a keyword and click the arrow to search the site

Liquidity risk and specialness



Publishing details

IFA Working Paper

Publication Year



Repo contracts, the most important form of collateralized lending, are widely used by financial institutions and hedge funds to create short selling positions and manage their leverage profile. Moreover, they have become the primary tool of money management and monetary control of several banks, including the Bundesbank and the newly born European Central Bank. This paper is an empirical study of this market. More specifically, we study the extent to which the repo market discounts the future specialness of Government bonds. We address this issue in the framework of the expectations hypothesis. We construct a daily data set of term repo rates on Government bonds containing both general collateral and "Special" repo rates for different tenors. We study the extent to which the current term structure of long term "special" repo spreads discount the future collateral value (specialness) of Treasuries. We ask if repo spreads embed a liquidity risk premium and whether this risk premium is time varying. We use a simultaneous regression approach and find that, allowing for a constant (term) risk premium, current long term "special" repo spreads strongly overestimate changes in the future collateral value (Specialness) of the underlying asset. Using Monte Carlo techniques, we find that these findings are not a statistical artefact due either to small sample bias or measurement errors. Is this due to an over-reaction of the market, or is this explained by a time-varying risk-premium? Using a Garch-in-mean specification, we find that the conditional volatility of special repo spreads is a priced risk factor. This seems indicative of the existence of a time varying liquidity risk premium.

Publication Research Centre

Institute of Finance and Accounting

Series Number

FIN 309


IFA Working Paper

Available on ECCH


Select up to 4 programmes to compare

Select one more to compare
subscribe_image_desktop 5949B9BFE33243D782D1C7A17E3345D0

Sign up to receive our latest news and business thinking direct to your inbox