Tomas Breach, Thomas B King
Review of Finance, Volume 29, Issue 1, January 2025, Pages 33–73, https://doi.org/10.1093/rof/rfae033
Short-term borrowing arrangements collateralized by financial assets—collectively referred to as “securities financing” transactions—are key mechanisms through which investors obtain leverage and engage in arbitrage. An important but opaque segment of this market involves dealers funding the securities positions of their clients, most often through bilateral repurchase agreements (repo), though such activity may also include margin lending and other similar transactions. Despite the theoretical interest in this market and its evident practical relevance, empirical facts are remarkably hard to come by. For example, while there is broad consensus that financing constraints had important effects on the liquidity and pricing of certain securities during the GFC, there is no systematic evidence on their impact during normal times or in the more-recent market deterioration around the advent of COVID-19.
We provide new evidence on bilateral dealer-to-client securities financing and its relationship to the respective cash markets for securities by exploiting the Senior Credit Officer Opinion Survey, or “SCOOS.” Every quarter, the SCOOS surveys the credit officers responsible for setting securities-financing terms at the roughly twenty broker-dealers with the largest presence in bilateral securities financing. The survey asks about the various terms on financing transactions and related information. Our data cover seven asset classes over the period 2010-2020.
The survey shows that dealers tend to change all types of terms (financing spreads, haircuts, credit limits, and maturity limits) together, suggesting that one or more common factors drive bilateral financing conditions. We present evidence on what those factors are by matching the SCOOS—by quarter and, where possible, by asset class—with a variety of data on market conditions, including financing and trading volumes, asset returns, securities issuance, and various measures of risk and volatility. The factor that emerges as most important is the liquidity of the underlying securities markets. Indeed, the inclusion of liquidity largely renders other measures of market conditions, such as volatilities and credit-risk spreads, insignificant in regressions. However, we do also find evidence that dealer balance-sheet constraints play a role in funding markets. In particular, controlling for other market conditions, we show that dealers tighten financing spreads and haircuts for less-liquid asset classes (consumer ABS, CMBS, and private-label RMBS) when their own equity positions worsen, suggesting a desire to preserve capital.
As an additional empirical exercise, we ask whether changes in funding terms themselves have explanatory power for market conditions, using as instruments self-reported changes in terms that are due to “market conventions.” We find that funding terms typically have little effect on asset-market liquidity and returns, although we do find some significant effects during the stress of early 2020.
Taken together, our results provide a number of facts about bilateral securities financing that may help to refine theoretical work in this area. Although our primary contribution is empirical, we also sketch a simple theoretical model that can rationalize our main results.