How Repo Rate Changes move across Collateral Classes

Views expressed are those of the authors and do not necessarily represent official positions or policies of the Office of Financial Research or the U.S. Department of the Treasury.

Do changes in interest rates on repurchase agreement (repo) borrowings against one type of collateral affect repo interest rates, quantities, and associated activities for other types of collateral? A truism in economics is that changes in prices or quantities for one good or service will affect prices and quantities for all other goods and services, but the strength of such spillovers differs. Often, the more specialized the good or service and crucial it is to the activities of its buyers and sellers, the more limited the spillover. Because the repo market may be divided into segments by collateral type that appear somewhat specialized, potential spillovers are uneven or small.

In the OFR working paper, “Repo Rate Spillovers: Evidence from a Natural Experiment,” OFR researcher Robert Mann focuses on the strength of spillovers and how a securities dealer’s liabilities can affect its operations and potentially lead to less stable securities markets.

A repurchase agreement is a form of secured lending. The borrower sells a security to the lender and simultaneously enters into an agreement to buy the security back, typically at a lower price. The price difference is like interest on the cash the borrower receives from the sale. Cash received is typically less than the value of the security (a haircut), which protects the lender from normal variations in the price of the security if the borrower does not buy it back.

The answer to the question at the beginning of this blog is “yes, but the strength of the spillovers varies”. The author provides evidence by using a surprise 2021 change in Federal Reserve policy. The Federal Reserve implements monetary policy in part by putting a floor on interest rates in the repo market through the Overnight Reverse Repurchase (ON RRP) facility, and in 2021, it increased the ON RRP interest rate by five basis points. The change only directly affected the Treasury-collateralized segment of the repo market but may have indirectly affected other repo financing as market participants adjusted to the change.

The author focuses on the impact of the increase in funding costs on dealers’ repo borrowing in non-Treasury collateral classes. For dealers predominantly funding themselves with repo at interest rates within 5 basis points of the Treasury repo rate before the policy change, the effect on funding costs was large, with less of an effect on those funding themselves at higher rates. In viewing the interest rate relative to the Treasury repo rate as a proxy for the degree of substitutability, the rate impact decreased as substitutability decreased.

Quantities of repo financing also changed. Dealers that saw the highest increase in their repo funding costs did relatively less repo borrowing during the two months after the policy was implemented. Also, such dealers reduced their amount of trading liquidity in other security markets (by decreasing securities inventories and increasing bid-ask spreads). The shock to dealer funding was transmitted to the asset side of the balance sheet. A sufficiently large funding shock might lead to financial instability in the form of impaired functioning of financial markets.