Repo Rates vs Central Bank Policy Rates: A Practical Guide

The same word for different transactions

“Repo” is used loosely. In a classic repo, one party sells securities to another and agrees to buy them back at a later date for a higher price. The difference between the sale price and the repurchase price is the repo interest. The securities are collateral; the cash is the loan.

The New York Fed describes the transaction from its own side: in a repo, the Desk “purchases securities from a counterparty subject to an agreement to resell the securities at a later date”. It calls this “economically similar to a loan collateralized by securities”. A reverse repo is the mirror image: the Desk sells securities subject to an agreement to repurchase them, which is “economically similar to borrowing collateralized by securities”.

A policy rate is a different kind of object. It is an administered rate, set by a central bank’s policy committee, that the central bank implements through its operations. The Federal Reserve’s target is for the federal funds rate, an unsecured overnight rate at which banks lend reserves to each other.

How the policy rate is implemented

The Fed does not set every overnight rate by decree. It sets a target range and uses tools to keep market rates inside it. The New York Fed’s page describes two standing facilities that bracket the corridor:

• Overnight reverse repo (ON RRP). Eligible counterparties lend cash to the Fed overnight against Treasury collateral. The facility “limits downward pressure and helps provide a floor under overnight money market rates”, because a counterparty can earn the offering rate rather than lend below it.
• Standing repo (SRP). The Desk lends cash against Treasury, agency and agency mortgage-backed securities, providing a ceiling for the secured rate.

Both run each business day at pre-announced rates set by the FOMC. The corridor is not a fixed box; the tools, the counterparties and the collateral all shape where the market actually trades.

Why repo can trade away from the policy rate

The link between the two rates is arbitrage, and arbitrage has costs.

1. Collateral. Repo is secured; fed funds are unsecured. A lender will accept a lower rate on secured cash, all else equal, because the credit risk is lower. The size of that discount depends on the collateral.
2. Balance sheet. A dealer that intermediates repo must fund the position and hold it against capital and leverage rules. At quarter-end and year-end, the cost of showing the position on a reporting date rises, and repo rates can move sharply.
3. Scarcity. When Treasury collateral is scarce, the rate to borrow cash against it can fall, and specials can trade well below the general collateral rate.
4. Counterparty limits. A cash lender may have a credit limit for a given borrower, which segments the market and drives a wedge between rates for different counterparties.
5. Timing and settlement. Overnight repo settles on a specific schedule. A late settlement or a failed trade can create a temporary premium.

None of these breaks the relationship. They explain why the observed repo rate is a market price, not a copy of the policy rate.

A worked example

Suppose a cash lender places 100 million in overnight repo at 4.30%, with Treasury collateral and a 2% haircut. The interest for one day, on an actual/360 convention, is:

100,000,000 × 0.0430 / 360 = 11,944.44

If the policy target midpoint is 4.25%, the repo rate is 5 basis points above it. On 100 million, 5 basis points for one day is about 138.89. Over a year of rolling overnight, the compounding of that spread is small but not zero, and the rollover risk is not.

The haircut matters separately. A 2% haircut means the lender advances 98 million against 100 million of collateral, so the effective return on cash is slightly higher than the quoted repo rate if the collateral is sold and reinvested, and the lender is protected against a small fall in collateral value. The borrower, in turn, has funded less than the face value of the securities.

Term repo versus overnight

An overnight repo is repriced every day. A term repo fixes the rate for a period, often one week to three months. The term rate embeds expectations of future overnight rates plus a term premium, and it can diverge from the overnight rate around expected policy changes.

For a borrower, the choice is between certainty and flexibility: term repo locks the funding cost but commits to the rollover, while overnight repo follows the market and exposes the borrower to a funding squeeze. For a lender, the trade-off is the reverse.

What the comparison is good for

The policy rate is the anchor for the cost of money. The repo rate is the price at which collateral actually changes hands. Comparing them tells you:

• Whether secured funding is tight or abundant. A repo rate persistently above the policy rate suggests collateral or balance-sheet pressure; below it suggests abundant cash and scarce collateral.
• Whether a quarter-end or year-end effect is at work. A temporary spike that reverses is a different signal from a persistent shift.
• What a cash holder is actually paid. The policy rate is not directly available to every institution; the repo rate is the rate at which they can lend.

What the comparison does not tell you is the direction of policy. A repo spike is a market event, not a committee decision. Treating every funding squeeze as a rate hike is the most common error in reading the two series together.

The practical checklist

Before comparing a repo rate with a policy rate, check the collateral, the counterparty, the tenor, the haircut and the settlement date. A 4.30% overnight rate against Treasury collateral and a 4.30% rate against agency mortgage-backed securities are not the same price, even when the number matches.

The policy rate is a reference. The repo rate is a transaction. The gap between them is where the money market’s frictions become visible.