Every so often someone asks me why the fed funds rate stays inside its target range without the Fed constantly buying and selling to keep it there, and the honest answer is that the mechanism people picture is about fifteen years out of date. Most explanations still describe the Fed as a scarcity manager, draining or adding a small amount of reserves each day to nudge the funds rate to a single point. That was true once. It stopped being true after the financial crisis, and the plumbing that replaced it works in a way that is genuinely different, not just a tweaked version of the old thing.
Once you understand the new setup, a whole category of commentary becomes readable. Talk about reserve scarcity, standing repo usage, the funds rate drifting toward the bottom of the range, all of it is describing the same small number of moving parts. So it is worth getting the parts straight.
The old corridor versus the new floor
In the pre-crisis world, banks held very few reserves at the Fed because reserves paid nothing. Holding idle cash there was pure opportunity cost. The Fed set a single target rate, and every morning the open market desk added or drained a modest amount of reserves to push the overnight funds rate to that target. Because reserves were scarce, small adjustments moved the price a lot. That is the corridor system, and it depended on reserves being genuinely tight.
After 2008 the Fed flooded the system with reserves through large-scale asset purchases. Reserves went from scarce to enormous. In that world, adding or draining a few billion does almost nothing to the overnight rate, because the marginal bank is already swimming in cash. The old lever stopped working, so the Fed switched to a floor system.
The idea behind a floor is simple. Instead of managing the quantity of reserves, you set the price the Fed pays on them and let banks hold as much as they want. If the Fed pays a decent rate on reserves, no bank has much reason to lend overnight at a rate meaningfully below that, because lending to the Fed is about as safe as it gets. The administered rate becomes a floor that the market rate rests on top of, and the Fed no longer has to fight the quantity every morning.
IORB and the ON RRP, the two rails
Two administered rates do most of the work now, and it helps to think of them as the two rails the funds rate rides between.
IORB, interest on reserve balances. This is the rate the Fed pays banks on the reserves they park at the Fed. It is the main tool. Raise IORB and you raise the whole overnight complex, because banks will not lend below what they can earn risk-free at the Fed. Lower it and everything eases. When people say the Fed hiked or cut, the IORB is usually the number doing the real steering.
There is a wrinkle, though. Not every player in the overnight market can earn IORB. Only banks have reserve accounts. Money market funds, government-sponsored enterprises like the Federal Home Loan Banks, and various other lenders cannot park cash at the Fed and collect IORB. So they will sometimes lend in the funds market below IORB, which is why the effective funds rate typically trades a touch under it rather than exactly on it.
ON RRP, the overnight reverse repo rate. This is the second rail, and it exists to catch the lenders that IORB cannot reach. The ON RRP facility lets a much wider set of counterparties, money funds especially, lend cash to the Fed overnight against collateral and earn a set rate. That rate sits a bit below IORB and functions as a harder floor under the whole market. If a money fund can get the ON RRP rate risk-free from the Fed, it has little reason to lend to anyone else for less. So the ON RRP props up the bottom of the range where IORB alone would leak.
The target range the Fed announces is the band those two rails hold together. IORB sits near the top of the operating band, the ON RRP sits near the bottom, and the effective funds rate settles somewhere in between. The Fed is not aiming a single arrow anymore. It is setting the edges of a pen and letting the rate wander inside it.
Why this only works with ample reserves, and where it wobbles
The floor system depends on one condition: reserves have to stay ample. As long as banks collectively hold more reserves than they strictly need, the marginal dollar earns the administered rate and the floor holds cleanly. The trouble starts when reserves drift back toward scarcity, usually because the Fed has been shrinking its balance sheet, letting bonds roll off without replacing them.
As reserves tighten, you move back up the demand curve toward the region where quantity starts to matter again. Individual banks begin to feel short at certain moments, month-ends and quarter-ends especially, when balance-sheet constraints bite and everyone wants cash at once. The classic warning sign is the funds rate and repo rates spiking above where the administered rates say they should sit. That kind of spike is the market telling you reserves are no longer as ample as the Fed assumed. The repo turmoil of a few years back was exactly this, a reserve-scarcity flare-up that the floor was supposed to prevent.
The Fed's answer to those flare-ups is the standing repo facility, a ceiling to go with the floor. It lets eligible institutions borrow cash from the Fed overnight against Treasuries at a set rate. If overnight rates try to spike above that rate, borrowers tap the facility instead of paying up, which caps the spike. So the modern system is really a floor plus a ceiling, ON RRP and IORB holding the bottom, the standing repo facility guarding the top.
Here is a quick way to read the signals without overthinking it:
- Effective funds rate drifting toward the bottom of the range, or ON RRP balances swelling, usually means cash is abundant and looking for a home. Reserves are more than ample.
- Effective funds rate creeping toward the top of the range, or repo rates poking above IORB around quarter-ends, suggests reserves are getting tight and the floor is under pressure.
- Standing repo facility usage jumping from roughly zero to something visible is the clearest tell that scarcity has arrived and the ceiling is doing real work.
None of this tells you where rates are headed, and I would not pretend it does. What it gives you is a way to tell whether the Fed is comfortably in control or quietly running short of the reserves its own system depends on. When you next read someone worrying about the plumbing rather than the policy, that distinction is usually what they are circling. The policy rate is set in a meeting. Whether the machinery can actually hold that rate is a separate question, and it lives entirely in these few numbers.