The U.S. Federal Reserve faces a monetary-policy challenge above and beyond determining the right level of short-term interest rates: how much and how quickly to reduce the more than $7 trillion in securities still on its balance sheet—holdings it amassed in previous years to help stimulate growth.

Back in September 2019, such quantitative tightening didn’t end well. Money markets buckled and short-term interest rates spiked as banks suddenly found themselves short of cash reserves. I expect that experience will make the central bank more careful this time.

The Fed’s securities portfolio plays a crucial role in the supply of bank reserves. When it declines, other investors naturally end up holding more securities, draining cash from their bank accounts and hence reducing the total amount of reserves that banks keep on deposit at the Fed. Since April 2022, the Fed has been in tightening mode, with its holdings of Treasuries and mortgage-backed securities currently running off at a rate of about $75 billion to $80 billion a month.

The goal is to reach what Fed officials call an “ample” level of reserves—high enough to meet banks’ needs and promote stable short-term interest rates, but no higher than needed to “efficiently and effectively” implement monetary policy. This means the interest rate the central bank pays on reserves—rather than the quantity of reserves—should determine other short-term rates, such as the federal funds rate at which banks lend to one another.

Problem is, nobody knows exactly where the “ample” level is. At about $3.5 trillion—more than double what they were in September 2019—bank reserves are almost certainly excessive. That said, short-term lending markets experienced upward pressure on rates at the end of last year, suggesting that banks’ liquidity needs might be greater than previously thought.

Hence, the Fed will have to move cautiously. Officials are already discussing when to slow the pace of quantitative tightening. Presumably they’ll act in the first half of this year, perhaps halving the runoff rate caps. But the impact of this change on the trajectory of reserves isn’t entirely within the Fed’s control. One example: Money-market mutual funds, which are responsible for most of the $700 billion parked at the Fed’s reverse repo facility, may continue to shift their cash into higher-yielding assets, increasing the supply of bank reserves.

The greatest danger is that markets will incorrectly interpret a decision to slow the runoff rate as a harbinger of rate cuts. If that happens, not only might investors be wrong-footed, but also the Fed might have to keep rates higher for longer to counteract excessive easing in broader financial conditions.

It’s also possible that the Fed will allow reserves to fall too much again, but this seems much less likely. The central bank is determined to avoid such an outcome. Moreover, even if this did occur, the risk of a rate spike is much lower now because the Fed has established a liquidity backstop, known as the standing repo facility, that banks can tap if reserves become too scarce.

The economic impact of slower quantitative tightening will be negligible: The destination in terms of the amount of bank reserves and the size of the Fed’s balance sheet is essentially the same. The only substantive difference is how long it will take to get there.

The Fed’s recent focus on the pace of quantitative tightening suggests it has learned the lesson of 2019. If it proceeds as I think it will, any adjustments should be nothing to fear. They also should be considered independent from the Fed’s decisions on interest-rate cuts.

Bill Dudley, a Bloomberg Opinion columnist, served as president of the Federal Reserve Bank of New York from 2009 to 2018. He is the chair of the Bretton Woods Committee, and has been a nonexecutive director at Swiss bank UBS since 2019.