Quantitative Tightening: What Happens When a Central Bank Unwinds Its Balance Sheet

Quantitative easing is reasonably well understood by now: a central bank creates reserves and uses them to buy bonds, pushing down long-term yields and adding liquidity.

The reverse is discussed far less, happens far more slowly, and is in some ways the more interesting problem - because nobody knows exactly where it has to stop.

What is actually being unwound

During quantitative easing, a central bank buys government bonds and, in some cases, mortgage-backed securities. It pays by crediting the seller's bank with newly created reserves - deposits held at the central bank.

Two things result. The central bank holds a large portfolio of bonds. Commercial banks hold a large quantity of reserves.

Quantitative tightening reverses both. As the bond holdings decline, the reserves created to acquire them are extinguished. The balance sheet shrinks on both sides at once.

Runoff, not selling

The obvious way to shrink a bond portfolio would be to sell the bonds. Central banks generally do not do this.

Instead they use runoff: when a bond matures, the issuer repays the principal, and the central bank simply declines to reinvest it. The holding falls passively as the portfolio ages.

Two reasons, both practical.

Selling would move prices. Dumping a very large quantity of bonds into the market pushes yields up sharply and disorderly. Runoff produces the same eventual reduction without the market impact.

Selling realises losses. Bonds bought when yields were low are worth less after yields rise. Held to maturity, they repay at face value and the loss is never crystallised. Sold, it is - with awkward consequences for the central bank's reported accounts and its remittances to the treasury.

Runoff has a drawback: the pace is set by the maturity profile of the portfolio, not by policy. Some months a large quantity matures; other months very little does. Central banks manage this by capping how much is allowed to roll off per month and reinvesting the excess, which smooths what would otherwise be a lumpy path.

How it actually tightens

The transmission is less direct than a rate change, and it runs through two channels.

Reserve scarcity. As reserves drain from the banking system, banks have less spare liquidity. They compete more for funding, short-term rates firm, and lending standards tighten at the margin.

Term premium. Quantitative easing compressed long-dated yields partly by removing bonds from the available supply. Withdrawing that support allows the premium investors demand for holding long-dated debt to rebuild, which steepens the curve.

Both channels are slower and less precise than moving the policy rate. This is why central banks describe the balance sheet as running on a separate track - set to a predictable path and left alone, while rates do the active work of responding to data.

The problem nobody can solve in advance

Here is the genuinely hard part.

Before quantitative easing, banks held minimal reserves. After it, they held enormous quantities. Quantitative tightening drains them back down - toward a level that is not known.

The banking system needs some quantity of reserves to function. Regulations introduced after the financial crisis require banks to hold substantial liquid assets, and reserves are the most liquid asset available. That has raised the floor considerably, and by an amount that cannot be calculated precisely from the outside or, apparently, from the inside.

So the central bank reduces reserves gradually and watches for signs of strain. The trouble is that the transition from ample to scarce is not gradual. It is a threshold, and it is discovered by crossing it.

What happened in 2019

The Federal Reserve's first attempt at balance sheet reduction ran from 2017. It proceeded quietly for around two years.

In September 2019, the overnight repo rate - the cost of borrowing cash against government bonds, normally one of the most stable rates in finance - spiked sharply. Reserves had fallen further than the system could comfortably absorb, and short-term funding markets seized.

The Fed intervened quickly and began adding reserves again. The episode ended the tightening programme.

The lesson was not that balance sheet reduction is impossible. It was that the floor is found by hitting it, and that monitoring money market conditions matters more than hitting any particular balance sheet target.

Subsequent programmes have been designed with that in mind - explicit caps, published plans, and standing facilities intended to relieve funding pressure before it becomes disorderly.

What to watch

Four indicators, all published.

The balance sheet itself. Central banks publish weekly or monthly. The trend and pace are directly observable.

Reserve balances. The quantity of reserves in the banking system, and the direction of travel.

Short-term funding rates. Repo rates relative to the policy rate. Persistent firmness here is the early signal that reserves are becoming scarce - this is what moved before the 2019 disruption.

Usage of standing facilities. Rising use of a central bank's overnight lending window suggests institutions are finding liquidity harder to source privately.

The bottom line

Quantitative tightening is the slow, deliberate unwinding of a decade of bond buying, done mostly by letting bonds mature rather than by selling them.

It tightens conditions through the quantity of reserves rather than the price of money, which makes it gradual and imprecise. And it runs toward a floor whose location nobody knows, which is why it is managed by watching funding markets rather than by aiming at a number.

The data needed to follow it is published on a schedule. Very few people read it, and it explains a good deal about why liquidity conditions change without any announcement.

This article is educational and is not financial advice. Monetary policy operations are complex and their market effects are contested.