Fundamental analysis · Advanced

Quantitative Easing and Quantitative Tightening

The last tightening cycle did not end at a balance-sheet target. It ended when the repo rate spiked to 10% and the Fed started buying bills again.

Last reviewed 15 September 2026 by the MarketPro research desk · Editorial policy

In short

Quantitative easing is a central bank creating reserves to buy bonds, which removes duration from the market and pushes long-dated yields down. Estimates from the US programmes put the effect near 15 to 25 basis points of ten-year yield for every $500bn purchased. Quantitative tightening reverses it, mostly by letting bonds mature without reinvestment, draining reserves from the banking system. Both work through the term premium and the portfolio balance channel rather than through money in circulation.

What to remember

QE is an asset swap, not money in circulation. Reserves are held by banks at the central bank, which is why the 2009 to 2019 programmes sat alongside inflation below target.
The commonly cited estimate is 15 to 25 basis points of ten-year yield per $500bn bought, with a wide error bar around it.
The announced tightening pace overstates the realised one. The mortgage runoff cap was rarely reached once refinancing stopped.
Tightening ends when the funding market breaks, not on a schedule. In September 2019 the overnight repo rate spiked to around 10% intraday.
Reserve balances, the reverse repo facility and the Treasury account matter more than the headline asset total.

Lesson 23 of 24 in Fundamental analysis

PreviousCommodity Prices and Their Ripple Effects NextYield Curves and Inversions

Quantitative easing is a central bank creating new reserves and using them to buy financial assets, usually government bonds, from the market. The purchases push longer-dated yields down and leave the banking system holding reserves where it used to hold bonds. Quantitative tightening reverses it, either by letting maturing bonds run off without reinvestment or by selling them, which drains those reserves back out.

What $2 trillion of bond buying is worth in basis points

The channel that matters is the term premium, the extra yield an investor demands for holding one long bond rather than rolling short ones. Central bank purchases remove duration from the market, and the commonly cited estimates from the US programmes put the effect at roughly 15 to 25 basis points of ten-year yield for every $500bn bought. Two trillion dollars of purchases is therefore something like 60 to 100 basis points, with a wide error bar and real disagreement in the literature about the upper end.

Attach that to a balance sheet and it stops being abstract. A ten-year government bond with a duration of 8.5 gains 8.5 multiplied by 0.70, which is 5.95% in price, for a 70 basis point fall in yield. A company refinancing $500m of ten-year debt saves $3.5m a year on the same move. A pension fund discounting its liabilities at that yield watches its deficit widen. Those are not side effects. They are the policy working as designed.

The scale is easiest to see in the balance sheet total. The Federal Reserve held about $900bn of assets in mid-2008, about $4.5tn by the end of 2014, and about $8.9tn in April 2022. Tightening then took it to roughly $7.2tn over the following two years. Every one of those moves changed how much duration the private sector was left holding.

Reserves are not money in anyone's pocket

This is the part that is routinely got wrong, including by people trading on it. QE is an asset swap. The pension fund that owned a bond now owns a bank deposit, and the bank that intermediated the trade now holds a reserve balance at the central bank. Reserves can only be held by banks, at the central bank. They do not circulate, they cannot be spent in a shop, and the private sector's stock of net financial assets is barely changed by the swap itself.

That is why a decade of asset purchases between 2009 and 2019 sat alongside inflation persistently below target in both the US and the euro area, and why the 2020 to 2021 episode did not. The difference was not the bond buying. It was direct fiscal transfers landing in household accounts at the same time as a supply shock, which is a mechanism QE does not contain.

What QE does do is push holders along the risk curve. Sell your bond to the central bank and you are left with cash earning the policy rate, so you buy the next asset along: corporate credit, then equities, then anything with a yield attached. That portfolio balance channel is how a government bond purchase ends up compressing credit spreads and supporting equity multiples, and it is the defensible version of the claim that QE supports asset prices.

12 more sections in MarketPro

Runoff caps, and why the announced pace overstates the real one

Tightening is usually not selling. The Federal Reserve ran it with monthly caps: up to $60bn of Treasuries and $35bn of mortgage-backed securities allowed to mature without reinvestment, phased in from mid-2022, with the Treasury cap cut to $25bn in June 2024. The Treasury cap was generally reached. The mortgage cap rarely was, because higher mortgage rates stopped households refinancing and prepayments dried up, so realised runoff often came in well below the headline number.

The balance sheet is also not the only dial on system liquidity. The Treasury's cash account at the central bank and the overnight reverse repo facility each move by hundreds of billions, and either can offset tightening or amplify it. The reverse repo balance drained from roughly $2.2tn in early 2023 towards near zero over the following eighteen months, feeding reserves back into the banking system at close to the pace runoff was removing them. Reading the asset total alone that year would have told you the opposite of what was happening to bank reserves.

September 2019, and the way tightening actually ends

The previous tightening cycle did not end at a target balance-sheet size. On 17 September 2019 the overnight repo rate spiked to around 10% intraday, because reserves had been drained past the level the banking system needed to fund itself day to day. The Fed was buying bills again within weeks.

The lesson generalises. Nobody knows the minimum workable level of reserves in advance, the central bank included, because it depends on how banks choose to hold liquidity under rules that keep changing. The floor gets found by hitting it. For anyone positioned on a liquidity view, that means the end of tightening has no scheduled date and tends to announce itself as stress in a funding market rather than as a policy statement.

The currency read, and its honest limits

The textbook chain runs from more base money to a weaker currency. In practice it is the weakest link in the subject. The Bank of Japan ran a balance sheet above 120% of GDP while the Federal Reserve's stayed nearer 35%, and the yen's direction across that period tracked the policy rate differential far more closely than the ratio of balance sheets did.

What moves a currency is the surprise. A programme that is larger, longer or broader than expected reprices the expected path of policy, and the currency moves on that path rather than on the purchases themselves. So use QE and QT to set the liquidity backdrop and to work out what the central bank is trying to do to the curve, then trade the rate expectations that actually price a pair. A balance sheet that has been shrinking at a known pace for eighteen months is not news, and it is not an edge.

Three lines on a Thursday afternoon

The Federal Reserve publishes its balance sheet in the H.4.1 statistical release every Thursday at 16:30 New York time. Three lines are worth more than the headline total: reserve balances held by depository institutions, the overnight reverse repo facility, and the Treasury's general account. Reserves are what the banking system funds itself with, and the other two are the tanks that reserves drain into and out of.

Track the four-week change rather than the week, because settlement timing makes single weeks noisy, and read reserves against the other two rather than on their own. Then treat a liquidity view as a sizing input before a directional one: periods of withdrawal tend to arrive with wider spreads and larger gaps, which is a position size problem first, and the arithmetic for that is in the risk rules.

Finish this lesson in MarketPro

The rest of Quantitative Easing and Quantitative Tightening, a quiz that checks it landed, and your place kept in the curriculum.

12 more sections of this lesson
All 320 lessons, with no cap and no paywall
Free on any account, including a free one

In your browserContinue on webor continue on web

Same account as the app · No install · Free to start

Or scan with your phone camera to get the MarketPro app.

Download on theApp Store Get it onGoogle Play

Free account, no card needed.

Not investment advice. Past performance is not indicative of future results.

Run the numbers while this is fresh

Work it through on your own position size. Nothing leaves your browser and there is no sign-up: Position size calculator.

Written and reviewed by the MarketPro research desk

MarketPro is a trading-signal and trading-education app operated by Harajuku Holdings LTD (Cyprus). Our signal desk publishes and tracks every trade idea in-app, and every page here is checked against the app's live behaviour before publishing. Last reviewed 15 September 2026. How we produce signals and content · About MarketPro

FAQ

Quantitative Easing and Quantitative Tightening: questions

What is the difference between quantitative easing and quantitative tightening?

Quantitative easing is a central bank creating reserves and buying assets, usually government bonds, which removes duration from the market and lowers longer-dated yields. Quantitative tightening is the reverse: the balance sheet shrinks, normally by letting bonds mature without reinvesting the proceeds rather than by selling them, and reserves drain out of the banking system. Easing adds liquidity and compresses the term premium; tightening withdraws liquidity and lets the premium rebuild.

Does quantitative easing cause inflation?

Not on its own, on the evidence available. The US and the euro area ran large purchase programmes from 2009 to 2019 and inflation stayed persistently below target throughout. The reason is that QE swaps one asset for another rather than putting money into household accounts, and bank reserves cannot be spent in the real economy. The 2020 to 2021 inflation came alongside direct fiscal transfers and a supply shock, which are different mechanisms.

How does QE affect a currency?

Less reliably than most explanations suggest. The textbook argument is that expanding the monetary base weakens the currency, but the measurable effect comes from the surprise: a programme larger or longer than expected reprices the expected path of the policy rate, and the currency moves on that. The Bank of Japan ran a balance sheet above 120% of GDP while the yen's direction tracked rate differentials far more closely than the balance sheet itself.

Trade · Learn · Earn

The whole academy is free in MarketPro

All 320 lessons, 12 guided paths, quizzes and a daily streak. Free on every account, in the app or your browser, with vetted free signals alongside it.

In your browserContinue on webor continue on web Download on theApp Store Get it onGoogle Play

Same account as the app · No install · Free to start

Not investment advice. Past performance is not indicative of future results.