The £895 billion question: why didn't QE cause inflation?
Britain spent over a decade pumping money into the economy. Prices barely moved. Here's what actually happened and why the answer is more complicated than it looks.
In March 2009, the Bank of England did something it had never done before. Faced with a collapsing economy and an interest rate already at 0.5%, effectively as low as it could go, it announced it would start buying financial assets using newly created central bank money. The programme would eventually total £895 billion across four separate rounds. It was called quantitative easing.
The stated goal was straightforward: inject money into the financial system, lower borrowing costs, encourage spending, and push inflation back up to the 2% target.
It didn’t work. Not in the way it was supposed to.
For over a decade, UK inflation stayed persistently weak. Wages barely grew in real terms. The economy limped along well below its potential. And the demand-driven inflation that QE was designed to generate, the kind that comes from people spending confidently, wages rising, businesses investing, never really materialised.
But here’s where it gets interesting: that doesn’t necessarily mean QE failed.
First, what was QE actually supposed to do?
The theory behind QE runs through several channels, each of which was supposed to feed into higher spending and, ultimately, higher prices.
The most important was the portfolio rebalancing channel. When the Bank of England buys government bonds, it drives up their prices and pushes down their yields. Investors holding those bonds suddenly find themselves sitting on cash, which they reinvest in other assets, such as equities, corporate bonds, or property. This flows through financial markets: asset prices rise, borrowing costs fall, and in theory, households feel wealthier and spend more.
The second was the credit channel. With banks flush with reserves from the Bank’s asset purchases, the expectation was that they’d lend more freely to households and firms, supporting consumption and investment.
The third was expectations. By signalling a long commitment to loose monetary policy, the Bank hoped to convince households and businesses that better times were coming, encouraging them to spend now rather than wait.
All three channels pointed in the same direction: more aggregate demand, higher wages, and eventually, inflation.
So why didn’t it happen?
What the data actually shows
UK CPI inflation spent most of the period from 2009 to 2020 at or below the 2% target. The exceptions, notably around 2011 and again after 2016, were driven not by strong domestic demand but by external shocks: a collapse in sterling, energy price spikes, and changes in VAT. Strip those out, and the underlying picture is one of persistent weakness.
Wages tell the same story. Nominal pay grew modestly throughout the decade, but once you adjust for inflation, real wage growth was anaemic, close to zero for much of the period. In a genuinely demand-driven economy, you’d expect wages to be pulling ahead of prices, with workers bargaining for more as employers compete for their labour. That dynamic simply didn’t emerge.
The output gap tells the same story from a different angle. Real-time estimates put the negative output gap at between -4% and -5% of potential GDP in 2009 - a vast ocean of spare capacity that should have made inflation almost impossible. Later OBR revisions scaled this back to around -2.5% to -3% as it became clear that the crisis had permanently destroyed potential output through the UK’s infamous “productivity puzzle.” But the conclusion was the same either way: the economy was running well below capacity. You cannot generate excess demand inflation without excess demand.
But here’s what the simple story misses
There’s a tempting narrative: QE was tried, inflation didn’t follow, therefore QE failed. It’s neat. It’s wrong.
The real question isn’t “did QE cause inflation?” It’s “what would have happened without it?” And that counterfactual is genuinely difficult to construct.
The best academic estimates suggest QE wasn’t doing anything. Work by Joyce, Tong and Woods (2011) at the Bank of England estimated that the first round of QE alone raised £200 billion of purchases, raised GDP by between 1.5% and 2%, and pushed inflation up by 0.75 to 1.5 percentage points relative to what it would otherwise have been. These are not trivial numbers. Without QE, the UK may well have experienced a deflationary spiral of the kind that caused so much damage in Japan during its Lost Decade, where prices fell for years, consumers delayed spending waiting for even lower prices, and the economy stagnated in a self-reinforcing trap.
The argument isn’t that QE was pointless. It’s that QE, operating alone, couldn’t generate excess demand inflation when structural forces were simultaneously suppressing demand. It stabilised the economy. It didn’t transform it.
And notably, the Bank kept expanding the programme precisely because it wasn’t generating enough inflation - QE1 (£200bn, 2009), QE2 (£175bn, 2011–12), QE3 (£70bn, post-Brexit 2016), and finally a massive Covid-era expansion (£450bn, 2020). Each round was a response to continued weakness. The escalation itself tells you something.
Why did the transmission break down
Two types of money and why that matters
Before examining each transmission failure, there’s a mechanical fact about modern banking that most QE commentary skips over and without it, the puzzle doesn’t completely resolve.
When the Bank of England buys gilts, it creates central bank reserves - digital balances held exclusively by commercial banks in their accounts at the Bank. These reserves settle transactions between banks. They cannot, by design, leave the interbank system. They are not the money in your current account. They are not the money businesses use to pay wages or invest in equipment.
Commercial bank deposits - the money that actually circulates in the real economy are an entirely separate circuit. The only way QE reserves become real-economy money is if commercial banks lend them out, thereby creating new deposits. That requires willing lenders and willing borrowers. As the following sections show, post-2008 Britain had neither.
"The pipes were full. The water wasn't flowing."
This distinction, financial money versus real-economy money, is the mechanical reason why the Bank could expand its balance sheet to £895 billion and still watch consumer prices barely move. The pipes were full. The water wasn’t flowing.
The wealth effect was real, but it went to the wrong people.
QE did push up asset prices. Bond yields fell. Equity markets recovered. Property values rose. But the consumption boost that was supposed to follow from all of this was far weaker than expected.
The reason is straightforward: financial assets are heavily concentrated among wealthier households, pension funds, and institutional investors, precisely the people with the lowest marginal propensity to consume. They don’t spend much of an extra pound of wealth. So, while QE made asset owners’ balance sheets look healthier, it didn’t translate into the broad-based spending surge needed to drive inflation.
Ben Broadbent, then Deputy Governor of the Bank of England, put it plainly in 2015: QE’s main effect was to accelerate a fall in real interest rates that would have happened anyway, given weak economic conditions, rather than essentially altering the dynamics of demand.
Banks weren’t lending - and when they did, it went to the wrong places
The numbers are stark. Total outstanding lending to UK residents grew by a mere 5% in nominal terms over the entire post-crisis decade, from approximately £2.07 trillion in 2008 to £2.2 trillion a decade later. In real, inflation-adjusted terms, that amounts to a significant contraction. Net lending to private non-financial corporations turned sharply negative by May 2009 as banks aggressively de-risked their balance sheets to meet tightening Basel III capital requirements.
But the deeper problem wasn’t just the volume of lending, it was where the lending went. Rather than channelling credit to the productive economy, UK banks quietly shifted their balance sheets toward residential mortgages, which carried lower risk weightings under the new regulatory frameworks. The proportion of bank lending allocated to residential mortgages jumped from 39.7% in December 2008 to 54% by the mid-2020s. Meanwhile, lending to non-financial companies fell from 11.6% of total lending in 2009 to 9.6%, and commercial real estate lending collapsed by 44.4%. The ratio of productive real-economy lending relative to mortgage lending dropped from 0.28 to 0.17 post-crisis.
QE flooded the system with reserves. The reserves went into mortgages and financial assets, not into business investment and wages.
The consequence was clear in the money supply data. Before the crisis, broad money (M4) was growing at 10–15% annually. Post-2008, M4 growth collapsed into negative territory by 2011–2012, and the more meaningful M4ex measure, which strips out intra-financial sector noise to show what was actually happening in the real economy, fell to just 1–2% annually between 2009 and 2014. QE was injecting money at one end; bank deleveraging was destroying it at the other.
The velocity of money - how many times each pound circulates through the economy- collapsed by an estimated 20–25% between 2008 and 2015. You can expand the money supply dramatically, but if each pound sits on a bank balance sheet or gets recycled into financial assets rather than changing hands in the real economy, prices don’t move. The Bank of England found itself in a classic liquidity trap: vast expansions of the monetary base failing to generate proportional increases in aggregate demand.
On the household side, the spike in the savings rate confirmed the same dynamic. Many households were already over-leveraged from the pre-crisis years. The rational response to uncertainty wasn’t to take on new debt - it was to pay down existing debt and save more. The credit channel, which required willing lenders and willing borrowers, found neither.
Fiscal policy was pulling in the opposite direction
This is arguably the most important factor, and the one that gets the least attention.
While the Bank of England was trying to stimulate demand through QE, the Treasury was doing the opposite. The coalition government’s austerity programme, launched in 2010, cut public spending and restrained public-sector wages at exactly the moment when monetary policy was trying to boost aggregate demand. The Office for Budget Responsibility confirmed that fiscal policy was contractionary throughout much of this period.
“QE and austerity were working against each other.”
Monetary policy was pushing on one end of the rope; fiscal policy was pulling on the other. The result was a kind of macroeconomic stalemate, neither deflation nor the inflation the Bank was targeting.
The labour market had too much slack
Demand-driven inflation requires wage growth. And wage growth requires a tight labour market - one where employers compete for workers and employees have the bargaining power to demand higher pay.
The UK had neither. Unemployment fell gradually, but underemployment remained high. The rise of zero-hours contracts and part-time work meant that headline unemployment figures understated the true degree of slack. Meanwhile, the UK’s productivity growth, already weak before the crisis, essentially flatlined after it, dropping from a historical 2% annual trend to just 0.3% a year. Without productivity gains, firms couldn’t afford to pay more without squeezing profits.
The result, as Nicholas Crafts documented, was a labour market that consistently failed to generate the wage-price spiral required by persistent demand-driven inflation.
Was this a UK problem, or did QE fail everywhere?
It’s worth asking whether the UK’s experience was unusual. It wasn’t.
The Federal Reserve ran three rounds of QE between 2008 and 2015, and counterintuitively, US inflation was actually lower than the UK’s for most of that period, despite both central banks expanding their balance sheets aggressively. US headline inflation hovered around 1.5–2%, while UK CPI repeatedly breached the target, peaking near 5.2% in 2011. But the UK’s higher inflation was driven entirely by supply-side factors - a 25% depreciation of sterling, which drove up import costs, and VAT hikes from 15% to 17.5% (2010) and then to 20% (2011). Strip those out, and the underlying demand pressures in both economies were similarly weak.
In the US, the Fed compounded the transmission problem by paying interest on excess reserves starting in October 2008, effectively giving banks a risk-free incentive to hoard liquidity at the Fed rather than lend it out. The money multiplier collapsed. The pattern was the same on both sides of the Atlantic: reserves created, reserves hoarded, real economy unaffected.
The ECB arrived even later, launching its broad-based asset purchase programme only in March 2015, years after the Fed and BoE. It successfully pulled the Eurozone out of its deflationary spiral, with headline inflation recovering from -0.6% in early 2015 to around 1.5–2% by 2017–18. But core inflation remained stuck around 1% for years, and the moment the ECB tried to taper purchases in late 2018, growth decelerated, forcing a restart. The pattern held: QE as stabiliser, not demand generator.
But the most powerful comparison isn’t the US or the Eurozone. It’s the UK itself in 2020.
When the pandemic hit, something fundamentally different happened. Rather than relying on banks to transmit central bank reserves into the real economy, the government bypassed the banking system entirely. Furlough payments, business grants, and direct support transfers created new commercial bank deposits, real-economy money straight into household and business accounts. Broad money growth surged. And when that extra spending power collided with supply-chain disruptions and an energy shock, inflation followed rapidly.
The contrast is stark. A decade of QE operating through financial intermediaries produced almost no demand-driven inflation. Months of direct fiscal transfers produced the fastest inflation surge in a generation.
Same country. Same central bank. Completely different transmission mechanism. Completely different outcome.
The lesson is not subtle: it matters enormously whether new money enters the real economy or stays trapped in the financial circuit.
The cross-country evidence points to the same conclusion: QE is highly effective at preventing financial crises from becoming deflationary spirals. It is much less effective at generating the self-sustaining demand interactions that produce persistent inflation.
The expectations paradox
One of the more intellectually interesting aspects of this story concerns inflation expectations and contains a genuine irony.
Part of QE’s purpose was to keep inflation expectations anchored at 2%, preventing deflationary psychology from taking hold. On this measure, it succeeded. Survey and market-based measures of inflation expectations remained broadly stable throughout the decade.
But here’s the paradox: that very credibility also helped explain why QE couldn’t generate persistent inflation. When households and businesses believe the Bank of England will keep inflation around 2%, they set wages and prices accordingly. There’s no self-fulfilling acceleration. The anchoring that made the monetary regime credible also made demand-driven inflation self-limiting.
This isn’t a failure of QE exactly; it’s more a reflection of what successful inflation targeting actually looks like. The Bank wasn’t trying to create runaway inflation. It was trying to prevent deflation while nudging the economy back toward target. The stability of expectations shows it largely succeeded on the first count, even if the second proved harder.
What to take from all of this
The UK experience between 2009 and 2020 and the parallel US and Eurozone stories tell us something important about the limits of monetary policy operating alone.
QE worked as a financial stabiliser. It prevented yields from spiralling, kept credit markets functioning, and almost certainly cushioned the recession’s depth. The counterfactual without it looks considerably worse.
But stabilisation is not transformation. QE couldn’t force money to circulate, couldn’t compel banks to lend productively, couldn’t override a fiscal policy pulling in the opposite direction, and couldn’t generate the wage growth that demand-driven inflation ultimately requires. The velocity of money collapsed. The credit went to mortgages, not businesses. And the economy closed its output gap not through triumphant recovery, but by accepting a permanently smaller version of itself.
The reckoning is still unfolding. As of 2026, the Bank of England’s attempt to unwind its QE holdings through aggressive quantitative tightening has cost taxpayers an estimated £125 billion - with markets adding roughly 70 basis points to UK borrowing costs in what analysts now call the “Bailey premium.” The experiment that failed to generate inflation between 2009 and 2020 is now generating fiscal pain in ways its architects never anticipated. Understanding why the transmission broke down in the first place is not just a historical question. It is essential context for every subsequent monetary policy decision.
Data sources: ONS Consumer Prices Index (D7G7); ONS Average Weekly Earnings (KAC2); ONS Household Saving Ratio (DGD8); OBR Economic and Fiscal Outlooks; Bank of England Asset Purchase Facility (RPQBVEUA); Bank of England M4ex data; Bank of England APF Quarterly Report (Q1 2026); Joyce, Tong and Woods (2011), Bank of England Quarterly Bulletin; Broadbent (2015), Society of Business Economists speech; Crafts (2017), National Institute Economic Review; Positive Money (bank lending composition data).
Full code and data available on [GitHub] · Connect on [LinkedIn]
This is the second post in a series using empirical data and structural analysis to dissect monetary policy, public finance, and sovereign debt dynamics. Subscribe below to stay updated.
Really interesting peice! One of the clearest explanations I've read of why outcomes can differ so dramatically even when the headline policy looks the same.