Inflation
in a Hurry

The Whole Thing in One Page

Inflation looks like a problem of expensive things. It is more slippery than that. A drought can make olive oil dearer. A refinery outage can lift petrol. A fashionable neighbourhood can push up rents. Those are price changes. Inflation is a sustained rise in the general price level, which means the unit in which wages, savings, debts, taxes and contracts are written buys less across the economy.

That distinction determines what can fix it. If one product is scarce, its higher price is part of the economy's message that supply is tight. If prices across many categories keep rising, the problem is broader. Nominal spending may be pressing against the economy's capacity to produce. A supply shock may be passing into wages, margins and other prices. Expectations may be changing how firms set prices and how workers bargain. Several forces can operate at once.

The inflation number is built rather than discovered. Statistical agencies collect huge numbers of prices, construct a representative basket and weight categories by household spending. They then adjust for changing products, substitutions and housing treatment. The result is useful, but it is an average. A renter, a heavily mortgaged homeowner, a pensioner and a family with three hungry teenagers can inhabit the same economy and feel different inflation.

The most important distinction is between the price level and the inflation rate. If a basket rises from £100 to £110, then inflation later falls to 2 per cent, the basket does not return to £102. It becomes roughly £112.20. Disinflation means prices are rising more slowly. Deflation means the general price level is falling.

Inflation can begin on the demand side or the supply side. Too much nominal spending relative to available labour, machines, housing, energy and imports creates pressure. So does a loss of supply, such as an energy shock or currency depreciation. The policy problem is to stop temporary shocks becoming a continuing process without causing more damage than necessary.

That is why central banks raise interest rates. They cannot produce gas, wheat or microchips. They can make borrowing dearer, reward saving, weaken demand and reduce firms' ability to raise prices without losing customers. The cost is part of the mechanism. Disinflation often means slower spending, weaker hiring and sometimes recession.

Inflation also redistributes. Unexpected inflation can reduce the real burden of fixed nominal debt, hurt creditors and cash holders, squeeze workers whose pay resets slowly, and hit poorer households harder when food and energy lead the rise. There is no single winner or loser. Contracts, assets, taxes and bargaining power decide the distribution.

The practical rule is diagnosis before prescription. Ask what moved first, how broad the rise is, whether the process is fading or propagating, what nominal demand is doing, where capacity is constrained, how wages and margins are responding, and what expectations imply. Then match the tool to the mechanism. The aim is not frozen prices. It is a stable enough measuring stick that economic decisions can once again mean what their numbers appear to mean.

That is the book.

Why You Should Care

In October 2022, UK CPI inflation reached 11.1 per cent. A worker could receive a 7 per cent pay rise and become poorer. A saver could earn interest and still lose purchasing power. A company could report higher revenue while selling fewer goods. A borrower with a long fixed-rate loan could repay the promised pounds while the real burden of those pounds shrank. Inflation creates a gap between the number on the page and the economic reality underneath it.

That matters because modern economies are webs of nominal promises. Salaries are written in pounds. Rents are agreed in pounds. Taxes, pensions, bonds, invoices and mortgages are stated in money units. A stable unit lets people compare choices across time. Rapid or unpredictable inflation changes the real content of those promises after they were made.

Inflation also forces a political question that arithmetic cannot answer. Suppose imported energy becomes permanently dearer. The country has lost real purchasing power. Monetary policy cannot undo that physical loss. Someone must bear it through lower real wages, lower profits, lower consumption, higher taxes, reduced public services or some mixture. If every group tries to restore its previous real claim at once, the struggle over distribution can help keep domestic prices rising after the original energy shock has faded.

The subtitle asks what to do. That wording is more demanding than it looks because inflation has no universal treatment. If excessive demand is the main force, restraining demand can work. If the problem is a temporary oil shock, suppressing demand cannot create oil. If expectations have become unanchored, policy credibility matters. If a narrow supply bottleneck is binding, investment or regulatory change may help, but usually too slowly for the next few quarters. If fiscal support is needed, targeting matters because broad support can protect incomes while also sustaining aggregate demand.

Both action and inaction impose costs. Let high inflation persist and contracts become harder to write, relative-price signals become noisier, and redistribution becomes arbitrary. Tighten too aggressively and firms can fail, mortgages reset at painful rates, investment collapses and unemployment rises. The central bank is not choosing between pain and no pain. It is choosing among paths that distribute pain differently through time.

Low and stable inflation is therefore an institutional achievement that is easiest to overlook when it works. It gives a five-year contract a chance of meaning roughly what both sides thought it meant. It gives a business owner a chance to interpret a rise in their own price as information about their product rather than merely the general drift of the currency unit.

The useful habit is to replace the word inflation with a chain of questions. Which prices moved? Why? How broad is the change? What is happening to spending and capacity? Are wages, margins and expectations spreading the shock? Which tool can affect that mechanism, and who will bear the cost? Once those questions are separated, the argument becomes much harder to reduce to a slogan and much easier to think about.

It also changes how you read everyday numbers. A mortgage rate is partly a price for waiting and risk, but its real burden depends on inflation. A business's 10 per cent revenue growth can conceal falling real output. A government can collect more tax in cash terms without gaining much more command over labour, medicines or concrete. Inflation is therefore less a specialised topic than a warning attached to every nominal number. Learn to strip out the movement of the unit and a large amount of economic fog disappears.

The Core Ideas

Inflation Is a Change in the Measuring Stick

A price is a ratio. A £4 coffee says that one coffee exchanges for four pounds. If coffee rises to £4.40 while most other prices are unchanged, coffee has become expensive relative to other things. Perhaps beans were scarce, rent rose on that street, or customers became willing to pay more. That relative-price movement contains information.

Inflation is broader. If coffee, haircuts, rail fares, restaurant meals, insurance and thousands of other purchases tend to cost more pounds, the pound buys less across the basket. The common measuring unit has shifted.

This distinction matters because an economy needs relative prices to move. If a poor harvest makes wheat scarce, a higher wheat price tells consumers to economise and producers that extra supply is valuable. Trying to keep every individual price fixed would obstruct that adjustment. Price stability does not mean a world in which nothing gets dearer or cheaper.

Three distinctions do most of the intellectual work.

First, the price level is different from the inflation rate. Suppose an index starts at 100 and rises to 110 in a year. That is 10 per cent inflation. If it then rises to 112.2, inflation has fallen to 2 per cent, but the price level remains 12.2 per cent above where it began. This is why households can hear that inflation has fallen and still feel no return to the old cost of living. Their complaint is about the level. The central bank's target concerns the rate of change.

Second, disinflation is not deflation. Disinflation means the inflation rate is falling while prices still rise. Deflation means the general price level is falling. Cheap computers or falling petrol prices can coexist with positive overall inflation. A single falling category does not make deflation any more than one expensive category makes inflation.

Third, nominal values differ from real values. A nominal wage is the pounds on a payslip. A real wage is what those pounds buy. A nominal interest rate is written into a contract. The real return depends on inflation. A 5 per cent pay rise during 7 per cent inflation is a pay rise in pounds and a pay cut in purchasing power.

These distinctions become important because economies are full of contracts fixed in nominal terms. Wages may reset once a year. Rents may be fixed for a tenancy. Bonds promise nominal coupons. Mortgages can have long fixed periods. Tax thresholds may remain unchanged while cash incomes rise. Inflation changes the real value of each contract according to how quickly it can adjust.

Expected inflation is less disruptive than surprise inflation because people can build it into decisions. A lender who expects 3 per cent inflation can demand a higher nominal interest rate. A worker can bargain over expected real pay. A firm can price a long contract accordingly. None of this makes inflation costless, but predictability reduces arbitrary redistribution.

Why, then, do many central banks aim for a small positive rate rather than zero? The answer combines several practical considerations. Some wages and prices are easier to raise than cut in nominal terms. Inflation measurement is imperfect. A positive target gives nominal interest rates more room above zero in normal times, allowing central banks to reduce real rates during downturns. The exact target is an institutional choice, not a law of nature, but many advanced economies have converged on about 2 per cent.

The best mental picture is a ruler. If the ruler's length changes slowly and predictably, people can compensate. If it changes quickly, unpredictably and differently from what contracts assumed, every measurement written with it becomes less trustworthy. Core Idea 7 will return to the same problem from the other end: stabilising the ruler means forcing some existing spending plans and nominal claims to adjust.

The Inflation Number Is Constructed

No agency can walk into a warehouse labelled "the economy" and read inflation from a dial. A consumer-price index is an engineered measurement system.

Statistical agencies choose a basket of goods and services, collect prices, group them into categories and apply expenditure weights. A category on which households spend a great deal matters more to the index than one on which they spend little. The weights are updated because consumption changes. New products appear. Old ones disappear. Shopping shifts from physical stores to online sellers. Energy use changes. Housing costs evolve.

The basic arithmetic is easy. The difficult part is defining like for like. A television may cost the same as five years ago while having a larger screen, better resolution and lower energy use. Treating the same nominal price as zero inflation would ignore the quality improvement. Statistical offices therefore make quality adjustments where possible. Those adjustments are necessary and inevitably involve judgement.

Substitution poses another problem. If beef becomes expensive, households may buy more chicken. A basket that never changes can overstate the cost of maintaining consumption because it assumes people keep buying the old quantities. Yet rapid substitution can also conceal a loss of welfare. A family forced from a preferred product to a cheaper one has adapted successfully, but that does not mean nothing was lost.

Housing is one of the hardest cases. Owner-occupiers consume housing services while also owning an asset. Mortgage interest is a financing cost. House prices are asset prices. Different indices answer different questions by treating these elements differently. In the United Kingdom, CPI is the measure used for the Bank of England's target. CPIH extends CPI by including an estimate of owner-occupiers' housing costs using rental equivalence and Council Tax. RPI remains embedded in some contracts but uses methods the ONS no longer recommends as its preferred measure.

The index is also not your household. National weights are based on broad expenditure patterns. A household spending a large share on heating and food can experience much higher inflation during an energy shock than one spending more on categories whose prices are stable. A commuter paying for fuel or rail travel feels a different basket from someone working at home. A renter and an outright homeowner have different exposures even before debt service is considered.

This creates a common political confusion. People say the official number is wrong because their bills rose faster. The statistical number can be correct for its defined average while the household's experience is also correct. The right response is to ask what the index measures and whether another distributional measure is needed.

Asset prices require another boundary. If house prices rise sharply, first-time buyers may become worse off even when consumer-price inflation is modest. Shares can boom with little movement in CPI. Consumer-price indices track the price of consumption under a defined methodology, not the price of every asset someone might want to own. A broader argument about affordability therefore cannot be settled by pointing to CPI alone.

As of the latest UK release available on 11 August 2026, CPI inflation was 2.6 per cent in the twelve months to June 2026 and CPIH inflation was 2.8 per cent. Those figures are useful current context, not eternal facts. The deeper lesson is methodological: every inflation number has an index, a population, a basket, a weighting method and a time horizon behind it.

When you hear that inflation is 3 per cent, translate the statement into a more precise one: the chosen weighted price index is about 3 per cent above its level a year earlier. Then ask which index and what that leaves out.

Demand Can Outrun Capacity

Imagine a city with enough chefs, tables and kitchens to serve one million restaurant meals a week at current prices. Now households suddenly try to buy 1.2 million. Restaurants can add shifts, hire people from other sectors, increase seating or improve productivity. At first, extra spending may produce more meals. As spare capacity runs out, restaurants start competing for the same chefs and premises. Wages and rents rise. Menus follow.

This is the core of demand-driven inflation. Nominal spending grows faster than the economy's ability to increase real output at existing prices.

Capacity is not a fixed ceiling. Firms invest. People enter the labour force. Productivity changes. Imports can expand. Workers can move between sectors. The difficulty for policymakers is that sustainable capacity is estimated rather than observed. Economists talk about an output gap, meaning the difference between actual output and an estimate of what the economy can produce without creating persistent inflation pressure. The estimate is uncertain and often revised after the event.

The same amount of extra spending therefore has different effects in different conditions. During a deep recession with unemployed workers and idle factories, stronger demand can raise output and employment with limited inflation. Near capacity, it is more likely to raise prices and wages. In a damaged economy whose supply has fallen, spending that once looked normal can become excessive relative to the new capacity.

Where does demand come from? Everywhere. Households can spend income or savings. Firms invest. Governments buy goods and services or transfer money that recipients spend. Banks create credit. Foreign buyers demand exports. Monetary policy influences borrowing costs, asset prices and expectations. The relevant object is aggregate nominal spending relative to real productive capacity, not one institution in isolation.

Money still matters. Sustained economy-wide inflation requires a nominal environment that allows total spending to keep rising relative to real output. Over long periods, a monetary and fiscal regime that continually validates higher nominal spending will show up in the price level. But the slogan "too much money" compresses several steps into one phrase. Modern money is largely bank deposits. Central banks set policy rates and supply settlement money. Governments tax, spend and borrow. Private banks lend. Households and firms decide whether to spend. The route from balance sheets to expenditure matters for both diagnosis and policy.

Fiscal policy can add demand pressure, but the effect depends on conditions. A transfer to unemployed households in a deep downturn can support spending when the economy has unused resources. The same transfer in an economy already facing scarce labour and supply bottlenecks can add more pressure to prices. Government investment can also expand future capacity, so the short-run demand effect and long-run supply effect need not point in the same direction.

The pandemic period made these interactions visible. Restrictions damaged supply and shifted spending from services towards goods. Governments in many countries replaced lost incomes at exceptional scale. Households accumulated savings. Monetary policy was highly supportive. When reopening arrived, demand recovered into ports, factories, housing markets and labour markets that could not instantly return to their old configuration. The result was neither a pure supply story nor a pure stimulus story.

This is why comparisons across countries matter. Countries had different fiscal packages, different energy dependence, different labour markets and different reopening paths. A theory that explains one country's inflation entirely through a single national policy should be treated with caution if similar inflation appeared elsewhere through different channels.

The practical implication is clean even when the causal weights are not. If nominal demand is persistently too strong relative to sustainable capacity, policy must cool it or raise capacity. The second route is usually slower. The first route works faster because spending can be reduced before new power stations, homes, workers or factories can be created.

That speed is also the source of the pain.

Supply Shocks Make the Economy Poorer Before They Make It Inflationary

Suppose a country imports much of its gas and the world price doubles. The country now has to give up more exports or financial claims to obtain the same quantity of energy. In real terms, it is poorer.

That is the fact policy cannot repeal.

The first effect is a relative-price change. Gas becomes dearer compared with other goods. Electricity, fertiliser, transport, glass, chemicals and heating may follow because energy is an input. Households have less income left for restaurants, clothes and entertainment. Firms face higher costs. The initial shock can raise the consumer-price index sharply even if domestic demand has not strengthened at all.

The distributional fight begins immediately. A firm can raise prices, accept a lower margin, cut other costs or reduce output. Workers can accept lower real wages or seek higher nominal pay. Government can subsidise bills, cut taxes or transfer income. Creditors and debtors can be affected through inflation and interest rates. Each choice changes who bears the loss.

What no policy can do is let every group preserve its previous real purchasing power when the economy has less real purchasing power to distribute. If workers fully restore real wages, firms may have to accept lower profits or raise prices. If firms fully protect margins, households bear more of the loss. If government reimburses everyone, taxes, borrowing or monetary conditions must adjust, and aggregate demand may remain stronger than the reduced supply can support.

This is why supply shocks create the hardest monetary-policy trade-off. A central bank can destroy enough demand to offset the inflationary effect of dearer energy, but it cannot create the missing gas. If the shock is temporary and expectations remain stable, an immediate attempt to force headline inflation back to target could require a large and unnecessary recession.

Central banks therefore distinguish first-round and second-round effects. The first-round effect is the direct rise in energy and related prices. The second-round question is whether the shock changes wage bargains, margins, expectations and broader price-setting so that inflation remains high after the original input stops rising.

Exchange rates can create a similar mechanism. A currency depreciation makes imported goods more expensive in domestic currency. The pass-through is rarely one for one. Firms may hedge currencies, absorb margins, price in dominant international currencies or face competition that limits increases. The effect also depends on how import-dependent the economy is. A country importing fuel, food and manufactured inputs has a different exposure from one producing more of those goods domestically.

Supply shocks also expose why "core" inflation is useful and dangerous. Measures that exclude volatile food and energy can help show whether price pressure is broadening into more persistent domestic categories. But food and energy are not peripheral to household welfare. A core measure can be useful for diagnosis while being a poor description of lived pain.

The recent European energy shock illustrates the distinction. Russia's invasion of Ukraine sharply altered energy markets and hit Europe harder than economies with less dependence on imported gas. Headline inflation rose dramatically. As energy prices later eased, attention shifted towards services, wages and domestic price-setting. The problem changed over time even though public debate often kept using the same word.

Supply policy can help, but it works on physical bottlenecks and usually on a slower clock. More power generation, better grids, housing reform, ports, training, storage and trade diversification can increase resilience and capacity. Those changes may lower the probability or impact of future inflation shocks. They are not substitutes for monetary restraint when inflation has already become broad and persistent.

The deepest lesson is political rather than technical. When supply falls, society is negotiating a real loss. Inflation can become the mechanism through which that negotiation occurs, especially when no group is willing to accept the first distribution of pain.

Expectations and Contracts Create Persistence

Inflation today can change inflation tomorrow because many prices are set for the future.

A restaurant prints menus that may last months. A worker agrees a salary for a year. A landlord sets rent for a tenancy. A lender writes a fixed interest rate. A supplier signs a contract. Each must form a view about future costs and prices.

If firms expect wages, energy and rents to rise rapidly, they may raise prices before every cost arrives. If workers expect consumer prices to rise, they bargain for higher nominal wages. If lenders expect inflation, they demand higher nominal interest rates. Expectations enter contracts, and those contracts then shape measured inflation.

This does not mean belief alone can create unlimited inflation. Firms still need customers. Workers need employers. Borrowers face budgets. Expectations work through institutions and spending constraints. Their role is to change the speed, breadth and persistence with which shocks pass through the economy.

That is the logic of anchored expectations. Suppose inflation rises temporarily to 6 per cent after an energy shock, but households and firms believe the central bank will return it to 2 per cent. Wage contracts may rise less than 6 per cent. Long-term interest rates may remain closer to the target plus a normal real return. Firms may treat the shock as temporary rather than continually repricing for another round. Credibility limits propagation.

If that belief weakens, behaviour changes. Contracts may shorten. Indexation becomes more common. Wage demands rise. Firms review prices more frequently. Lenders seek protection. A temporary shock can then become a process embedded in the rules people use to protect themselves.

Indexation makes the mechanism easy to see. If a wage, pension or rent automatically increases with last year's inflation, the contract protects one side from erosion. It also carries some of yesterday's inflation into tomorrow's nominal payments. In economies with histories of high inflation, widespread indexation can become a rational defensive response that makes stabilisation harder.

The famous wage-price spiral is one possible form of persistence, not a universal law. Higher wages do not automatically produce higher inflation. If productivity rises by the same amount, unit labour costs may not rise. Firms can absorb higher wages through lower margins. An economy with spare capacity can increase output. Conversely, margins can widen during an inflation episode when firms have pricing power and demand remains strong.

This is why arguments over whether wages or profits "caused" recent inflation often become confused. Wages and profits are both claims on the value generated by production. An initial import-cost shock can reduce the real income available to domestic workers and firms. The subsequent path depends partly on how they divide that loss. IMF research on the euro area found that import prices and profits were large counterparts of price increases during part of 2022 and early 2023, while warning that this did not establish a general rise in markups across firms. Later wage growth could then restore some labour income without implying that workers initiated the original shock.

The Phillips curve belongs here too. In broad terms, tighter labour markets can put upward pressure on wage growth and inflation, but the relationship is neither fixed nor sufficient by itself. Expectations, productivity, supply shocks, labour-market institutions and the credibility of policy all matter. Milton Friedman's 1968 argument was important because it attacked the idea of a permanent menu in which policymakers could buy lower unemployment with predictably higher inflation forever. Once expectations adjust, the trade-off changes.

Expectations are difficult to measure. Surveys ask households and firms what they expect. Financial markets contain inflation compensation, but that also includes risk premiums and liquidity effects. Wage settlements reveal some expectations after negotiation. No single measure is decisive.

Policy therefore watches a bundle of evidence: services inflation, wage growth, pricing intentions, medium-term expectations, labour-market slack, margins and measures of underlying inflation. The question is not whether any one indicator is high. It is whether the network of contracts is beginning to behave as if high inflation is normal.

Once that happens, stopping inflation becomes more expensive because policy must change behaviour as well as the original price shock.

Inflation Redistributes Through Timing

Calling inflation a tax on everyone is tempting because it sounds simple and suitably unpleasant. It is also wrong in an important way. Inflation changes real claims unevenly.

Consider a fixed-rate borrower. A household owes £200,000 at a fixed nominal rate. Unexpected inflation pushes up wages and prices while the contractual debt remains £200,000. If the household's nominal income eventually rises, the real burden of the debt can fall. The lender receives the promised pounds, but those pounds buy less than expected. Surprise inflation transfers real value from the creditor towards the debtor relative to the original contract.

Expected inflation changes the deal. If lenders foresee higher inflation, they can demand higher nominal interest rates. This is why persistent inflation rarely offers governments or borrowers a free way to erase debt. Once the pattern is anticipated, future contracts become more expensive, shorter or indexed.

Cash and conventional fixed nominal bonds are exposed because their money values do not automatically rise with the price level. Equities and property may offer partial protection over long periods because profits and rents can adjust, but they are not guaranteed inflation hedges. Higher inflation often brings higher interest rates, which can depress asset valuations. The 1970s provide ample warning against the idea that any asset labelled "real" must protect purchasing power over every horizon.

Workers experience inflation according to bargaining power and reset timing. A worker whose pay is reviewed once a year may endure many months of lower real income before catching up. A worker in a scarce occupation may recover quickly. A pensioner with an indexed benefit faces a different path from one living on a fixed nominal annuity.

Households also have different spending shares. Poorer households often devote more of their budgets to necessities such as food and energy. When those categories lead inflation, the same national CPI rate can impose a larger welfare loss on them because there is less discretionary spending to cut. Yet it is unsafe to declare every inflation episode mechanically regressive. Debt, benefits, wages, taxes and asset ownership can reverse parts of the distribution.

Tax systems add another channel. If nominal wages rise while tax thresholds stay fixed, more income can enter higher tax bands without a corresponding real improvement. This is sometimes called fiscal drag. Indexing thresholds can reduce that effect, but it also changes government revenue and therefore the fiscal stance.

Firms divide too. A company with long fixed-price sales contracts and rapidly rising input costs may be crushed. Another with flexible pricing and weak competition may protect or increase margins. A third may lose customers if it tries. Talking about "business" as one side of the inflation conflict hides these differences.

Governments can gain from unexpected inflation if the real value of fixed-rate nominal public debt falls. They can also lose because benefits, wages, procurement costs and future borrowing rates rise. Investors demand compensation once inflation risk is visible. A government that acquires a reputation for using surprise inflation to reduce debt may find the trick cannot be repeated cheaply.

At high and unstable rates, the costs go beyond redistribution. Firms spend time repricing and renegotiating. Households shorten planning horizons. Long-term lending becomes harder. Accounting systems can record nominal profits that are partly illusory. Tax can fall on gains that disappear after inflation adjustment. Relative-price signals become harder to read because every price is moving against a moving background.

Hyperinflation is the extreme case in which these problems destroy money's ability to work as a unit of account and store of value. It usually appears with severe fiscal and political breakdown, not as the inevitable next step after a few years of ordinary inflation. Weimar Germany is therefore useful for understanding monetary collapse and poor as a casual analogy for every uncomfortable CPI release.

The common thread is timing. Prices, wages, debts, taxes and benefits reset on different schedules. Inflation rewards some contracts and punishes others before the system catches up. The distributional question is therefore not "who does inflation hurt?" It is "which claims adjust first, which adjust last, and who has the power to rewrite the contract?"

Stopping Inflation Means Breaking the Spending Process

Return to the measuring stick. Persistent inflation means nominal spending, costs and contracts are evolving in a way that keeps the general price level rising faster than the chosen stability target. To stop that process, policy has to make nominal demand grow more slowly relative to sustainable real output, improve supply, change expectations, or some combination.

The fastest general tool is usually monetary policy.

A central bank raises its policy interest rate. Money-market rates respond. New mortgages and business loans become dearer. Existing fixed-rate borrowers feel the change when they refinance. Saving becomes more attractive for some households. Asset valuations can fall because future cash flows are discounted at higher rates. Exchange rates may strengthen, reducing imported-price pressure, though currencies respond to many forces beyond policy rates.

The next stage is spending. Some households postpone large purchases. Some indebted households have less disposable income. Firms cancel projects whose expected returns no longer clear the financing cost. Hiring slows. Demand weakens relative to capacity. Firms find it harder to raise prices without losing customers. Labour markets cool. Wage growth may ease. Inflation pressure falls.

This process is uneven and delayed. A household with a five-year fixed mortgage does not react like one refinancing next month. A cash-rich company does not react like a leveraged property developer. Service sectors respond differently from manufacturing. Exchange rates can move quickly while employment moves slowly. Monetary policy therefore acts with long and uncertain lags.

The Bank of England's current position illustrates the forecasting problem without providing a timeless template. At its meeting ending 29 July 2026, the Monetary Policy Committee kept Bank Rate at 3.75 per cent. The latest available ONS release showed June CPI inflation at 2.6 per cent. Policy was therefore being set with inflation above the 2 per cent target but far below the 2022 peak, amid continuing uncertainty over domestic persistence and energy. The relevant lesson is not the particular rate. It is that central banks set policy for the future path of inflation rather than mechanically matching today's CPI number.

Supply shocks complicate this because interest rates cannot restore lost supply. If energy prices rise, suppressing enough demand can still prevent the shock spreading, but it does so by making the rest of the economy weaker. The central bank is choosing how much real activity to sacrifice now to reduce the risk of more persistent inflation later.

Fiscal policy can make the job easier or harder. Broad tax cuts or transfers can sustain demand while monetary policy is trying to reduce it. Fiscal contraction can cool demand, though composition matters because cutting productive investment may reduce future capacity. Targeted support can protect households facing severe energy or food shocks with less demand leakage than universal support, especially when designed around need rather than total consumption.

Price controls operate differently. A legal ceiling can stop the market price from rising, but it does not remove scarcity. If demand exceeds available supply at the controlled price, allocation shifts to queues, rationing, waiting lists, lower quality, informal markets or government subsidy. Controls can be sensible in narrow emergencies or regulated monopoly settings. They cannot by themselves reconcile economy-wide nominal spending with scarce real resources.

Supply reform is the least painful cure when it is feasible and timely. More energy, housing, transport capacity, workers, competition and productivity can allow demand to be met with more output rather than higher prices. The problem is timing. Building a grid, training nurses or reforming planning can take years. Inflation expectations and contracts can change in months.

This returns us to Core Idea 1. The economy began with nominal promises written against a changing measuring stick. Stabilisation means changing the conditions under which those promises are fulfilled. Some borrowers refinance at higher rates. Some firms lose demand. Some workers find jobs harder to obtain. Some public spending is restrained. The mechanism works because someone spends less than they otherwise would.

That does not mean recession is always required. If supply recovers quickly, expectations remain anchored and policy acts early enough, inflation can fall with limited output loss. A "soft landing" is possible. It is difficult because policymakers are steering with uncertain data, delayed effects and a moving estimate of capacity.

The essential policy question is therefore not whether a government or central bank is "tough on inflation". It is whether the instrument chosen can interrupt the mechanism that is keeping inflation alive, at a cost proportionate to the problem.

How It Actually Works

Inflation becomes easier to understand when followed through one episode from the first shock to the policy response.

A shock enters the economy

Imagine a country that imports natural gas. A geopolitical disruption drives wholesale gas prices sharply higher. Energy retailers face higher replacement costs. Households see bills rise as fixed deals expire or regulated caps reset. Factories using gas pay more. Electricity prices may rise if gas-fired power plants influence the marginal cost of generation. Fertiliser, transport, glass, chemicals and food processing all feel the shock.

At this stage, the economy has mainly experienced a relative-price change and a real-income loss. Gas is scarcer or dearer from abroad. The country must surrender more purchasing power to obtain it. Calling this inflation is statistically correct if it lifts the consumer index, but the policy diagnosis should begin with the physical shock.

The statistical office records it

Price collectors and data systems observe the new tariffs and prices. Each category enters the index with a weight. Energy's direct weight may be modest compared with the whole basket, but indirect effects spread through other goods and services.

The annual inflation rate compares today's index with its level twelve months earlier. This creates base effects. If gas prices jump in January and then remain flat, they can make annual inflation high for much of the year. Once the comparison month catches up with the jump, their contribution to annual inflation falls even though the gas price itself remains high.

This is why inflation can fall rapidly without anyone seeing lower bills. The rate has changed because the comparison base changed. The level remains elevated.

Households decide where the loss goes

A household whose energy bill rises by £150 a month has choices, none attractive. It can reduce heating, cut restaurant meals, save less, draw down savings, borrow, work more, seek higher wages or rely on government support.

These responses matter for aggregate demand. If the household spends £150 more on energy and £150 less elsewhere, total nominal spending may not rise. The shock reallocates spending. If government reimburses the full energy increase without offsetting taxes or spending cuts, the household can keep buying other goods too. That protects living standards but also preserves aggregate demand despite the economy's reduced real purchasing power.

Different households respond differently. A high-income household may absorb the shock through lower saving. A low-income household may have no saving margin and cut food or heating. This is one reason distribution is part of macroeconomics rather than a separate afterthought.

Firms choose prices, margins, quantity and quality

A restaurant now pays more for electricity, ingredients and deliveries. It can raise menu prices, accept lower profit, alter portion sizes, change ingredients, reduce opening hours, invest in efficiency or cut staffing.

Competition shapes the choice. If every restaurant faces the same energy shock and customers have few alternatives, more of the cost may be passed through. If one restaurant alone has high costs, a price increase may lose customers. Strong demand can also make pass-through easier because customers are less price-sensitive when tables are full.

This is where the debate about profits enters. A rising profit share during an inflation episode can mean firms passed through more than their own unit costs, that sectors with high profits gained weight, that wages adjusted more slowly, or several other accounting changes. It is evidence about distribution and pricing, not automatic proof of one causal story.

Workers bargain after prices have moved

Workers see the cost of living rise and seek higher nominal pay. A 6 per cent wage settlement may be catch-up after an 8 per cent price rise rather than the cause of the original inflation. Yet once higher wages are embedded in firms' costs, they can contribute to future price increases if productivity does not rise and margins do not fall.

The timing therefore matters. The first energy shock may come from abroad. The persistence may later become domestic through wages, services prices and expectations. By then the economy is dealing with a different problem from the one that started the episode.

Wage data are also easy to misuse. Average pay can rise because high-paid sectors expand or low-paid workers leave. Bonuses can distort monthly figures. Settlements differ across industries. Policymakers care about broad labour-cost pressure rather than one noisy number.

Expectations decide whether the shock is treated as temporary

Suppose firms and workers believe inflation will return near 2 per cent within a few years. A worker may ask for partial catch-up rather than indexing wages fully to current inflation. A firm may avoid a large pre-emptive price increase if it expects input costs to stabilise. Lenders may keep long-term rates closer to normal.

If nobody believes the target, defensive behaviour becomes stronger. Contracts shorten. Indexation spreads. Price reviews become more frequent. The economy starts carrying inflation forward through its own institutions.

Central-bank communication matters because policy works partly through those beliefs. Credibility cannot be declared into existence. It is accumulated through a history of decisions that convince people the target will guide policy even when doing so is uncomfortable.

Policymakers test for broadening

A central bank does not ask whether inflation is high. It asks why it is high and whether the forces are likely to persist.

Officials examine services inflation because services often depend heavily on domestic wages and adjust slowly. They examine pay growth, labour-market slack, surveys of expectations, business pricing intentions and measures that exclude volatile categories. They compare current output with estimates of capacity. They look at credit, fiscal policy, exchange rates and global commodity prices.

Every measure is imperfect. Core inflation can fall while food bills remain painful. Wage growth can be legitimate catch-up. Survey expectations can jump with petrol prices. Potential output is unobservable. The task is inference from several partial gauges.

Interest rates transmit through balance sheets

Suppose the bank raises its policy rate. The announcement can change market rates almost at once. Some asset prices and exchange rates move quickly. The effect on household cash flow arrives later.

A new mortgage becomes more expensive. A fixed-rate borrower feels nothing until refinancing. A landlord with floating debt may face an immediate cost increase. A firm considering a factory finds the hurdle rate higher. A saver receives more interest and may spend some of it, which is one reason monetary transmission is not uniform.

The dominant intended effect is weaker aggregate demand. Fewer marginal investments go ahead. Interest-sensitive purchases are postponed. Hiring slows. Some households cut consumption to service debt. Firms meet more resistance to price increases. Over time, wage and price pressure eases.

A 2024 Bank of England review of monetary transmission emphasised that policy rates affect financial conditions, expectations, activity and inflation through several channels rather than one direct lever. That description matters because public debate often treats a rate rise as though the central bank is choosing a new inflation number. It is choosing conditions that may alter future spending.

The mortgage channel makes the delay visible in Britain. When many households hold fixed-rate mortgages, a rise in Bank Rate does not hit every borrower at once. It rolls through the economy as fixed deals expire. That can extend the tightening effect long after the central bank stops raising rates. The reverse is also true when rates fall. Transmission depends on the stock of contracts already written, which is why identical policy moves can produce different effects across countries and periods.

The 1970s show how persistence raises the eventual cost

The US Great Inflation, conventionally dated from 1965 to 1982, combined several forces: expansionary demand, repeated policy accommodation, the breakdown of the Bretton Woods framework, oil shocks in 1973-74 and 1978-79, and changing expectations. No single element explains the whole period.

By the late 1970s, inflation had become persistent enough that Paul Volcker's Federal Reserve chose severe restraint. The resulting 1981-82 recession was deep. Inflation fell, but disinflation imposed large costs in unemployment and lost output.

The historical lesson is not that every modern inflation requires Volcker-style treatment. It is that allowing inflation to become embedded in contracts and expectations can make eventual stabilisation more expensive. Credibility is valuable because it can reduce the amount of unemployment needed to convince people that the inflation process will not continue indefinitely.

Britain's 1970s experience carried different institutions and policy choices but a similar warning. Oil shocks, wage bargaining, fiscal strain and policy experimentation interacted. Attempts at direct wage and price restraint could alter settlements for a time, yet they could not permanently solve the conflict between nominal claims and real capacity.

Hyperinflation shows the fiscal limit

At the extreme, inflation can accelerate because the state cannot finance spending through sustainable taxation or borrowing and repeatedly relies on money creation while confidence collapses. People try to hold money for shorter periods. Prices rise faster. Tax revenue loses real value between assessment and collection. The fiscal position can deteriorate further.

This feedback appeared in several famous hyperinflations, including Germany in 1923. But the conditions were extreme. Hyperinflation is not the normal destination of a credible central bank missing a 2 per cent target for a few years. Its value as an example is to show the real-resource and fiscal limit of monetary promises when institutions break down.

The pandemic shock shows why one-cause stories fail

The inflation of 2021-23 began from a different environment. Pandemic restrictions disrupted factories, ports and services. Consumers shifted spending towards goods. Governments replaced lost incomes and supported firms. Monetary policy remained highly accommodative. Household savings rose. Reopening released demand into supply chains and labour markets that had not returned to their previous shape.

Then Russia's invasion of Ukraine produced a major energy shock, especially in Europe. Commodity and shipping pressures later eased, but domestic services prices and wages adjusted more slowly. Firms' margins also changed, with IMF work showing a significant profit counterpart in euro-area inflation during part of the episode without establishing a universal increase in markups.

Countries differed. The United States experienced particularly strong domestic demand support and labour-market tightness alongside supply constraints. Europe faced a larger imported-energy shock. The United Kingdom combined imported energy exposure, labour-market changes, fiscal interventions and domestic persistence. A good model has to explain why the mix differed while the headline phenomenon looked similar.

Central banks tightened after inflation had already risen sharply. That timing became a major source of criticism. Some officials had expected inflation to be more temporary than it proved. Yet retrospective certainty is cheap. The more useful question is which signals were available, how quickly the balance of risks changed, and how policy frameworks should respond when a supply shock and strong demand arrive together.

Disinflation leaves the price level behind

Suppose the basket was £100, rose to £111 during the inflation surge, then inflation returned to 2 per cent. A year later the basket is roughly £113.22. The crisis in the inflation rate can be over while the cost-of-living level remains permanently higher.

Real incomes recover only if nominal wages, benefits or other incomes subsequently rise faster than prices, or if taxes and debt service ease. That catch-up can take years and differs across households. This is why the social consequences of inflation outlast the headline statistic.

The central bank is not normally trying to reverse the whole price-level increase. Doing so would require a period of deflation, which could raise real debt burdens and damage demand. Most inflation-targeting regimes instead aim to stabilise the future rate of increase.

A policy meeting is a forecast under uncertainty

By the time policymakers meet, much of the inflation they observe was caused by decisions made months earlier. The latest CPI release is backward-looking. Wage settlements span different periods. Mortgage refinancing schedules spread the effect of old rate rises into the future. Commodity prices can reverse between one forecast round and the next. Fiscal policy may change after a Budget.

The committee therefore has to decide where inflation is going rather than where it has been. A simple reaction to the latest headline number would be dangerous. If energy prices have already fallen but services inflation is persistent, the headline may overstate progress. If domestic demand is weak and the annual rate is high only because of an old base effect, further tightening may arrive after the need has passed.

Forecast errors are unavoidable because the relevant quantities are partly hidden. Potential output is estimated. The neutral interest rate is uncertain. Expectations are measured imperfectly. The effect of a given Bank Rate depends on how much debt is fixed, how banks price loans, how firms finance investment and how households respond. Good policy is therefore not policy that never forecasts wrongly. It is policy that updates when the mechanism changes and explains why the balance of risks has moved.

Fiscal and monetary policy can pull in opposite directions

Imagine the central bank raises rates to weaken demand while the government introduces a broad tax cut financed by borrowing. The two arms of policy are pushing against each other. The fiscal expansion supports household spending; the central bank may then need tighter financial conditions than otherwise to produce the same disinflation.

That does not mean fiscal deficits are automatically inflationary. During a recession, government support can prevent a collapse in demand and employment. Public investment can expand future capacity. The relevant question is the same one that runs through the book: what happens to total nominal demand relative to real supply, and when?

The interaction can become more serious if fiscal credibility breaks down and investors doubt the government's ability or willingness to stabilise debt without inflation. In advanced economies with credible institutions, that is a tail risk rather than the normal explanation for ordinary inflation. In severe historical episodes, however, the monetary and fiscal regimes cannot be separated. A central bank cannot maintain price stability indefinitely if the state requires it to validate unlimited nominal claims against limited real resources.

How we know

Consumer-price inflation is measured more directly than many macroeconomic concepts, but its causes are inferred rather than observed in one series. Statistical agencies measure prices and expenditure weights. National accounts connect prices with output, wages and profits. Labour data track pay and employment. Surveys and financial markets provide imperfect evidence about expectations. Policy institutions then combine timing, sector patterns, cross-country differences and economic models to judge causation.

That is why serious researchers can agree on the inflation rate while disagreeing about the share attributable to demand, supply, fiscal policy, monetary conditions, wages or margins. The disagreement matters when it changes the prescription. Current UK policy statements and inflation data used here were rechecked against the ONS and Bank of England through 11 August 2026. Historical claims were checked against Federal Reserve History and institutional and academic sources. The evidence is strongest on what happened and weaker on exact causal percentages.

What People Get Wrong

"Inflation means everything is getting more expensive"

Inflation is a rise in the general price level, not a requirement that every price rises. Some goods can become cheaper while overall inflation remains positive. A bad harvest can make coffee dearer without creating broad inflation. A technology improvement can make computing cheaper during a general inflation episode.

The mistake is persuasive because households encounter prices one at a time. The petrol station, supermarket and rent renewal are vivid. An index is abstract. Yet policy should not suppress every relative-price movement. Those movements tell people where scarcity and demand have changed.

The correction matters because different problems need different tools. If one crop fails, higher interest rates do not grow it. If demand is excessive across the economy, expanding one crop is irrelevant. The first diagnostic step is always to separate the relative price from the general price level.

"If inflation falls, prices should go back down"

A lower inflation rate means prices are rising more slowly. It does not mean the old price level returns.

The confusion is built into ordinary language. We say inflation "fell" and the natural image is of prices falling. In most inflation-targeting systems, policy aims to stop the rapid rate of increase, not reverse every previous rise.

If a basket rises from £100 to £110 and next year's inflation is 2 per cent, the new level is about £112.20. Returning the basket to £100 would require deflation. That can sometimes happen in particular sectors without difficulty, but broad deflation can raise real debt burdens and interact with weak demand.

This correction matters for judging policy. A central bank can succeed in bringing inflation back to target while households still need wage growth or tax changes to recover lost real income.

"Inflation is always caused by printing money"

The long-run connection between sustained inflation and the nominal monetary-fiscal environment is real. The phrase "printing money" is still a poor short-run model of modern inflation.

Most money used by households is bank deposits, not banknotes. Credit conditions, fiscal transfers, interest rates, supply shocks, exchange rates and expectations affect spending. A drought can raise food prices without any unusual money creation. The pandemic combined supported demand with impaired supply. Europe's energy shock had a large imported component.

The slogan becomes useful only after the missing chain is supplied: how did monetary and fiscal conditions affect nominal spending, and why did spending remain too high relative to real output? Without that chain, the phrase identifies a long-run constraint while hiding the episode's mechanism.

The distinction matters because policy must act through the route that is keeping inflation alive, not through a ceremonial attachment to one monetary aggregate.

"Corporate greed causes inflation"

Firms usually prefer higher profits in low-inflation years too. A sudden outbreak of greed cannot by itself explain why economy-wide inflation accelerates in one period and not another.

That does not make profits irrelevant. Firms with pricing power can widen margins when demand is strong or disruptions weaken competition. Profit shares can rise during an inflation episode. IMF analysis of the euro area found profits were a large counterpart of price increases during part of 2022 and early 2023, while cautioning against treating that accounting result as proof of a universal increase in markups.

The stronger model asks what gave firms room to raise prices, how costs moved, how competition changed, what happened to demand and how the resulting income was split between wages and profits.

The correction matters because "greed did it" and "profits did not matter" are both shortcuts. Pricing power can propagate and distribute a shock without being a complete theory of the price level.

A useful check is counterfactual: if firms were equally profit-seeking two years earlier, what changed to make larger price increases feasible now? The answer may be stronger demand, common cost shocks, disrupted competition, scarce capacity or expectations that customers will accept frequent repricing. That changed condition belongs in the causal story.

"Wage rises cause inflation"

Wages are a cost to firms and income to workers. Rapid wage growth can sustain services inflation when it outpaces productivity and firms pass higher labour costs into prices. Yet wages often rise after prices because workers are trying to recover purchasing power already lost.

Sequence matters. If energy prices jump first and wages follow a year later, blaming the original inflation on wages reverses the chronology. If wage growth then remains high after energy prices stabilise, labour costs may become part of the persistence mechanism.

Unit labour cost is therefore more informative than wage growth alone. Productivity can offset part of a pay rise. Margins can absorb part. Weak demand can prevent pass-through.

The correction matters because restraining wages may distribute the real-income loss towards workers without being the only route to lower inflation. Diagnosis should separate origin from propagation.

"Higher interest rates fix inflation immediately"

Interest rates influence inflation through financial conditions, spending, labour demand, exchange rates and expectations. Those channels take time.

A rate rise today does not create gas, unclog a port or harvest wheat. Many borrowers are on fixed rates and feel the change only when they refinance. Firms cancel investment gradually. Hiring changes with delay. The strongest effects can arrive after headline inflation has already started falling.

This lag explains a pattern that otherwise looks absurd: central banks can keep tightening while inflation is falling, or leave rates high after the headline rate has returned near target. They are trying to influence future persistence, not today's petrol price.

The correction matters because evaluating policy solely against the next month's CPI release confuses the instrument's time horizon with the statistic's publication cycle.

There is also an asymmetry in public experience. The pain from a rate rise can be immediate for a household refinancing a mortgage, while the benefit arrives diffusely through inflation that is lower than it might otherwise have been. That makes monetary policy politically awkward even when it works. The counterfactual is invisible. Nobody receives a statement saying which future price increases did not happen because demand weakened.

"Two per cent inflation means prices are stable"

At 2 per cent annual inflation, a price level compounds upward. After ten years it is roughly 22 per cent higher if the rate were exactly maintained.

Central banks still use the language of price stability because the practical goal is a low, predictable inflation rate rather than a frozen price level. Relative prices need to move. A small positive target creates room for nominal wages to adjust and for real interest rates to become negative during downturns when nominal rates are low.

The phrase can therefore mislead if interpreted as a fixed price level. Stability means that the unit's erosion is slow and predictable enough to be incorporated into contracts and decisions.

The correction matters because debates about "why not zero?" are legitimate. They should be debates about wage rigidity, measurement, monetary-policy space and institutional trade-offs, not arguments caused by a semantic misunderstanding.

A low positive target also avoids promising a price-level path that policymakers may not be able to deliver without large swings in employment. Price-level targeting is a serious alternative framework, but it is different: past misses are later offset. Ordinary inflation targeting normally lets a past price-level shock remain in the base and focuses on the future rate. Confusing the two makes policy look more inconsistent than it is.

Use It

Separate the level from the rate

Whenever someone says inflation is falling, ask which object they mean. The inflation rate can fall while the price level keeps rising. A household complaining that groceries remain expensive and a central bank reporting disinflation can both be correct.

This single distinction cleans up a remarkable amount of public argument. It also prevents the false expectation that successful inflation policy should restore every old sticker price.

Start with the first relative price

Ask what moved first. Energy? Food? Imported goods? Rents? Wages? Broad consumer demand? The answer does not settle causation, but it gives the chain a starting point.

Then ask whether the first movement represented scarcity, excess demand, tax change, exchange-rate movement, market power or some other force. A useful inflation explanation should be able to state the opening mechanism without using the word inflation as its own cause.

Follow the propagation chain

The origin is only half the problem. Trace the shock into costs, margins, wages, expectations, fiscal support and spending. A gas shock that fades after one price-level jump is different from one that changes wage bargains and services prices for two years.

This lens is especially useful when political arguments assign blame. A group can be central to propagation without having caused the initial shock. Another can have caused part of the original demand pressure without explaining why inflation later persisted. Separating stages prevents one-cause stories from swallowing time.

Translate nominal numbers into real ones

A pay rise, savings rate, investment return, tax receipt or company revenue can look strong during inflation while shrinking in purchasing-power terms.

A 5 per cent nominal savings rate with 6 per cent inflation is a negative real return before tax. Revenue growth of 8 per cent in an economy where relevant prices rose 10 per cent may represent lower real sales. A government collecting more tax in cash terms may not have more real resources to buy public services.

The correct deflator depends on the question. Household CPI is not automatically the right adjustment for every business or investment. The habit is what matters: do not confuse a bigger money number with a bigger real outcome.

Ask who carries the loss

When imported necessities become dearer, the country cannot collectively preserve every previous real claim. Look for the adjustment among wages, profits, creditors, debtors, taxpayers, benefit recipients and public services.

This lens turns many inflation arguments into distributional arguments. A wage settlement may protect workers and squeeze margins. An energy subsidy may protect households and shift cost to taxpayers or public borrowing. Higher interest rates may protect future price stability while imposing losses on refinancing borrowers and interest-sensitive firms.

Economics can trace the transfer. It cannot determine the morally correct distribution on its own.

Match the tool to the mechanism

Interest rates are strong against excessive nominal demand and persistence because they weaken spending and influence expectations. They are weak against the physical source of a supply shock. Fiscal policy can add or subtract demand and redistribute losses. Targeted transfers can protect vulnerable households with less aggregate-demand support than universal subsidies. Supply measures can relax bottlenecks but often work slowly. Price controls change allocation and the measured price without abolishing scarcity.

A policy claim should therefore name its transmission mechanism. "Cut rates", "raise taxes", "cap prices" or "build more" is incomplete until it explains which part of the inflation process the instrument is meant to change and on what timescale.

A useful test is to write the claim as a before-and-after chain. If rates are cut, whose borrowing changes first, how much extra spending follows, whether supply can respond, and what happens to expectations? If taxes rise, which households or firms change spending, and with what lag? If a price is capped, who supplies at the capped price and who receives the scarce quantity? The chain exposes policies that are emotionally satisfying but mechanically incomplete.

Look at the distribution before judging the average

The national inflation rate is necessary for macroeconomic policy and insufficient for household welfare. Ask which categories drove it and which groups spend most on those categories. Then add debt structure, wage reset dates, benefits and taxes.

This prevents two opposite errors. The first is dismissing personal hardship because the national average is lower. The second is assuming that one vivid household proves the official measure is false. Averages and distributions answer different questions. Good analysis uses both.

Distinguish a forecast from a promise

Central banks publish forecasts because policy works with lags. Those forecasts are conditional judgements, not guarantees. When energy prices, fiscal policy, exchange rates or wage behaviour change, the forecast should change too.

Judge an institution less by whether every projection was right than by whether it identified the main risks, updated when evidence changed and kept its policy consistent with the target. Forecast error is inevitable. Refusing to revise the model is not.

The limits

Inflation analysis cannot identify one perfect price index for every person. National consumer indices are averages built for defined purposes. Personal exposure differs by housing, debt, age, location and spending patterns.

Causal attribution is also uncertain. Demand and supply can move together. Expectations are difficult to measure. Potential output is estimated. Fiscal and monetary policies interact. Researchers can agree on the observed inflation rate and disagree about causal shares.

Policy has unavoidable value choices. A real energy loss has to be distributed somehow. Economics can show that protecting one group shifts the burden towards another or towards the future. It cannot decide the morally correct split without political and ethical premises.

Nor does every inflation episode obey the institutions of the United Kingdom, United States or euro area. Exchange-rate regimes, energy dependence, labour bargaining, indexation, debt structure and central-bank credibility vary. Importing a policy lesson without importing those conditions can turn a true historical observation into a false universal rule.

The one thing to keep

Keep the chain.

A useful inflation explanation should tell you what moved first, how the change entered the general price level, why nominal spending and real capacity became inconsistent, how contracts and expectations spread or damped the pressure, and which policy can interrupt that process.

Once you demand that chain, the loudest slogans lose much of their power. "Money", "greed", "wages", "oil" and "deficits" can each name a real part of an episode. None is a substitute for explaining how the part connects to the whole.

The final practical change is simple: stop asking whether inflation is good or bad in the abstract and start asking what mechanism is operating now. That is the difference between reacting to a number and knowing what to do.

Terms

Inflation. A sustained rise in the general price level, usually measured by the percentage change in a price index. It means money buys less across the measured basket.

Price level. The overall level of prices represented by an index. Inflation is the rate at which this level changes, so a lower inflation rate does not imply a lower level.

Disinflation. A decline in the inflation rate while the general price level is still rising. It is the usual goal after an inflation surge.

Deflation. A sustained fall in the general price level. Broad deflation can raise real debt burdens and interact with weak demand.

Consumer Price Index (CPI). A weighted measure of consumer prices. In the United Kingdom it is the inflation measure used for the Bank of England's 2 per cent target.

CPIH. The ONS consumer-price measure that extends CPI to include owner-occupiers' housing costs using rental equivalence and Council Tax. The ONS describes it as its most comprehensive consumer measure.

HICP. Harmonised Index of Consumer Prices, the comparable consumer-price framework used across the European Union and the reference for the ECB's target.

Basket. The selected collection of goods and services used to represent household consumption in a price index. The contents and weights change over time.

Weight. The importance assigned to a category in an index according to expenditure patterns. A heavily weighted category moves the overall measure more.

Base effect. A change in an annual inflation rate caused partly by what happened in the comparison month a year earlier rather than by a large new monthly move.

Nominal. Expressed in current money units without adjusting for changes in purchasing power. Wages, debts and interest rates are usually contracted nominally.

Real. Adjusted for inflation or another relevant price change. Real values distinguish purchasing power or quantity from money amounts.

Real interest rate. A nominal interest rate adjusted for expected or realised inflation. It is the rate that matters for the purchasing-power return to lending or saving. Ex ante real rates use expected inflation; ex post real rates use the inflation that later occurred.

Core inflation. An inflation measure excluding selected volatile components, often food and energy, used to examine underlying persistence. It is not a measure of household hardship.

Demand-pull inflation. Inflation pressure associated with nominal demand growing faster than sustainable real supply. The term is useful when excess spending is central to the episode.

Supply shock. An unexpected change in productive capacity or input availability, such as an energy disruption, harvest failure or shipping interruption.

Cost-push inflation. A label for price pressure transmitted from higher production or input costs. It describes one channel but does not by itself explain persistence.

Output gap. The difference between actual output and an estimate of sustainable potential output. It is important for inflation analysis and difficult to observe directly. A positive gap suggests demand may be pressing beyond sustainable capacity; a negative gap suggests slack.

Potential output. An estimate of how much an economy can produce sustainably at normal utilisation. It can change when labour supply, capital, productivity or disruptions change. Because it is estimated rather than observed, policy can discover capacity limits only after pressure appears.

Inflation expectations. Beliefs held by households, firms and markets about future inflation. They influence wage, price and interest-rate decisions.

Anchored expectations. Expectations that remain close to the monetary authority's target over medium and long horizons despite temporary inflation shocks. Anchoring reduces the chance that a temporary shock rewrites wage, price and lending contracts for years.

Indexation. Automatic adjustment of a payment, wage, tax threshold or contract according to a price index. It protects real value while carrying past inflation into future nominal payments.

Second-round effects. The spread of an initial price shock into wages, margins, expectations and broader domestic prices after the first direct effect.

Pass-through. The degree to which a change in input costs, taxes or exchange rates appears in final prices. Pass-through depends on contracts, competition, demand and margins.

Unit labour cost. Labour compensation per unit of output. Wage growth creates less cost pressure when matched by productivity growth.

Monetary policy. Central-bank action that influences financial conditions, demand and expectations, usually through a policy interest rate and related tools.

Policy rate. The short-term interest rate set or targeted by a central bank. Changes transmit through market rates, borrowing, saving, asset prices and expectations. It is an instrument, not the price of every loan in the economy.

Fiscal policy. Government decisions on taxation, spending, transfers and borrowing. It can change aggregate demand, distribution and future productive capacity.

Price control. A legal restriction on a price. If the controlled price prevents supply and demand from clearing, scarcity must be allocated through another mechanism such as queues, rationing, quality changes or subsidy.

Hyperinflation. Extremely rapid inflation associated with a severe breakdown in monetary, fiscal and political credibility. It is not a useful default analogy for ordinary inflation above target.

Go Deeper

For an accessible modern overview: Stephen D. King, We Need to Talk About Inflation: 14 Urgent Lessons from the Last 2,000 Years (Yale University Press, 2023). King moves between historical episodes, institutions and current policy with little formal mathematics. Read it for the recurring political economy of inflation and the temptation to believe each generation has solved the problem permanently. His policy judgements are arguments rather than settled doctrine, which is useful because it forces the reader to separate mechanism from recommendation. The historical range also shows why inflation debates repeatedly become moralised: the economic mechanism is inseparable from arguments over debtors, savers, workers, firms and the state.

For the classic expectations argument: Milton Friedman, "The Role of Monetary Policy", American Economic Review 58, no. 1 (1968): 1-17. This presidential address explains why policymakers cannot assume a permanent trade-off between inflation and unemployment once expectations adapt. It is short enough to read in one sitting and more careful than the slogans later attached to Friedman. Read it as a historical turning point in macroeconomic thinking, not as a complete theory of every inflation episode. Pay particular attention to the distinction between temporary real effects and the long-run limits of monetary policy. That distinction is the part most often lost in later summaries.

For the institutional framework: Ben S. Bernanke, Thomas Laubach, Frederic S. Mishkin and Adam S. Posen, Inflation Targeting: Lessons from the International Experience (Princeton University Press, 1999). This explains why explicit targets, communication, accountability and credibility became central to modern monetary policy. The cases predate the pandemic and energy shock, which makes the framework easier to inspect before its latest stress test. It is more technical than King but still accessible to a determined general reader. Its value is institutional: it explains why targets work through expectations, accountability and a framework for decisions rather than through the magic of announcing a number.

For the Great Inflation and the cost of ending it: Robert J. Samuelson, The Great Inflation and Its Aftermath (Random House, 2008). Samuelson gives a readable US-centred account of the 1960s and 1970s, the politics surrounding inflation and the eventual Volcker disinflation. Use it as a case study in how inflation can become institutional and politically entrenched. It should not be treated as a universal template for countries with different labour, fiscal and monetary systems.

Notes and Sources

Definition and measurement

Definitions of CPI, CPIH, the distinction between rates and levels, and current UK consumer-price methodology follow Office for National Statistics consumer-price releases and methodology. The latest release available on 11 August 2026 was Consumer price inflation, UK: June 2026, published 22 July 2026, reporting CPI inflation of 2.6 per cent and CPIH inflation of 2.8 per cent. The July 2026 release was not yet published and is not treated as known.

The 11.1 per cent UK CPI peak used in the opening comes from the ONS October 2022 release. ONS later described that October 2022 figure as the highest annual CPI rate in the accredited series beginning in 1997 and, on modelled estimates, the highest in more than forty years.

Housing treatment, weighting, substitution and quality adjustment follow ONS methodology. The manuscript avoids treating the national CPI basket as a personal cost-of-living measure and does not use consumer-price indices as measures of asset-price affordability.

Targets and current policy

The Bank of England's target is 2 per cent CPI inflation over the medium term, set by the UK Government. At the meeting ending 29 July 2026, the Monetary Policy Committee voted 6 to 3 to maintain Bank Rate at 3.75 per cent. The decision was published 30 July 2026. Current-policy references in the narrative are used only to illustrate transmission and forecasting, not as durable prescriptions.

The European Central Bank aims for 2 per cent HICP inflation over the medium term. The manuscript states this only where useful for comparison and does not turn the book into a survey of central-bank mandates.

Demand, supply and monetary transmission

The demand-capacity framework, real-versus-nominal distinction and treatment of expectations follow standard modern macroeconomics and central-bank practice. The Bank of England's 2024 review, About a rate of (general) interest: how monetary policy transmits, supports the description of policy operating through financial conditions, expectations, activity and inflation rather than through an immediate direct control of prices.

Current Bank of England Monetary Policy Reports from 2026 were checked for treatment of energy shocks, labour-market slack and persistence. The manuscript keeps time-sensitive forecasts out of the conceptual core.

Pandemic and energy inflation

The 2021-23 account is deliberately multi-causal. Pandemic restrictions disrupted supply and shifted spending patterns; fiscal income support and accommodative financial conditions supported demand; reopening encountered constrained supply; Russia's invasion of Ukraine added a major European energy shock.

Niels-Jakob Hansen, Frederik Toscani and Jing Zhou, IMF Working Paper 2023/131, is used for the euro-area distribution between import prices, profits and wages. The paper's decomposition shows a substantial profit counterpart during part of the episode but does not establish a universal increase in markups. The manuscript therefore rejects both the claim that profits were irrelevant and the claim that greed alone explains the inflation burst.

The Great Inflation and Volcker disinflation

Federal Reserve History dates the US Great Inflation from 1965 to 1982 and documents the interaction of demand, policy, oil shocks, changing monetary arrangements and the eventual anti-inflation shift under Paul Volcker. Its separate history of the 1981-82 recession supports the statement that tight monetary policy used to fight inflation produced a severe downturn.

The manuscript treats the episode as evidence that persistent inflation can raise the later cost of stabilisation, not as proof that every modern inflation episode has the same causes or requires the same treatment.

Distribution and debt

The redistribution from unexpected inflation follows nominal-contract arithmetic: if the money value of a fixed debt is unchanged while the price level and incomes rise unexpectedly, its real burden falls relative to what was contracted. Expected inflation can be priced into future nominal rates, limiting the ability to repeat the transfer.

Household exposure is described conditionally because expenditure shares, wages, debts, assets, taxes and benefits differ. No universal claim is made that inflation is always regressive or always favourable to debtors.

Policy tools

Interest-rate transmission follows Bank of England material. Fiscal policy is treated according to its effects on demand, distribution and capacity rather than through a rule that every deficit is inflationary. Price controls are treated as allocation mechanisms: if a ceiling prevents market clearing, scarcity must appear through rationing, queues, subsidy, quality changes, informal markets or shortage. Supply reform is presented as valuable where it removes real constraints, with the warning that its timetable is often slower than inflation persistence.

Bibliography

Primary, official and institutional sources

Bank of England. About a rate of (general) interest: how monetary policy transmits. Quarterly Bulletin, 12 July 2024.

Bank of England. Inflation and the 2% Target. Current online edition, accessed 11 August 2026.

Bank of England. Monetary Policy Report. February 2026.

Bank of England. Monetary Policy Report. April 2026.

Bank of England. Monetary Policy Report. July 2026.

Bank of England. Monetary Policy Summary and Minutes. 30 July 2026.

European Central Bank. Two per cent inflation target. Current online edition, accessed 11 August 2026.

European Central Bank. Measuring inflation and consumer prices. Current online edition, accessed 11 August 2026.

Federal Reserve History. The Great Inflation. Current historical resource, accessed 11 August 2026.

Federal Reserve History. Recession of 1981-82. Current historical resource, accessed 11 August 2026.

Friedman, Milton. "The Role of Monetary Policy." American Economic Review 58, no. 1 (1968): 1-17.

Hansen, Niels-Jakob H., Frederik G. Toscani, and Jing Zhou. "Euro Area Inflation after the Pandemic and Energy Shock: Import Prices, Profits and Wages." IMF Working Paper 2023/131. Washington, DC: International Monetary Fund, 2023.

Office for National Statistics. Consumer price inflation, UK: October 2022. 16 November 2022.

Office for National Statistics. Consumer price inflation, UK: June 2026. 22 July 2026.

Office for National Statistics. Consumer Price Inflation, Quality and Methodology Information. Current 2026 edition.

Modern works

Bernanke, Ben S., Thomas Laubach, Frederic S. Mishkin, and Adam S. Posen. Inflation Targeting: Lessons from the International Experience. Princeton: Princeton University Press, 1999.

King, Stephen D. We Need to Talk About Inflation: 14 Urgent Lessons from the Last 2,000 Years. New Haven: Yale University Press, 2023.

Samuelson, Robert J. The Great Inflation and Its Aftermath: The Past and Future of American Affluence. New York: Random House, 2008.