Opening

Opening

Okay. Hello. Today we are going to talk about inflation. Simply put, inflation measures how fast prices increase over time. And we’ll answer a few big questions: What is inflation, and how do we measure it? Is inflation good or bad for society? And if it’s bad, how can central banks control it — and at what cost?

Usually people think inflation is bad because the same money buys you fewer goods. Groceries cost more. Bus tickets cost more. And if you still have a student loan, inflation can actually help you a little — you pay the loan back in money that buys less, so in real terms the burden is lighter. Still, inflation often feels unfair: you didn’t do anything differently, but your money gradually loses its value.

The receipt

I’ll start with something both of us are familiar with: the receipt. This slide shows two receipts. One is from 2002 at JUSCO, and the other is from 2026 at AEON. Since JUSCO later became AEON, they are actually the same store. So we are looking at how prices change for the same store, the same items, over about 24 years.

Now look at the table. Five-kilogram rice rises from 980 yen in 2002 to 4,200 yen in 2026 — up around 300%. One-liter milk rises from 98 yen to 260 yen — up 165%. The other groceries rise by a similar amount, roughly 200% to 300%. In total, the whole bundle roughly triples, from 1,200 yen to 5,200 yen.

If we average those five percentage increases, we get about 259%. That’s the average price increase across products — but not yet a yearly rate. For average inflation per year, take final expense over initial expense, take the 24th root, and subtract 1. That number is around 6%.

Is 6% the inflation rate?

So — do you think 6% is the inflation rate? Of course not. We only looked at one shop, two days, and five items. That’s way too narrow.

Also, not everyone buys the same bundle, so not everyone feels inflation the same way. Look back at the table: milk rose less, rice rose more. If your basket is mostly milk, you feel less of a price increase than if your basket is mostly rice. Different shopping lists mean different inflation stories.

So to give everyone a shared benchmark, we need a basket that stands in for consumers in general. That basket is tracked by the Consumer Price Index, or CPI.

The Consumer Price Index

Formally, CPI is the cost of a fixed basket of goods and services relative to a base period. Right now the base period is 2025, so CPI in 2025 is set to 100. In 2026, CPI asks: how much more does that same fixed basket cost?

In Japan, the Statistics Bureau builds this basket. They survey families about what they actually buy, then construct an average consumer basket. It’s not your basket or my basket — it’s the average consumer’s basket.

In Japan, food is about 28%, housing about 22%, and the rest about 32% — things like education, clothing, or a movie ticket when you go out with friends.

Now look at how we compute CPI. The basket has q-j units of each good j. The numerator is what that same basket costs today — today’s prices times those fixed quantities. The denominator is what it cost in the base period. Divide today by the base period, multiply by 100, and that’s CPI. If CPI is 103, the basket costs 3% more than in the base year.

Inflation, Deflation, and Disinflation

Often we care not just about the price level — the CPI — but about how fast that level is changing. Policymakers and newspapers talk about the inflation rate: this year’s CPI minus last year’s CPI, divided by last year’s CPI.

Inflation means the price level is rising: CPI goes up, and the inflation rate is positive. Deflation means CPI is falling — on the graph, the line slopes down over time.

You might think deflation is good, because the same money buys more goods. In reality, it can be dangerous. When people expect cheaper prices next year, they delay spending. With less consumption, firms earn less and stop hiring. Incomes fall, spending falls again — a vicious cycle. Japan lived through this for more than 20 years, until very recently.

Finally, disinflation. Disinflation is still inflation — prices are rising — but the rate of increase is getting smaller and smaller. So keep the distinction clear: deflation means the price level itself drops; disinflation means the price level is still rising, just more slowly.

Example: milk and rice

Let’s walk through one example so everyone is clear. Suppose the basket has two liters of milk and two bags of rice. We want CPI at each period, the inflation rate, and whether period 2 is deflation.

Look at the table. Milk starts at 100 yen, rises to 120 in period 1, then to 125 in period 2. Rice starts at 200, then 240, then 250.

First, total spending in the base period: 2 times 100 plus 2 times 200 — that’s 600 yen. We set CPI to 100. In period 1, spending is 720 yen, so CPI is 100 times 720 over 600 — which is 120. In period 2, spending is 750 yen, so CPI is 125.

Now the inflation rates. Period 1: 120 minus 100, over 100 — 20%. Period 2: 125 minus 120, over 120 — about 4.2%.

Is period 2 deflation? No. Inflation is still positive — prices are still rising — but the rate fell from 20% to 4.2%. That is disinflation, not deflation.

Three costs of inflation

We’ve learned how to measure inflation. Before we ask what causes it: is inflation bad? It does have real costs.

First, shoe-leather costs — the hassle of holding cash. When prices keep rising, cash in your wallet loses value. To protect that value and earn interest, people hold less cash and make more trips to the bank or ATM. That time is wasted — time you could spend on something productive, like homework. Estimates suggest even 3% inflation can cost roughly half a percent of GDP through shoe-leather costs alone.

Second, from the firm’s side: menu costs. Changing prices is not free — reprint menus, update websites, notify customers. And when prices jump around, customers get confused about what things are really worth. If a good jumps from 50 dollars to 100 dollars in two days, decisions get much harder.

Third, and most important: unexpected inflation. Think of a fixed-rate loan. You borrow 100 dollars, and next year’s inflation turns out to be about 10%. You pay back money that buys only about 90 dollars’ worth of goods. You, the borrower, gain; the lender loses. The gains and losses cancel in size — so the issue is fairness, not waste.

The real waste comes earlier. When nobody can predict prices, lenders protect themselves: higher interest premiums, shorter loans, or no lending at all. But many projects — a factory, a railway — need money locked up for 20 years. Unpredictable inflation stops those investments. It’s like a game where the rules keep changing. If you can’t trust the rules, you don’t play — and valuable projects never get funded.

Benefits of inflation

Inflation is not only costs. If it were, central banks would aim for zero. Today the consensus is that mild, stable inflation around 2% is beneficial.

First, it encourages investment. When inflation is low but positive, cash sitting in a bank account slowly loses value. That pushes people to put money into the real economy — equipment, factories, goods people use — instead of letting cash sit idle.

Second, it slowly dilutes the real value of government debt. Public debt is usually written in nominal yen or dollars. A few percent inflation each year shrinks the real burden of repayment without rewriting the contract.

Third, it helps the job market. When demand for labor in one sector falls, real wages need to adjust. Companies can cut nominal pay, or keep pay and lay people off. Layoffs are costly: incomes fall, spending falls, and we risk the vicious cycle again. But workers hate nominal pay cuts. Imagine doing nothing wrong, and your boss suddenly takes 2% off your wage. With a little inflation, you might still get a 2% raise while inflation is 5%. It still feels like a raise — but your purchasing power fell a bit. That soft adjustment is why modest inflation can “grease the wheels” of the labor market.

Hyperinflation

So far we’ve talked about ordinary inflation — a few percent a year. The extreme case is hyperinflation, as in Zimbabwe. Prices can double in hours or days. Look at the right-hand side: 100-billion notes. Every extra zero is evidence of how far printing money can go.

The pattern is similar in almost every hyperinflation. The government spends more than it collects, so it borrows. Investors buy bonds while they still trust repayment. Then trust collapses: the government can no longer refinance, and no one will buy new bonds. The only remaining option is to print money.

As more money chases the same goods, prices rise. Once people expect that, they stop holding cash and spend it as soon as they get it — which pushes prices even higher. Workers demand higher wages, the government prints more to cover spending, and the cycle feeds on itself.

Tools against inflation

We’ve seen the costs — and in the extreme, hyperinflation. So what can central banks do?

The tool you hear about most is the interest rate — the price of borrowing. If the central bank wants to push prices down, it raises rates. Loans get more expensive. Households spend less; firms invest less. Demand cools. With supply unchanged, prices ease. The classic trade-off: fighting inflation often means slower growth. Lower rates can boost growth, but usually with more inflation pressure.

Sometimes the problem is deflation, not inflation. Then the central bank wants to cut rates to stimulate demand. But when short-term rates are already near zero, cutting further is hard. So the bank turns to long-term rates — the rates that matter for mortgages, business investment, and government borrowing.

To bring long-term rates down, the central bank buys long-term government bonds. More demand for bonds raises bond prices and lowers long-term yields. That is quantitative easing, or QE.

A related tool is yield curve control, or YCC. Under QE, the bank targets how many bonds to buy — say 1 or 2 trillion. Under YCC, it buys whatever it takes to pin the long-term rate near a target.

The third tool is reserve requirements. Banks cannot lend out every yen they take in; regulators make them hold a share of deposits as idle reserves — often around 10%. Raise that ratio, and banks lend less: credit tightens, inflation pressure falls. Lower it, and banks can lend more: credit loosens, money flows into the economy.

Diagnosing the cause

We know the toolkit. Before picking a tool, ask what is driving inflation. Three channels: demand, supply, and expectations. I’ll use the United States as examples.

First, demand. During COVID, factories shut down. Then governments tried to boost the economy with what we call helicopter money — cash sent straight to households, as if dropped from the sky. People had more money, spending rose faster than firms could supply, and prices were pulled up.

Second, supply. An oil shock — say conflict around the Persian Gulf or the Strait of Hormuz — makes energy scarce and expensive. Energy enters almost everything, so costs pass through to final prices. A trade war works the same way: tariffs and shipping costs get passed into prices, even without a spending boom.

Third, expectations. Suppose inflation starts high because of a supply shock — a war in the Middle East pushes up oil prices. Even after the war ends, people may still expect prices to keep rising. Workers ask for higher wages; firms raise prices to cover those wages. Higher prices lead to higher wage demands again. A temporary shock can become a long-lasting, self-fulfilling inflation.

The output–inflation tradeoff

How policy should respond depends on the cause.

If inflation is mostly demand-driven — helicopter money, over-generous fiscal stimulus — then tightening monetary policy helps: raise rates or raise reserve ratios. Spending and investment fall, demand cools, and inflation eases.

If inflation is supply-driven — oil or a trade war — then tightening demand alone may not fix prices much, but it can cut output. Rising prices plus falling output is stagflation: stagnation plus inflation. In that case, cutting demand hard can hurt the economy without much gain. So before you choose the tool, identify the cause.

Case study: Japan through the CPI chart

Finally, one case study: Japan’s CPI inflation over time, starting from 1990.

Around 1990, inflation hovered near 3%. The story begins a bit earlier. In 1985, under the Plaza Accord, the U.S. pushed Japan to let the yen appreciate so American goods could compete better. A stronger yen hurt Japanese exports. To prevent a recession, the Bank of Japan cut rates. That cushioned the economy — but cheap money poured into stocks and real estate. At the peak, land in Tokyo alone was estimated to be worth more than the entire United States.

When the BOJ raised rates around 1989–90, the bubble burst. The economy slowed, and inflation fell toward zero. Then came roughly two decades of near-zero or negative inflation — Japan’s deflation era.

You can also see sharp dips in crisis years: around 2008–09, after the global financial crisis and an oil-price crash; and around 2020–21 during COVID, when much of the economy shut down.

There was a bump around 2013–14 under Abenomics. Some credited the stimulus. But the Statistics Bureau noted that much of the 2014 jump came from raising the sales tax from 5% to 8%. Strip out the tax, and underlying inflation was still well below 2%.

Only very recently did inflation rise more persistently. A weak yen made imports — especially energy, which Japan lacks — much more expensive. After about 20 years of fighting falling prices, Japan now has to manage inflation that has finally returned.