Who steers the economy?

Nearly a century ago, the British economist John Maynard Keynes pointed out a simple problem. Shops and factories can only make so much, limited by their staff and machines. But how much people want to buy swings up and down with how confident or scared they feel and how easily they can borrow. So the total amount people want to buy doesn’t automatically match what a country can make.

When people want to buy much less than shops can make, shops sell less and cut staff. Those people now have less pay, so they buy less too, and the downturn feeds itself. A bad time like this is called a slump. When the economy actually shrinks for a while, it’s called a recession, and the share of people who want a job but can’t find one, called unemployment, goes up.

When people want to buy much more than shops can make, the shops can’t keep up. Lines form and shelves empty, so shops raise their prices. Their staff ask for higher pay to keep up, and prices rise again. A hot time like this is called a boom. When it goes too far, economists call it overheating, and it shows up as high inflation.

Keynes’s answer was that someone should step in: speed buying up in a slump, and slow it down in a boom. Evening out the swings like this is called stabilisation policy, and today the main “someone” is the central bank (chapter 2). Try both extremes in the toy town below.

The first models were built by looking at past numbers and copying the patterns, as if they were fixed laws. One pattern said that when lots of people were out of work, prices stayed calm, and when almost everyone had a job, prices rose faster. Economists drew it as a line called the Phillips curve. Another pattern said that people spend a fixed share of whatever they earn this year. (Robert Hall, one of this year’s winners, later tested that idea. That’s chapter 4.)

Those patterns made the dial easy to set. Want more jobs? Push spending up and accept a little more inflation. For about twenty years after the Second World War, that seemed to work.

Then, in 1973, oil prices suddenly jumped to about four times what they had been. Oil runs factories, trucks and heating, so almost everything got more expensive to make and move. Businesses raised their prices to cover the extra cost, and some made less and cut staff. So prices rose fast and people lost their jobs, at the same time. The old pattern said that couldn’t happen. This mix got its own name: stagflation, a stalled economy plus inflation.

The old rulebook only knew one kind of slump: people buying too little, so shops sell less and cut staff. Its fix was to push spending up, by keeping interest rates low and having governments spend more. The goal was to bring jobs back: more customers means shops sell more and hire again. That works when shops have staff and machines standing idle, waiting for customers.

But this slump was a different kind. People hadn’t stopped buying. Making things had got more expensive. Extra spending can’t make oil cheaper, so shops and factories couldn’t make much more, however many customers came. When governments and central banks pushed spending up anyway, the extra money mostly pushed prices even higher, and the jobs still didn’t come back. The problem was costs, not too little buying, so the old fix aimed at the wrong thing.

After years of rising prices, people started to expect more of the same. Workers asked for bigger raises to keep up, shops raised prices again to pay them, and inflation fed itself. A second oil jump in 1979 made it worse. In the US, high inflation was finally brought down in the early 1980s, after the central bank raised its rate to around 20%, which made loans so expensive that people spent much less. Price rises slowed, but the cost was a deep recession, with unemployment above 10% by the end of 1982.

Why did the patterns break? An earlier winner of this prize, the economist Robert Lucas (1995), gave the main answer. Patterns in past numbers only show how people behaved under the conditions of that time. For example, when prices had been steady for years, shops rarely raised prices in hard times. Once everyone had lived through years of 10% inflation, shops raised prices anyway, because they assumed everyone else would. People had changed how they behave, so the old pattern vanished.

His fix was to build the model from the bottom up: start from how single households decide what to spend and how single businesses set their prices, then add them all together. Economists call the small scale micro (one person, one shop) and the whole country macro. That is what the prize’s official wording means: “microeconomic foundations of macroeconomic stabilisation”. Building such a model that was also simple enough to use took economists decades, and it’s what Hall and Woodford helped do.

Think of a café. It can’t print a new menu every morning. Changing prices takes time and money, and customers hate it. So the café changes its prices only now and then, maybe once or twice a year. Economists call these sticky prices.

Because the café is stuck with its price for months, it has to look ahead when it sets one. If the owner expects milk, rent and other cafés’ prices to rise over the coming months, she sets a higher price now. So what businesses expect drives prices, just as what households expect drives spending.

In 2003, Woodford (then at Princeton University, now at Columbia) put all the pieces together in a book called Interest and Prices. His model has three parts: households that spend based on what they expect (Hall’s result), businesses with sticky prices that look ahead, and a central bank that sets the interest rate. The Nobel committee calls it “a kind of laboratory for policymakers”: you can ask it what happens to jobs and prices if the rate goes up from, say, 3% to 4%, before trying it on a real country.

This kind of model is called New Keynesian: Keynes’s old idea of steering, rebuilt on the choices of real people. In it, as in real life today, the central bank steers by setting the interest rate.