Opening
HOST
Welcome back. Today we are on the big questions of macroeconomics, and on the models economists use to answer them. In 1989, prices in Argentina rose by about 3,080 percent. What does that mean?
ECONOMIST
A rise of 3,080 percent adds 30.8 times the old price on top of the old price. So prices were on average about 32 times as high as a year before. A price rise that large is called hyperinflation.
HOST
And this is the kind of thing macroeconomics studies?
ECONOMIST
Yes. Macroeconomics is the study of the economy as a whole: a country's total income, its jobs and its prices. Whole economies can grow for more than a century. They can stall, when output, everything the country produces, falls and jobs are lost. And they can inflate, when prices rise fast.
HOST
So here is today's question. Why do whole economies grow, stall and inflate, and how can a model help us answer?
ECONOMIST
By the end of this episode, you'll be able to do three things. First, describe four big facts, and explain why the share of adults with a job can fall for two reasons. Second, explain what goes into a model and what comes out, and use a small model of the market for work to predict what a minimum wage does. Third, tell the difference between the long run and the short run, and use the rule of 70 to turn a growth rate into a doubling time.
HOST
And we'll follow the chapter's mind map part by part?
ECONOMIST
Yes. Part one is the questions and the facts. Part two shows how a model works. Part three is about the long run, the short run and policy. Each part has four leaves, small topics we take in order.
The questions & the facts
HOST
Part one of the map: The questions and the facts. The first leaf is income gaps. How big are they?
ECONOMIST
Huge. Economists measure a country's income with GDP, short for gross domestic product. It's the total income a country makes in a year. Divide it by the population, and you get income per person. In 2022, income per person in the United States was about 25 times as high as in Ethiopia.
HOST
25 times! How can one chart show incomes that different?
ECONOMIST
With a ratio scale, where each equal step up the chart doubles the value: from 1,000 dollars to 2,000, then to 4,000. On it, a country that grows at a steady rate rises by the same distance every year, so it draws a straight line.
HOST
Next leaf: recessions cost jobs. What exactly is a recession?
ECONOMIST
A recession is a period when a whole economy produces less and jobs are lost. To measure the damage, we use the employment-population ratio, the share of adults who have a job. Before 2007, no recession lowered it by more than 2.4 percentage points.
HOST
And the two latest recessions?
ECONOMIST
Much deeper. The recession from 2007 to 2009 lowered the ratio by 4.5 points, almost twice the old record.
HOST
What about Covid in 2020?
ECONOMIST
Even deeper. In February, about 61 percent of American adults had a job. Two months later, only about 51 percent did. That's a fall of 9.9 points, almost 10.
HOST
So all those people became unemployed, I guess.
ECONOMIST
A lot of students think so, but not all of them did. We need three terms. The labor force is everyone who has a job or is looking for one. The unemployment rate is the share of the labor force that is looking for work but has no job. And the participation rate is the share of adults who are in the labor force.
HOST
Three terms. How do they fit together?
ECONOMIST
Take the share of adults who are in the labor force. Multiply it by the share of the labor force that has a job. That gives the share of adults with a job. So it can fall for two reasons. People lose their jobs and keep looking, so unemployment rises. Or people stop looking and leave the labor force, so participation falls. In April 2020, about 70 percent of the fall came from higher unemployment, and about 30 percent from people leaving.
HOST
Next leaf: inflation comes in waves. First, what exactly is inflation?
ECONOMIST
The inflation rate is how fast prices rise: the percentage change in consumer prices over one year. In rich countries, it has come in waves. In the 1970s, inflation was high in much of the world. In Britain, for example, prices rose 24 percent in 1975 alone.
HOST
What happened after that?
ECONOMIST
From 1990 to 2019, inflation was low and calm. It averaged about 2.5 percent a year in the United States, in Britain and in the euro area, the countries that share the euro.
HOST
And then it came back.
ECONOMIST
Yes. In 2021 and 2022, inflation jumped in the United States, Britain, the euro area and Japan. In the United States, it reached 8 percent in 2022.
HOST
Last leaf of this part: unemployment differs. My guess is that Europe has always had higher unemployment than the United States.
ECONOMIST
Only half right. Recently, yes: from 2006 to 2025, Europe's unemployment rate was on average about 1.6 times the U.S. rate. Here Europe means 12 countries of the euro area. But not always.
HOST
So when was it different?
ECONOMIST
Before 1979, it was the opposite. In every year from 1960 until then, Europe's rate was below the U.S. rate, on average just over half of it. Europe's rate first rose above the U.S. rate in 1979. Since the mid-1980s, it has been higher in almost every year.
HOST
So what changed?
ECONOMIST
The facts alone can't tell us. People suggest rules on firing, unemployment benefits and taxes on work. To test stories like these, we need a model.
HOST
So in part one we saw four big facts. Income gaps are huge, and recessions cost jobs. Inflation comes in waves. And unemployment can stay higher in one place for decades.
ECONOMIST
Yes. Next branch: how a model answers the questions that facts raise.
How a model works
HOST
Part two of the map: How a model works. The first leaf is four steps. What are they?
ECONOMIST
First, collect the facts with data, as we just did. Second, build a model: a small, simple economy written as a few equations. Third, compare the model's predictions with the facts, including the numbers. Fourth, use the model for experiments: change one input and compare the outcomes.
HOST
Next leaf: inputs, model, outcomes. What goes in?
ECONOMIST
Two kinds, both chosen by the model's builder. Parameters are fixed numbers that change only in an experiment, like a minimum wage set by law. Exogenous variables come from outside the model: exo means outside. They may change over time, but along a path set in advance, like a population that grows 1 percent a year.
HOST
So what comes out?
ECONOMIST
The endogenous variables, the outcomes the model works out inside: endo means inside. In a model of the market for work, they are the wage and the hours worked. You choose the inputs, and the equations do the rest.
HOST
Next leaf: the labor market.
ECONOMIST
Let's visit Islandia, the small imaginary island country we follow through the book. Islanders sell hours of work in a single labor market, at a single wage, to one firm, Harbor Boatworks. The first rule is labor supply: the higher the wage, the more hours islanders offer. Even at a wage of zero, they would offer some hours, the work they like to do anyway.
HOST
Is the second rule the boatworks' side?
ECONOMIST
Yes, labor demand. At a wage of zero, the boatworks would hire a certain number of hours. Each extra dollar of wage makes it hire one hour less. The wage settles where the hours offered equal the hours wanted. Economists call that point the equilibrium.
HOST
Let's try it with numbers.
ECONOMIST
For each extra dollar of wage, islanders offer 2 more hours. We'll call that the supply response. At a wage of zero, islanders would offer 30 hours, and the boatworks would want 60. These three numbers are the model's parameters.
HOST
So at a zero wage, the boatworks wants more hours than islanders offer.
ECONOMIST
Yes: 60 minus 30, a gap of 30 hours. It's just by chance that this gap equals the islanders' own 30.
HOST
How does a higher wage close the gap?
ECONOMIST
Each extra dollar closes it by 3 hours: 2 more hours offered, and 1 hour less wanted.
HOST
So the gap of 30 hours is gone at 30 divided by 3. A wage of 10 dollars?
ECONOMIST
Exactly, that's the market wage, and it's the leaf's formula in words. Divide the gap at a zero wage by 3, which is one plus the supply response. At that wage, the boatworks wants 60 minus 10, so 50 hours, and islanders offer the same.
HOST
Now an experiment: we change one input. Suppose more islanders reach working age, so they offer 3 more hours at every wage. There's still only one boatworks. So some of them end up unemployed?
ECONOMIST
That's the natural worry, but no. At the old wage of 10 dollars, islanders now offer 53 hours, but the boatworks wants only 50. With hours left unsold, the wage falls, and a lower wage makes the boatworks hire more.
HOST
Where does it stop?
ECONOMIST
The gap at a zero wage is now 27 hours. Divide by 3, and the wage is 9 dollars. Hours worked rise to 51, and nobody is left out. So a supply shift moves the wage and jobs in opposite ways.
HOST
What if the boatworks wins a big order, and would now hire 63 hours at a zero wage?
ECONOMIST
Then at the old wage, the boatworks wants more hours than islanders offer, so the wage rises, and islanders offer more hours. The wage goes from 10 to 11 dollars, and hours from 50 to 52. A demand shift moves the wage and jobs together.
HOST
Last leaf of this part: a minimum wage above the market wage. Suppose Islandia passes a law that nobody may be paid less than 12 dollars an hour. That's 2 dollars above the market wage. My guess is 2 hours of unemployment.
ECONOMIST
That's the tempting answer, but the model says 6 hours. A minimum wage works like a floor: the wage cannot go below it. This floor has two effects at once. First, each hour now costs the boatworks more, so it hires 2 hours less: 48 instead of 50.
HOST
What's the second effect?
ECONOMIST
The higher wage makes islanders want to work more. They offer 2 more hours for each extra dollar, so 4 more hours in total. Add the 2 hours the boatworks no longer hires, and you get 6 hours that islanders want to sell but nobody buys. That's unemployment. This time, the floor stops the wage from falling.
HOST
So each dollar above the market wage adds 3 hours?
ECONOMIST
Yes. Here is the leaf's formula in two steps. Take one plus the supply response: that's 3. Multiply it by how far the minimum wage is above the market wage, here 2 dollars. That gives 6 hours. A minimum wage at or below the market wage changes nothing at all.
HOST
So should Islandia have a minimum wage? Who gains, and who loses?
ECONOMIST
In the model, islanders who keep their hours gain, and those who lose hours or can't find work lose. But the model leaves things out. Our boatworks takes the market wage as given. In some real markets, a firm can set the wage, because workers have few other choices. In such markets, a moderate minimum wage can raise both pay and jobs.
HOST
What does the evidence say?
ECONOMIST
Many studies find small job losses from moderate rises. Others find larger losses, especially for big rises or young workers. The debate is still open.
HOST
So in part two we saw that a model is a small machine: we choose the inputs, and the equations work out the outcomes.
ECONOMIST
Exactly. In Islandia, a demand shift moved the wage and jobs together, a supply shift moved them apart, and the minimum wage left 6 hours unsold. Next branch: the long run, the short run and policy.
Long run, short run & policy
HOST
Part three of the map: Long run, short run and policy. The first leaf is the long run. What does that mean?
ECONOMIST
The long run is about growth over decades: why living standards rise, and why some countries grow faster than others. Since 1870, income per person in the United States has grown about 1.9 percent a year.
HOST
1.9 percent doesn't sound like much. Over 150 years, I'd guess income maybe tripled.
ECONOMIST
That's the natural guess, but it's far too low. By 2022, U.S. income per person was almost 17 times as high as in 1870, because growth compounds, as the next leaf shows.
HOST
That brings us to the next leaf, the rule of 70.
ECONOMIST
It turns a growth rate into a doubling time: divide 70 by the growth rate in percent.
HOST
Why 70? At 1 percent a year, I'd need 100 years to add 100 percent.
ECONOMIST
Without compounding, yes. But each year's growth is taken from a bigger base, which already includes all the growth before it. So income doubles sooner, in about 70 years. Grow twice as fast, and it doubles in half the time.
HOST
So for the United States, 70 divided by 1.9?
ECONOMIST
That's about 37 years. So American income per person has doubled about every 37 years.
HOST
Then 150 years divided by 37. That's about four doublings?
ECONOMIST
A little over four. Each doubling multiplies income by 2. So after two doublings, income is 4 times as large, and after three, it's 8 times as large.
HOST
So after four doublings, 16 times.
ECONOMIST
Right. And a little over four doublings gives a bit more than 16 times. That's the almost 17 we saw in the data.
HOST
Next leaf: the short run. What's different there?
ECONOMIST
The short run is about the rises and falls around the long-run path. To measure them, we compare actual output, meaning GDP, with potential output. Potential output is what the economy would produce in normal times, with its workers and machines in normal use. The output gap is how far actual output is from potential, in percent of potential.
HOST
How big do the gaps get?
ECONOMIST
The deepest gap came in the Great Depression, the economic crisis of the 1930s. In 1933, U.S. output was 17 percent below potential. Before 1949, potential output is only a rough trend.
HOST
What about a more recent crisis?
ECONOMIST
In 2009, during the global financial crisis, which came with the deep recession of part one, output was 4.7 percent below potential. That's about 3,100 dollars per person, in a single year.
HOST
So potential output is the most an economy can produce. Output can only be below it.
ECONOMIST
That's a common mistake. Potential means normal use, not the maximum. Output can rise above potential. In 1944, during World War Two, U.S. output was 23 percent above potential, because wartime production pushed people and machines past normal use.
HOST
Next leaf: policy without a parallel universe. Science fiction?
ECONOMIST
In a lab, a scientist can run a controlled experiment: one group gets a treatment and one group does not. Macroeconomic policy happens once, in one economy. Think of the U.S. stimulus program of about 800 billion dollars, during the global financial crisis. A fiscal stimulus is extra government spending or lower taxes, meant to support the economy.
HOST
Did it work?
ECONOMIST
Over the next two years, unemployment rose by more than the experts had predicted. Was it because the stimulus did not work? Or was the crisis worse than anyone knew? To answer, we would need a second United States, the same in every way but without the stimulus. Economists call that missing world the counterfactual: what would have happened otherwise.
HOST
We can never see that.
ECONOMIST
Never. But in a model, we can build both worlds. We run the model economy once with the stimulus and once without it, and we compare the outcomes. That's the experiment from part two. A model also states its assumptions openly, so others can check them. Models don't end policy debates, but they make the disagreements precise.
HOST
One last idea for this part: small growth differences compound. Does half a percentage point a year really matter?
ECONOMIST
Think of Argentina again. Since 1870, its income per person has grown about 1.4 percent a year, compared with about 1.9 percent in the United States. That's roughly half a point less each year, and that half point compounds too.
HOST
So how did Argentina end up?
ECONOMIST
In 1870, its income per person was about half the U.S. level. By 2022, it was less than a third. Over the long run, growth matters far more for living standards than the rises and falls of the short run.
HOST
So in part three we saw two time scales. In the long run, growth compounds, and income doubles every few decades. In the short run, output moves around potential. And since policy has no second universe, we test it in models.
ECONOMIST
Exactly. Let's sum up.
Wrap-up
HOST
Let's go back to our three goals. First, the four facts: income gaps, job losses in recessions, inflation in waves, and unemployment gaps. But why can the share of adults with a job fall for two different reasons?
ECONOMIST
Because it's the share of adults in the labor force, times the share of that labor force with a job. So it falls when people lose jobs and keep looking, or when they stop looking.
HOST
Second, the model. I'd say parameters and exogenous variables go in, and endogenous variables come out. And a minimum wage above the market wage creates unemployment.
ECONOMIST
Yes. In our example, each dollar of minimum wage above the market wage adds 3 hours of unemployment.
HOST
Third, the long run is growth over decades, and the rule of 70 gives its doubling time: about 37 years at 1.9 percent. The short run is the rises and falls around potential output, which the output gap measures.
ECONOMIST
Yes. So here is the big message. Macroeconomics turns big facts into small models. We choose the inputs, the model works out the outcomes, and we run the experiments that the real economy cannot run for us.
HOST
Here's a question to think about. Suppose Islandia ends its minimum wage. Instead, the government pays Harbor Boatworks 6 dollars for every hour of work it hires. What happens to the wage, to the hours worked and to unemployment? And who pays?
ECONOMIST
Pause the episode and think about it for a moment.
HOST
Okay, we're back. What's the answer?
ECONOMIST
Each dollar off the cost of an hour makes the boatworks want one more hour. So 6 dollars off means 6 more hours at every wage. That's a demand shift: the gap at a zero wage grows from 30 to 36 hours.
HOST
So the wage is 36 divided by 3, which is 12 dollars?
ECONOMIST
Yes, the same as with the minimum wage. But hours worked rise to 54, instead of falling to 48. And with no floor, nobody is left unemployed.
HOST
So who pays?
ECONOMIST
Taxpayers: 6 dollars for each of the 54 hours. With a minimum wage, some workers and the firm pay instead.
HOST
Where can students try all this themselves?
ECONOMIST
Open the chapter's Colab notebook to work with the data and the model yourself. The Model Apps run the labor market with sliders and no code, and the Data Apps show this episode's charts for many more countries.
HOST
And next time?
ECONOMIST
In Chapter 2, we look closely at GDP, the number behind most of these facts. We ask how we count a whole economy, and how we split its growth into higher prices and more goods.
HOST
That's it for today. Thanks for listening, and see you next time.