Opening
HOST
Welcome back. Today we are on consumption: what households spend. It's more than two thirds of GDP, the total value of what a country produces in a year. And it holds a puzzle. In the years before the financial crisis of 2008, American households saved very little from their income, and their debts kept climbing. Yet their wealth, compared with their income, kept climbing too. Were they careless with money?
ECONOMIST
Or were they planning for a whole life? So here is today's question. How do households decide how much to spend today and how much to save for the future?
HOST
And what will we be able to do by the end?
ECONOMIST
By the end of this episode, you'll be able to do three things. First, explain why households spread their spending evenly over life, and work out how much a student spends and borrows. Second, use the Euler equation to explain why interest rates follow growth, and predict how spending today changes with a windfall or a risky future. Third, describe what the data since 1970 show about debt, saving and wealth, and what they mean for stimulus checks and tax cuts.
HOST
And we'll follow the chapter's mind map part by part?
ECONOMIST
Yes, in three parts: a lifetime budget, then the Euler equation and its lessons, then debt, saving and wealth in the data. We take the leaves, the small topics of each part, in order.
A lifetime budget
HOST
Part one of the map: A lifetime budget. The first leaf is total wealth, financial plus human wealth. Wealth sounds like something only rich people have.
ECONOMIST
Everyone has some. Financial wealth is what you own today, like money in the bank, stocks or bonds. Human wealth is the value today of all the wages you will earn.
HOST
The value today? A wage I earn in ten years is still a wage.
ECONOMIST
But a dollar later is worth less than a dollar now, because a dollar now can earn interest. Say the real interest rate, the rate after inflation, is 5 percent. One dollar today grows to 1.05 dollars later. So one dollar is the present value of 1.05 dollars later: what that later money is worth today.
HOST
Let's make it concrete.
ECONOMIST
Meet this chapter's islander. Irving is a young student on Islandia, our imaginary island country. Irving has no savings. He earns 1 unit now and expects to earn 5 units later, once he works. Think of one unit as 10,000 dollars.
HOST
So his total wealth is 1 plus 5, which is 6.
ECONOMIST
Only if we ignore interest, as the map's example does to keep things simple. With interest, his later 5 units would be worth a bit less today.
HOST
Next leaf: Irving earns 1, spends 3, and borrows 2. Spending three times his income? That sounds unwise.
ECONOMIST
A lot of people think spending should follow this year's income: low income now, low spending now. But in this model, spending follows total wealth. Irving plans for his whole life.
HOST
So how does he get to 3?
ECONOMIST
Let's build it step by step. His life has two periods, today and later. He cares about both equally, and he wants to spend the same amount in each. Why the same? That's the next two leaves. So his total wealth of 6, split evenly, gives 3 for each period.
HOST
But he only earns 1 today.
ECONOMIST
So he borrows the difference, 2 units. Later he earns 5, repays the 2, and has 3 left, the same as today. This works only if someone will lend to him. Part two shows what happens if no one will.
HOST
Next leaf is a formula: utility equals the log of consumption. What's utility?
ECONOMIST
Utility is the happiness people get from consuming. The log, short for logarithm, gives a curve that always rises, but more and more slowly. Marginal utility is the extra happiness from a little more spending. Diminishing marginal utility means it shrinks as you spend more. Your first meal of the day matters a lot. A fifth meal adds very little.
HOST
And the second half of the formula?
ECONOMIST
With the log, marginal utility is one divided by what you already consume. So at 5 units of spending, an extra dollar brings only one fifth of the happiness it brings at 1 unit.
HOST
So if Irving just spent his income, a dollar today would be worth five times a dollar later. Then he should move as many dollars to today as he possibly can!
ECONOMIST
That's the tempting step, but not quite. Each dollar he moves raises his spending today and lowers it later. So the next dollar today is worth a bit less, and a dollar later a bit more. He stops when the last dollar is worth the same in both periods. With no interest, that's at 3 and 3.
HOST
Next leaf: smoothing wins, 1.10 versus 0.80. What's being compared?
ECONOMIST
Two plans that cost the same. In the uneven plan, Irving spends his income as it comes. In the smooth plan, he spends 3 in each period. With the log formula, the smooth plan gives a utility of about 1.10 per period. The uneven plan gives only 0.80 on average.
HOST
Same money, more happiness. Why?
ECONOMIST
Because the curve bends. Going down from 3 to 1 costs a lot of happiness, since the curve is steep at low spending. Going up from 3 to 5 adds less, since it's flatter there. So the two units he'd lose today hurt more than the two units he'd gain later help. That's why people smooth consumption: they keep spending steady while income goes up and down.
HOST
So in part one, Irving's budget covers his whole life, even wages he hasn't earned yet. And because each extra dollar adds less happiness, a steady 3 and 3 beats spending income as it comes.
ECONOMIST
Exactly. Next branch: the rule behind that even split, and its lessons.
The Euler equation and its lessons
HOST
Part two of the map: The Euler equation and its lessons. The first leaf is the Euler equation itself. What does it say?
ECONOMIST
Try a small test. Irving spends one dollar less today and saves it. Today he loses the happiness that dollar would bring. Later, the dollar comes back with interest and brings him happiness then. But happiness later may count a little less for him, depending on his patience.
HOST
Patience?
ECONOMIST
How much he counts happiness later, compared with happiness now. A patience of 1 means he counts both the same. A patience of 0.9 means future happiness counts only 90 percent as much.
HOST
So when is his plan the best one?
ECONOMIST
If the saved dollar brings more happiness later than it costs today, he should save more. If it brings less, he should borrow and spend more. His plan is best only when the happiness lost today equals the happiness gained later. That's the Euler equation.
HOST
Next leaf: spending today is total wealth divided by the sum of one and patience. Half, when patience is 1.
ECONOMIST
With the log formula, add one to his patience, and divide his total wealth by that number. That's what he spends today. A more patient Irving cares more about later, so he keeps more for later and spends a smaller share today. With a patience of 1, he divides by two: half of 6 is 3, the plan from part one.
HOST
Next leaf: interest rates track growth, that is, move with it. What does Irving's spending have to do with interest rates?
ECONOMIST
For one person, the interest rate is fixed: he can't change it. He only chooses how fast his spending grows. For a whole economy, it's the other way round. Growth models, like the one in Chapter 6, set how fast spending grows. The Euler equation then tells us the real interest rate at which households are willing to spend along that path.
HOST
Can we build that interest rate with numbers?
ECONOMIST
Let's do it step by step. Suppose spending per person grows 2 percent a year, and patience is 1. So next year, Irving spends 2 percent more. Marginal utility is one divided by spending, so next year an extra dollar brings about 2 percent less happiness.
HOST
Then why would he save at all?
ECONOMIST
Only if a saved dollar grows by about 2 percent to make up for that. So the interest rate must be 2 percent, the growth rate.
HOST
And with a patience of 0.99, a little impatient?
ECONOMIST
Then he counts next year's happiness at 99 percent of today's: about 1 percent less. Saving must make up for both, about 2 percent for higher spending and about 1 percent for impatience.
HOST
So about 3 percent?
ECONOMIST
Yes. To be exact, take one plus the growth rate, 1.02, and divide it by his patience, 0.99. That gives about 1.03, which is one plus the interest rate. So the interest rate is about 3.0 percent. When patience is close to one, the extra for impatience is small. That's why real interest rates and growth rates are numbers of a similar size, a few percent a year.
HOST
Next leaf: the MPC, one divided by the number of periods if you can borrow, and 1 if you cannot. What's the MPC?
ECONOMIST
The marginal propensity to consume: the share of an extra dollar that a household spends right away. Say Irving gets 100 extra dollars today, a windfall, money he didn't expect.
HOST
Easy. He's a student, so he spends nearly all of it.
ECONOMIST
That's the natural guess, but if he can borrow, the model says he spends only 50 dollars. The windfall adds 100 dollars to his total wealth, and he spends half of his total wealth today. So his MPC is one half.
HOST
And over a longer life?
ECONOMIST
Then the share is one divided by the number of periods left. With 50 years ahead, 100 extra dollars raise his spending by only about 2 dollars a year.
HOST
And if he can't borrow?
ECONOMIST
That's the second half of the leaf: an MPC of 1. Irving wants to spend 3 units, but he earns only 1 unit today. If no bank will lend to him, he spends what he earns, so any extra dollar he gets today is spent at once.
HOST
Last leaf of this part: precautionary saving, from 3 to 2.44. Saving as a precaution?
ECONOMIST
Yes, saving for bad times. So far, Irving knew his future income for sure: 5 units. Now suppose it will be 3 or 7, with equal chances. On average, it's still 5.
HOST
Then nothing changes. The average is the same, so he still spends 3 today.
ECONOMIST
That's the tempting answer, but he spends only about 2.44.
HOST
Why less?
ECONOMIST
Look at the bad case, an income of only 3. If he spent 3 today, he'd owe 2, and have just 1 left to live on.
HOST
That's painful. And the good case?
ECONOMIST
He'd have 5 left. But a bad year hurts more than a good year of the same size helps, because an extra dollar matters most when spending is low. So he protects himself by borrowing less. That pulls his spending from 3 down toward his income today, as for a household that can't borrow: his spending stays closer to the cash he has now.
HOST
So in part two, one comparison, a dollar spent today against a dollar saved, taught us every lesson. If you can borrow and have a long life ahead, you save most of a windfall, and interest rates follow growth.
ECONOMIST
But a household that can't borrow spends every extra dollar it gets. And households with little cash that fear a bad year act much like it. Next branch: what real households did.
Debt, saving and wealth in the data
HOST
Part three of the map: Debt, saving and wealth in the data. The first leaf is debt: 98 percent of GDP in the first quarter of 2008, that is, its first three months. Whose debt is that?
ECONOMIST
The debt of American households, mostly mortgages, the loans people take out to buy a home. In the 1970s, it was less than half of GDP. It climbed from the mid-1980s, and fastest in the housing boom of the 2000s. At the peak, it was almost a full year of the country's output.
HOST
And after the peak?
ECONOMIST
After the crisis, households cut their debts, by paying them down or by failing to repay. Cutting debt is called deleveraging.
HOST
Next leaf: saving. It fell from 12.2 percent in the 1970s to 2.5 percent.
ECONOMIST
That's the personal saving rate: the share of their disposable income, their income after taxes, that households save. So in the 1970s, households saved about 12 cents of every dollar. From 2005 to 2007, they saved two and a half cents. After the crisis, saving went back up, partly as a precaution: many people feared for their jobs.
HOST
Next leaf: wealth. I can guess this one. Less saving and more debt, so households must have become poorer.
ECONOMIST
A lot of students think that, but household wealth grew. Wealth here means net worth: everything households own, like homes, stocks and bank deposits, minus everything they owe. In 1970, households owned wealth worth about 5 years of their disposable income. In the second quarter of 2026, it was worth 8.3 years, a record high.
HOST
How can wealth grow when people save so little?
ECONOMIST
In two ways. You can put aside part of your income, or the things you already own can go up in price. Stock prices climbed fast in the 1990s, and house prices in the 2000s. Saving is what you put aside each year. Wealth is what you own minus what you owe, at one moment.
HOST
So that answers our opening puzzle?
ECONOMIST
Partly. A leading explanation is that rising stock and house prices made households feel richer, so they saved less from their income. In the data, saving tended to be low when wealth was high, but that link alone doesn't prove the cause.
HOST
Next leaf: saving across countries, China and Japan. Let me use Irving's logic. China has grown very fast for decades. Households that expect to be much richer should borrow now and repay from future income, like Irving. So I predict Chinese households save very little.
ECONOMIST
The simple model says so, but the data say the opposite. In 2023, Chinese households saved 34.9 percent of their disposable income, more than a third. Economists debate why. One possible reason is precautionary saving: when you must pay for health care and education yourself, with little help from the state, you save more as protection.
HOST
And Japan?
ECONOMIST
Japan fits the life-cycle idea better: people save while they work and spend their savings in retirement. As Japan aged, more people were retired, so its household saving rate fell, to only 0.7 percent in 2024.
HOST
Last leaf: policy, with stimulus checks, tax timing and retirement saving. In a recession, governments sometimes mail checks to households to raise spending. From part two, households that can't borrow should spend the most. Is there evidence?
ECONOMIST
Yes, from the United States in 2001. The law passed in June, and weeks later most taxpayers got a tax rebate check, money back from the government.
HOST
And what did they find?
ECONOMIST
Within three months, households spent 20 to 40 percent of the check on everyday goods, like food. For a household that can borrow, the model predicts far less.
HOST
Far less? Irving spent half of his windfall.
ECONOMIST
His windfall was a surprise, and his life had only two periods. These checks were known weeks ahead, so in the model their arrival is no news and shouldn't change spending. And with many years ahead, a household spends only about 2 dollars of each 100 in a year. Households with little cash spent the most, as if they couldn't borrow. Wealthier ones smooth much as the model says, and the average share spent is still debated. But for policy, model and evidence agree: a check raises spending most when it reaches households with little cash.
HOST
Next, tax timing. What if the Islandia Treasury, the government's money office, cuts taxes today and borrows to pay for it?
ECONOMIST
It must repay the loan with interest, so taxes rise later. Say the tax cut is 1,000 dollars, at an interest rate of 5 percent. Later, taxes rise by 1,050 dollars, which is worth exactly the 1,000 today. So total wealth doesn't change, and an islander who can borrow saves the whole tax cut to pay the later tax.
HOST
So for that islander, the timing of taxes doesn't matter.
ECONOMIST
Right, as long as government spending doesn't change. That's Ricardian equivalence, after the economist David Ricardo. But it fails for an islander who can't borrow, like Irving when no bank will lend to him: he spends the whole tax cut. It also fails when the later taxes fall on other people, like future generations.
HOST
And retirement saving?
ECONOMIST
In the model, the default, the option you get if you do nothing, doesn't matter. In real life, people tend to stay where they start. With automatic enrollment, almost no new worker leaves a firm's retirement plan. If workers must sign up themselves, fewer than half join in their first year.
HOST
So in part three, the model fits some facts, like Japan's retirees and saving that fell as wealth grew. But China saves far more than predicted, and many households spend a check when it arrives.
ECONOMIST
That's a fair summary. Let's sum up.
Wrap-up
HOST
Let's go back to our three goals. First, why does a household want smooth spending, and how much does a student spend?
ECONOMIST
Each extra dollar adds less happiness, so a dollar is worth most when spending is low. So Irving spreads his total wealth of 6 evenly: he spends 3 today and borrows 2.
HOST
Second, interest rates, windfalls and risk.
ECONOMIST
The interest rate must make up for rising spending and a little impatience: about 3 percent when spending grows 2 percent. If you can borrow, you spend only part of a windfall. If you can't, you spend all of it. And a risky future makes you spend less today.
HOST
And third, the data.
ECONOMIST
Saving fell and debt climbed, yet wealth grew. And checks were spent most by households with little cash. So here is the big message. Households spend out of their lifetime wealth, not this year's income, so they smooth. In the model, with many years ahead, they save most of a windfall, and interest rates follow growth. But borrowing limits and risk make many households with little cash spend almost every extra dollar. That's why the timing of taxes and of government payments, like stimulus checks, can matter.
HOST
Here's a question to think about. Islandia sets up unemployment insurance, which pays islanders who lose their jobs. Irving's future income is still 5 on average, but now it can only be one unit above or below that, not two. Does he spend more or less today than the 2.44 we found? And why?
ECONOMIST
Pause the episode and think about it for a moment.
HOST
Okay, we're back. What's the answer?
ECONOMIST
He spends more, about 2.84, but still less than 3. The bad year is now less bad, so he needs less protection. Some risk is left, so he still saves a little extra. The insurance raises his spending today, though his average income hasn't changed.
HOST
Where can students try all this themselves?
ECONOMIST
Open the chapter's Colab notebook to run Irving's plan yourself. The Model Apps let you move sliders on the utility and smoothing charts without any code. And the Data Apps show debt, saving and wealth for dozens of countries.
HOST
And next time?
ECONOMIST
In Chapter 17, we turn to investment, the other big private spending decision. We ask: when is a machine, a share or a house worth buying, and why does investment swing so much?
HOST
That's it for today. Thanks for listening, and see you next time.