And most of the time there isn’t; most of the time scarcity and opportunity cost – the cost of passing up the next best choice when taking a decision – is what economics is largely about. But not always. Sometimes, there are depressions.

When the economy is in a depression, scarcity ceases to rule. Productive resources sit idle, so that it is possible to have more of some things without having less of others; free lunches are all around. As a result, all the usual rules of economics are stood on their head; we enter a looking-glass world in which virtue is vice and prudence is folly. Thrift hurts our future prospects; sound money makes us poorer.

Moreover, that’s the kind of world we have been living in for the past several years, which means that it is a kind of world that students should understand.

The same cannot be said about depressions. Are they simply very bad recessions? Sometimes people do try to define depressions that way, offering arbitrary criteria such as a fall of 10 percent or more in output. But when we talk about the Great Depression we don’t just mean the plunge from 1929 to 1933, we mean the whole period from 1929 to 1939 or 1940, a period that included episodes of growth – in fact, growth at 8 percent a year between 1933 and 1937 – as well as periods of decline. What made it the Great Depression, as opposed to just two recessions plus two recoveries, was the fact that national economies were clearly operating far below capacity the whole time, even when they were expanding – and the fact that there did not seem to be an easy way to get out of the trap.

That last point is crucial. Garden-variety slumps end fairly quickly, in large part because there is a simple technocratic answer: central banks cut interest rates, and the economy pops back up. But during the Great Depression those banks couldn’t do that, because short-term rates – which are the rates conventional monetary policy affects most directly – were very close to zero, and could not go any lower. Hence the situation that John Maynard Keynes described and analyzed in his 1936 master-work The General Theory of Employment, Interest, and Money (Keynes 1936), in which the economy remained ‘in a chronic condition of subnormal activity for a considerable period without any marked tendency either towards recovery or towards complete collapse’. Japan entered that condition in the 1990s; America, Britain, and much of Europe joined it there in 2008.

So depressions should really be defined in functional terms: they are situations in which the economy is persistently below capacity but the usual medicine for that ailment, monetary expansion, is ineffective, because short-term interest rates are already near zero.

How do economies get into such situations? In the case of both Japan in the 1990s and the North Atlantic economies today, the story involved a large debt-financed real estate bubble that eventually burst, leaving both an overhang of excess construction and an overhang of debt, both of which exert a persistently depressing effect on the economy – sufficiently depressing that even a zero short-term interest rate isn’t low enough. But one should not get too hung up on the details of the boom that preceded the slump.

Actually, that is a general caution. Ever since people began noticing the phenomenon of business cycles, there has been a strong tendency to cast them as morality plays – to lavish attention on the excesses of the boom and assume, at least implicitly, that the slump that follows is a necessary consequence. Yet there is no good reason to believe that this is true – in fact, as we will see in this chapter, there is every reason to believe that it is not true, that persistent slumps can and should be avoided, no matter what went on in the previous boom. And the habit of seeing economic ups and downs in moral terms can get in the way of the right policies.

So, to return to the main line of argument – a depression, I have asserted, is a situation in which the economy is operating persistently below capacity, and in which ordinary, technocratic policy cannot restore full employment. Now, let’s look at the difference that such a situation makes to the usual rules of economics.

Suppose, for example, that you would like to see the government build more infrastructure – new roads, bridges, rail tunnels under New York’s Hudson River (that’s a personal sore spot, in case you haven’t guessed). Ordinarily we would say that doing this would require doing less of something else – either consumers must be persuaded to consume less, or the government spending would have to ‘crowd out’ private investment. If the economy is in a depression, however, no such trade-off is required: the tunnels may be built by workers who would otherwise be unemployed, using construction equipment that would otherwise have sat idle. If you ask where the funds to pay for this come from, well, they may be raised by selling government bonds – but this borrowing does not compete with other uses of funds, because the very act of spending government money raises income, some of that increased income is saved, and the spending therefore creates the savings that are used to buy those government bonds.

You may find this implausible; you may believe that large-scale government borrowing must surely drive up interest rates. If so, you have a lot of company: many prominent people, some of them believing that they know a lot about economics, insisted loudly back in 2009 that the large budget deficits that were already looming would lead to much higher interest rates. But they didn’t. I’m writing this chapter more than three years after some of those debates, and the interest rate on long-term US government bonds is less than 2 percent. We are living in an economy to which the usual rules do not apply.

Nor is that all. Depression economics is marked by paradoxes, in which seemingly virtuous actions have perverse, harmful effects. Two paradoxes in particular stand out: the paradox of thrift, in which the attempt to save more actually leads to the nation as a whole saving less, and the less-well-known paradox of flexibility, in which the willingness of workers to protect their jobs by accepting lower wages actually reduces total employment.

On the paradox of thrift: for the economy as a whole, savings are always equal to investment, as a matter of accounting. So suppose that some group in the economy, say households, becomes more future-oriented, more willing to defer present gratification and save for the future. Ordinarily we would expect this willingness to translate into higher overall investment spending, so that the economy would indeed make more provision for the future – building more factories, server farms, and office buildings in order to raise future productivity.

It is important, however, to understand how that process is supposed to work. How does a decision by John Q. Public, in Ho-Ho-Kus, New Jersey, to spend less and save more persuade Google to expand the Googleplex in Mountain View, California? The standard answer is that it works through interest rates: higher desired savings translate into lower rates, which either directly reduce borrowing costs or indirectly lead to higher stock prices; either way, the cost of capital falls for corporations considering expansion, and the result is more investment.

Now think about what happens in a depression. Interest rates, or at least short-term rates, cannot fall because they are already at zero. So the transmission mechanism from desired savings to investment is broken.

Instead, when John Q. Public cuts his spending, the result is just a fall in total spending – the economy becomes more depressed. This, in turn, will reduce, not increase, the amount that corporations are willing to invest, since there is less reason to expand capacity. Yet it is still true that savings equal investment in the aggregate, which means that overall savings fall. Hence the paradox of thrift: an attempt by some people to save more ends up reducing saving by the economy as a whole.

What about the paradox of flexibility? We normally imagine that the way to get people to buy more of something is to cut its price. So if there are not enough jobs, aren’t wage cuts the answer? Many people think so, and in fact many conservative historians assert that the Great Depression would have ended much sooner if the US President Franklin Roosevelt had not allowed unions to oppose wage cuts and win wage increases.

Even in normal times, however, the relationship between wages and employment doesn’t work in the way many imagine. It is true that workers in a particular industry or company can save their jobs by taking wage cuts. That’s because by reducing wages their labor and the products they produce are made cheaper in relation to the labor and products of other workers. When the overall level of wages falls, however, no one gains a relative advantage. If there is any positive effect on employment, it happens via interest rates. In practice, the way this works tends to be through policy at the central bank: lower wages mean lower inflation, which encourages the central bank to cut interest rates, leading to higher demand and hence higher employment.

In a depression, however, interest rates cannot be cut. So there is no channel through which lower wages can raise employment! And the likely effect of wages cuts is in fact to reduce employment.

Why? Debt. Recall that the overhang of debt is probably one major reason that we’re in a depression right now. Some 80 years ago, the great American economist Irving Fisher explained how this works (Fisher 1933). When individuals have run up debts that are now considered excessive, these debtors end up being forced to slash spending in order to pay that debt. Meanwhile, creditors face no comparable pressure to spend more. So debt that is seen as excessive creates a ‘deleveraging’ environment in which overall spending is depressed, possibly depressed enough to ...