Consumption shocks and debt
By Scott Sumner
Before discussing consumption, let me illustrate the general concept with an analogy. Suppose you have a population that experiences a major increase in life expectancy. Should that create a problem for public pensions? Many people would say “yes”, but it’s not obvious why. In general, improvements in life expectancy go hand in hand with improvements in the number of healthy years of life. So if your policy is for people to work 80% of their expected adult life, and then retire, it doesn’t matter whether life expectancy is 80, 160, or 800 years.
On the other hand, if the policy if for people to retire at age 65, then increased longevity does create a fiscal problem. But that’s not because of the longevity itself, rather it’s due to the arbitrary decision to fix the retirement age at 65, even as life expectancy increases.
Now suppose that an epidemic forces the shutdown of 20% of the economy. For simplicity, I’ll assume all economic activity is consumption, but the broader points will still hold in a more complex model.
As you may recall from EC101, aggregate output equals aggregate income. If aggregate output falls by 20%, then so does aggregate income. In a pure consumption economy that’s not a problem at all. If consumption falls hand in hand with falling income, then people with 20% less income are still able to consume the current consumption basket, which is 20% less costly than before the epidemic.
Of course the impact of an epidemic is very uneven, and some people are hit much harder than others. But in aggregate, people still have enough income to buy the prevailing consumption basket. Perhaps the government might want to redistribute that income somewhat, in order for the unemployed to be able to consume, but there’s no aggregate shortage of income.
In the US, we are responding to the epidemic as if it is essential to prevent people’s aggregate income from falling by 20%, even though their consumption expenditures are falling sharply. Thus while state and local governments take in far less sales and income tax revenue; they continue to pay roughly the same nominal salaries to public employees. Unemployment insurance has also been made much more generous. Lucky private sector employees like me have an unchanged nominal wage, even though my expenses have plummeted sharply.
Governments at all levels have decided to prevent a 20% fall in incomes via massive borrowing. That may be the way to go, but the motivation for this decision is not obvious. So let’s consider some possibilities:
1. Fairness: The impact of Covid-19 is very uneven, and it’s not fair for those hurt by epidemic to suffer disproportionately. I have sympathy for this argument. But on close inspection it calls for redistribution, not borrowing. Pay teachers and police and firefighters 20% less, and give that money to the unemployed. Tax people like me to provide aid for the unemployed.
2. Stimulus: All this borrowing helps to stimulate the economy. But recall that we are assuming that 20% of the economy must shut down for health reasons, so what good does stimulus do? Perhaps the goal is to provide stimulus for when the economy re-opens, but that’s done much more effectively via money creation, which doesn’t impose a debt burden on future taxpayers.
In my view, the real reason for this orgy of borrowing, this attempt to make everyone “whole”, is fairness combined with complexity and money illusion. The government naturally wants to avoid a situation where millions are suffering greatly from a steep drop in income. In theory, this could be accommodated in the way I describe above, but the redistribution would be highly complex and difficult to ramp up quickly. Furthermore, because of money illusion, the public is not keen on 20% pay cuts even if their “cost of living the way we live now” has fallen by 20%.
A policeman or teacher thinks he or she is just as productive as before, even though Covid-19 makes them far less productive. People have trouble grasping this concept, so an international example might help. In Pakistan, barbers do roughly the same job as in America, in a physical sense. But while an American barber is embedded in a highly productive economy, and thus might make $15/hour, a Pakistani barber is embedded in a low productivity economy, and thus might make $1/hour (or $3/hour adjusted for price level differences.).
The coronavirus epidemic makes America less productive, a little bit more like Pakistan. Because labor can flow between markets, people move at the margin until wages get equalized, adjusted for difference in skill levels, work environment, etc. Thus over the years, people in America left low paying jobs like cutting hair until wages rose from $1/hour to $15/hour, at which point alternative jobs no longer seemed so attractive. Pakistani barbers don’t have that $15/hour alternative.
Because most Americans don’t know why barbers here make 15 times more than in Pakistan, they are not likely to be happy seeing their wage fall when the productivity of the American economy falls sharply. And it’s not just barbers; teachers, police, and other jobs also pay for more in America than in Pakistan, much more than could be explained by differences in physical productivity. It’s the privilege of being embedded in a highly productive economy.
I’m trying to steer clear of normative claims, and focus instead on why this is all happening. It’s perfectly fine to argue that we should add trillions of dollars in debt in this sort of crisis. But it’s not true that we need to do so in order to maintain adequate levels of consumption. The epidemic is causing a big drop in both aggregate income and consumption, and debt doesn’t change that fact. The debt is being created for a complex mix of other reasons, of which the general public is largely unaware. Future generations will have to deal with the consequences.
PS. Just to be clear, I do support maintaining stable growth in nominal GDP, in order to help stabilize credit markets. This post is discussing real income and real output.
PPS. The standard model says the budget deficit should rise during recessions, in order to smooth out tax rates (and thus the deadweight loss from taxation) over time.