US households should be resilient

  • There is currently much (justified) concern about a possible US recession
  • If there is one, it is likely to be short and shallow
  • Because US households, contrary to perception, are not overburdened with debt

Will there, won’t there be a recession in the US? Some signs are certainly pointing to it; and the Fed’s new-found aggressive anti-inflation rhetoric and policy are adding to these concerns. On the other hand, the National Bureau of Economic Research (NBER), the official arbiter of the US business cycle, has hinted that even if GDP contracts in two consecutive quarters (e.g., Q1 and Q2 2022), that will not necessarily mean that there really was a recession. 

But, in one sense, the question of whether there is an ‘official’ recession or not, is irrelevant. What is more important is how resilient the US consumer is, and therefore, how prolonged, or otherwise, the downturn will be, and what the eventual recovery will be like. And here, there are some good news.

One reason to fear a recession is because the Fed is raising interest rates to combat inflation. The impact of higher interest rates depends partly on how indebted households (and companies) are. The Great Recession of 2007/08 was so severe, partly because households entered into it loaded with excessive debt. But, this is not currently the case. That means that households should be able to withstand higher interest rates.

This mildly optimistic view is based on two series of data. One is the Federal Reserve’s series on the Financial Obligations Ratio (FOR) of households. This shows that in Q1 2022, US households’ debt service costs amounted to 9.52% of disposable income. This is above the 8.38% seen in Q1 2021; but that was the lowest figure since at least 1980, when the series starts. The average for the entire period 1980-2022 is 11.11%; in 2007, it was above 13%. So around 9.5% is not too bad.

Of course, this comes after an extended period of very low interest rates, so it might be misleading. However, there is another measure which helps to reassure. This is household debt. In the wake of the Great Recession, I did some calculations on how much debt US households can ‘safely’ take on. Without going into details, it seemed that under ‘normal’ interest rates, US households could live with debt in the interval 100-110% of personal disposable income (PDI). Above that level, problems would appear (in early 2008, the level was just under 130% of PDI after five years above 100%). Since 2018, the debt total has been below 100% of PDI; in Q1 2022, it had just crept back up to exactly 100%. Higher interest rates and cautious banks mean that, although bank lending is currently growing strongly, debt is unlikely to grow very much over the coming year. Moreover, current inflation will erode part of the real value of the debt.

This means that when the policy cycle turns, US households should be poised to return to strong consumption, with healthy balance sheets. If so, the recovery from the downturn, whether a recession or not, could actually be quite strong. That being said, the very near-term outlook remains bleak.

P.S. The standard definition of a recession is two consecutive quarters of real GDP falling. But this is, like so many other standard definitions, both silly and arbitrary. It is useful, because it is easy to grasp and to show. But it really only makes sense if a country’s trend growth is sufficiently above zero for a contraction to be meaningful. This is in many advanced countries no longer the case. Perhaps the best example is Italy. The Italian economy’s trend growth rate is probably around zero. In other words, Italian GDP is as likely to contract as it is to expand. But if that is the case, then proclaiming a recession each time the economy contracts twice in a row, becomes meaningless. It would be far better to define a recession as, say, growth significantly below trend for a period of time. How significant? Perhaps more than one standard deviation? But trying to explain that is probably doomed. So we’re stuck with the current, unsatisfactory definition.