US recession ahead

I can’t share professional work here because these missives go out to a broad audience, many of which are not paying for my research published behind a paywall (which begs the question, why aren’t you paying for research?). As you might imagine, questions about the possibility of recession and the path forward for inflation are the key topics everybody wants to discuss with me. I make no secret of the fact that I think a recession in the US is all but inevitable, but when it starts and how deep and long-lasting it will be, remains an open question.

This is where a simple analysis by Edward Leamer from UCLA comes in handy. He looked at the development of several key macroeconomic indicators in the three years before the onset of a recession and compared it with the development of these indicators in 2021 and 2022. I will focus on his results based on the data for the first half of 2022, but you can download the full results here.

Leamer looked at the development in US interest rates, inflation, the unemployment rate, and the housing market and estimated the probability that the US will enter recession in the next 12, 24 and 36 months. The table below shows a summary of his findings.

Probability for US recession

Source: Leamer (2022)

In general, the probability that a recession in the US will start in the next 12 months is very high. Inflation data would suggest that it is all but inevitable. Advocates of a soft landing claim that the job market is so strong that a recession is by no means certain and may likely never come. My response to that is and always has been that we learn in Economics 101 that the labour market is a lagging indicator. If you wait for the labour market to show you a recession, you will be 3 to 6 months late. Indeed, Leamer’s analysis suggests that current strong labour market data is perfectly in line with the onset of a recession in the next 12 months. Better not believe the Fed and other economists when they point to employment data as a sign of an economy far away from recession.

Also, housing starts have weakened so much, that we have now roughly a one in three chance of seeing a recession in the next 12 months based on that data alone. But Leamer in his paper also shows that if we look at housing starts not just in the first half of 2022 but in the last 12 months, the probability for the onset of a recession in the next 12 months rises to 77%.

In the end, the only indicator that points to a recession in late 2023 is the steepness of the yield curve which is the most reliable of them all. As I have written here, an inversion of the yield curve is a remarkably good indicator for a recession some 18 months down the road. And the yield curve has inverted in the US since in spring. However, I would argue that this time around, the lead time from the yield curve inversion to a recession is likely shorter than normal, because short-term interest rates were extremely low at the beginning of this year and the Fed needed to hike rates a couple of times before it would move short-term bond yields even close to the 10-year Treasury yield.

To me at least, the data of Edward Leamer confirms my suspicion that the US will enter recession in the next 12 months and possibly already in 2022. Equity markets certainly seem to have that priced in and in my view, they are correct.