The busy recession?

It is Fed Day (again) and the members of the Fed Open Market Committee have to make up their minds whether to continue to hike interest rates and fight inflation or not. As I write this, the market is split about a rate hike, pricing in roughly a 50/50 chance at the meeting today. One of the things the Fed decision makers will look at is the jobs data which has been extremely resilient so far. Could we really see a slowing US economy and a recession without weakness in the job market?

The recovery after the 2001 recession in the US is sometimes called a jobless recovery because economic growth accelerated after the recession, but job creation was very low. This contributed to the Fed keeping interest rates too low for too long and helped fuel the real estate bubble that came back to haunt us in 2008.

Now it seems we may be heading into the opposite situation, something I would call a ‘busy recession’ where the economy slows down and even shrinks but jobs remain plentiful. A recent paper by Jose Asturiaz and his colleagues showed that business applications may be a good leading indicator for job creation. It is widely known that small businesses create collectively more jobs new jobs each year than large ones, so if people start more new businesses, one expects to see more jobs being created a little later.

But not all new businesses are created equal. Many business applications are for companies that then lay dormant for a while as a shell in case someone wants to start an active business in the future. This is why it is helpful to look at a little-known dataset from the Bureau of Labor Statistics that counts so-called high-propensity business applications. These only count applications for new businesses that are likely to result in active businesses.

The chart below shows the annual change in high-propensity business applications moved ahead by 12 months together with the annual change in nonfarm payrolls. If we ignore the shock to nonfarm payrolls due to the pandemic in early 2020, we can see that the fit is remarkable. Business applications were able to predict changes in nonfarm payrolls a year ahead both during times of rising and declining employment. And while there was a bit of a soft patch in business application a year ago, the chart indicates that we should have seen much weaker employment data by now. In fact, looking ahead to the next 12 months, the chart implies that we might see jobs growing by some 3% over the next twelve months. Outside of the pandemic, that would be jobs growth we have last seen in 2005/2006.

Business applications indicate a strong labour market in the next 12 months

Source: Bureau of Labor Statistics

Now, I would take this chart with a grain of salt since it is just one indicator and other indicators point to a decent chance of job market weakness ahead, but just like the jobless recovery in the early 2000s contributed to policy mistakes by the Fed, so too may a busy recession.

The Fed will be watching not just inflation data but also labour market data to assess if they have to cut interest rates or if they can keep them high without creating a recession. If we really get a busy recession, the Fed may keep interest rates too high for too long, thus creating long-lasting damage to the economy by curtailing investments. Let’s hope it will not come to that.