Economists cannot forecast recessions. Back in 2000 Prakash Loungani, an International Monetary Fund (IMF) economist, said their “record of failure to predict recessions is virtually unblemished”. And this has remained largely the case since: economists did not see the 2008-09 recession coming until it was on top of them.
One reason for this is that recessions are not always caused by the sort of events that macroeconomists focus upon – things such as rising interest rates or commodity prices. Instead, some of them have microeconomic causes. Harvard University’s Xavier Gabaix has shown that they can arise from failure of specific firms. The 2008-09 slump fitted this pattern: it was caused by failures in a handful of banks. So, too, did the previous recession: Paul Gregg and Paul Geroski showed that in the 1989-92 downturn a mere 10 per cent of firms accounted for 83 per cent of the fall in sales. That’s more consistent with recessions being due to failures at a few companies rather than to economy-wide troubles.
The lesson here is clear: if we want to be on guard against the risk of recession we need to do more than look only at macroeconomic data. We must look for industry-specific problems.
One candidate here is the car industry. Honda’s decision to close its Swindon plant reflects widespread troubles in the industry such as a Europe-wide fall in sales, especially of diesel cars, and the difficulties of some companies in adapting to electrification. These have contributed to a 4.6 per cent drop in industrial production generally in Germany since May.
In itself, of course, the industry is tiny: even in Germany it accounts for less than 2 per cent of total employment. But spill-over effects matter. As Daron Acemoglu has shown, recessions arise from network effects – when troubles in one firm reduce demand for others’ products. Electricity generation represents less than 2 per cent of UK GDP, but if it were to cease GDP would fall by considerably more than 2 per cent.
The issue here is not merely that the closure or shrinkage of one firm reduces demand down the supply chain – an important part of which is the reduced consumer spending of laid-off workers. Business confidence also matters. Christopher Carroll at Johns Hopkins University has shown that this spreads through an economy much like diseases spread through populations. Pessimism is infectious. If a big employer closes or shrinks, other firms even in different industries might ask why they should expand.
Also, remember, it’s not just the car industry that is in trouble. So too are retailers, which are struggling with the legacy of overexpansion, the rise of internet shopping and changing customer tastes. And the output of the construction and even pharmaceutical industries has fallen in recent months.
None of this is to say definitively that we are on the verge of recession. It’s just that macroeconomists' protestations that there won’t be a recession carry little weight, given their inability to forecast them. If you want a better picture of the chances of recession, you must look beyond macroeconomic aggregates towards particular key firms and industries. And the picture here isn’t especially pretty.
