Unemployment is rising. Economists expect the official unemployment rate to be almost 8 per cent by December – twice what it was before the pandemic: yes, there’s huge uncertainty around this, but few doubt the direction of change.
This, however, might ultimately be good for equities. I say so because of a simple statistic. There has been a significant positive correlation between the jobless rate and returns on the All-Share index in the following 12 months – of 0.35 since 1985. High unemployment in 1986, 1992 and 2009 led to shares rising nicely, for example, while low unemployment in 1989, 2000, 2007 and last year led to prices falling.
Mass unemployment, then, tends to lead to good equity returns. This isn’t because capitalists rejoice in the misery of workers. In fact, exactly the opposite. My chart shows the contemporaneous correlation between the dividend yield on the All-Share index and the jobless rate. It’s clear that for most of the past 35 years high unemployment has been accompanied by a high yield (that is, low share prices) and low unemployment by a low yield – that is, high prices. Granted, this correlation seems to have broken down after 2015. But I suspect it has recently re-emerged. The latest data we have on unemployment is for February-April, but the rate is almost certainly higher now.
What’s going on here is quite simple. It’s that share prices overreact. In bad times (when unemployment is high) they fall too far causing the dividend yield to be high when unemployment is high. From such low levels, prices subsequently recover, which means that high unemployment leads to shares rising (the opposite happens in good times).
Exactly why the market overreacts is, however, unclear.
One possibility is that investors are slightly irrational. They extrapolate too much from current economic conditions, believing that these will persist for longer than they in fact do. That’s the recency bias. They also fail to foresee that the subsequent recovery will raise their appetite for risk and demand for equities – this is the projection bias.
It’s also possible, though, that investors are in fact being quite rational. Recessions are risky times. Not only is there a heightened chance of firms going to the wall, but even if they survive their prospects might be permanently worsened.
One reason for this is that recessions can reduce long-term growth. In creating spare capacity and depressing animal spirits (even years later) they might reduce capital spending and firms’ willingness to expand.
Also, recessions can accelerate creative destruction by changing patterns of supply and demand. Personally, I suspect some of the desire to 'build back better' after the pandemic is over is as credible as new year’s resolutions – but I don’t blame anybody for attaching some weight to them.
Such dangers justify an additional risk premium upon equities – which means a higher yield and rising prices later as compensation for taking on such extra risk.
It’s hard to tell which of these explanations is more valid. If we look at sectors’ returns since 2000 we see significant correlations between unemployment and subsequent returns on IT and media stocks. This is consistent with bad economic times depressing investor sentiment. But there are also significant correlations between unemployment and returns on cyclical sectors such as chemicals, general retailers and support services. This is consistent with bad times increasing economic risk (for banks, miners and oil stocks, however, there’s no link between the jobless rate and subsequent returns).
All this tells us that rising unemployment – while it happens – is bad for equity investors as well as workers. There is, though, a big difference between the two groups. For shareholders, higher unemployment eventually provides some nice opportunities. The same cannot be said for workers.
