The western world is heading into recession while interest rates rise. As two-headed monsters go, that’s about as scary as it gets for investors. While recession sucks value from their equities, rising interest rates do the same for their fixed-coupon bonds. Then the ensuing stagflation feeds a vicious spiral that drags down values still further. Welcome to the bear market just around the corner.
Yet it is worth asking whether a recession is all but certain and whether its investment concomitant – a bear market – is all but present. True, on bad days – for instance, the FTSE 100 index closed 2.3 per cent down the day before I wrote this – the manic depressives among the investment commentariat aren’t in any doubt. However, perhaps they are simply proof that Benjamin Graham’s ‘Mr Market’ is always with us, ready to grab at the extreme outcome if at all possible. So it might pay to be a bit circumspect and ask simple questions, such as: what are the chances of either or both a recession and a bear market?
Start with recession since that is more likely to lead to a bear market than the other way around. To help with perspective, let’s use a rule of thumb labelled the Sahm Rule, devised by a US economist, Claudia Sahm, a couple of years ago. Granted, this indicator was framed within the context of the US economy so it may work less well applied to UK data. Even so, it is a good way to get the thought process going.
The first thing to note, as the chart below indicates, is that a recession is a comparative rarity – where a recession is defined, conventionally but arbitrarily, as a quarter-on-quarter fall in the real value of gross domestic product (GDP) two quarters running. Starting in 1972, the UK’s economy has fallen into recession on only six occasions. Put another way, out of the 204 quarters since then, the economy has been in recession for just 21 quarters. In addition, recessions have become more infrequent – just two in the 30 years since 1992.
Less frequent, but possibly more to be feared; at least, the previous two have been the deepest of the six. UK GDP dropped 6 per cent in the wake of the US sub-prime crisis of 2008 over the five quarters to mid-2009. Much worse was the 21 per cent drop in output over the first two quarters of 2020 as the global economy cowered in fear of Covid-19. In contrast, the worst of the four recessions that hit in the final 30 years of the previous century took just 4.2 per cent off GDP over the five quarters to March 1981.
The Sahm Rule rests on the idea that a rise in an economy’s unemployment rate is a good predictor of impending recession if only because monthly unemployment data is released before quarterly GDP data. It matters little whether the unemployment rate is high or low; what counts is whether the rate is rising quickly. When it is, that’s a sign that economic activity is dropping off, which will work through to lower GDP before long.
Putting the idea into practice, the Sahm Rule says that, for any month’s data, if the three-month rolling average unemployment rate is at least 0.5 percentage points above its minimum level in the previous 12 months then the economy is already in recession. For the US, which has also had six recessions since 1972, the rule works well. It has a clean record on false positives (ie, it does not forecast recessions that don’t happen) and, more often than not, made its forecast just before the GDP data announced a recession.
The chart applies the rule to the UK where the shaded columns cover the six periods of recession; the straight horizontal line is the 0.5 percentage point rise in the unemployment rate that is the threshold of recession and the jagged line is the recession predictor (ie, the difference between the latest unemployment rate and the minimum rate over the past 12 months).
The rule’s overall record is acceptable although, as the chart shows, it did signal two false positives either side of the recession of 2008-09. Nor did it start off too well. It missed the short but deep recession triggered by the response of Middle Eastern oil producers to 1973’s Yom Kippur War. Then again, no one saw that one coming. But it timed the mild recession of mid 1975 nicely; as it did the recessions of 1980-81, 1990-91 and 2008-09.
Then it got caught out by the short but ultra-deep downturn shaped by Covid-19. That recession was over before the unemployment rate made its signal. It was not until July 2020 that the jobless rate rose to more than half a percentage point above its minimum rate for the previous 12 months, but in the third quarter of that year – when we were ‘eating out to help out’ – GDP recovered strongly.
True, there were exceptional circumstances that snarled the rule’s predictive capability, the major one being a state-subsidised job-protection scheme on a scale hitherto unknown. As a result, the UK’s unemployment rate, while on a brief upward trend, stayed low throughout 2020-21 and has been falling for the past 12 months. The latest data, the three-month average for December to February, show a rate of 3.8 per cent, which is actually the lowest the rate has been in the past 12 months. That does mean a low unemployment rate could still signal recession. The rate would only have to rise to 4.4 per cent – most recently seen in July 2021 – to make the signal.
But perhaps the question for 2022 is, which will come first, the signal or the recession itself? The latest quarterly data for GDP shows 1.3 per cent growth in 2021’s fourth quarter and eking out further growth is proving a real struggle even before the first effects of war in Ukraine kick in. The provisional monthly data show that GDP grew 0.8 per cent in January and scraped 0.1 per cent growth in February. So it looks unlikely that UK GDP will shrink in 2022’s first quarter. In which case, the earliest point at which recession could be triggered is with the GDP data for the third quarter. That isn’t due until the end of the year, which seems to give plenty of time for the unemployment figures to make their prediction. Certainly, jobs data for March, out next week, will make interesting reading.
Meanwhile, will there be a bear market in equities by the year end? One difficulty in forecasting bear markets is that, unlike a recession, there is no conventional definition. I could suggest a bear market is where there is much wailing and gnashing of teeth among market makers, but what’s unusual about that? The closest we get to a consensus is that a bear market is entrenched when an index falls 20 per cent from its recent peak. In which case, a level of just below 6,200 for the FTSE 100 index would do it since the three peaks over the past five years have each been around 7,700.
At 7,260 currently, and 5 per cent below 2020’s high, the index still has some way to fall to make bear territory, although it could cross the remaining gap easily enough. It has been way below 6,200 twice since the start of 2020; once when folk in the affluent world realised that SARS-CoV-2 infected them too and once when the second wave of the virus started rolling in.
It is unclear whether present worries are more prosaic or more apocalyptic than fearing a virus. More prosaic if issues about supply chains and a shortage of microchips are the major factors; more apocalyptic if that nice Mr Putin really is as mad as many would have us believe.
Obviously, it does not help that central bankers don’t know what to do. It is easy enough to define their dilemma – should they raise interest rates further and usher in the recession they don’t want, or leave rates where they are and appear to abandon their remit to control inflation? Part of the problem is their own conceit, their belief that monetary policy has a big influence on inflation in the long term. Or perhaps central bankers feel obliged to give the impression that their monetary tool box is clean and sharp and fit for purpose. Behind the front, intuitively they know they are damned if they do, damned if they don’t.
At times like these it can be sensible to reach for the tried and tested means of navigation; in particular to ask, what does mean reversion suggest? If markets move towards their average position, where does that leave investors?
It depends on the reference point. Take the so-called CAPE rating, which uses 10-year average earnings as the denominator in a price-to-earnings ratio, and US equities look expensive on a 32.5 times ratio; not scarily so, but with lots more downside than upside (the average CAPE multiple of the past 50 years is 21 times). Simply look through the lens of current year’s forecast earnings and the S&P 500 index of US shares is a bit below its five-year average rating at 17 times, while the FTSE 100, rated at about 10 times earnings, is well below its 13.6 times average. This may help explain Footsie’s comparative resilience since February. So I might be fussy about committing cash into new equity holdings, but I certainly would not be dumping holdings and running for the cover of cash willy-nilly.
bearbull@ft.com
