The writer is chairman of Fulcrum Asset Management
The Covid-19 economy has had many unprecedented events packed into a few months, but none more extraordinary than the switchback in global equities.
The advanced economies experienced a savage but brief bear market up to 23 March, followed by a spectacular recovery that eliminated all the year-to-date losses within 17 more weeks. Any investor who succeeded in navigating both legs of this reversal was either very skilled, or very lucky.
Economists are already investigating the episode — what have they learnt?
The first question is whether the market behaved “rationally”? in the sense that the entire decline and rebound in equity prices can be explained by alterations in growth expectations, driven by the economic shutdowns to control the virus in March, followed by the start of unlocking in April and May. That does not appear to be the case.
To understand why, start with the idea that the stock market’s total value equals the present value of the benefits equity holders expect in the future. This changes for two reasons. First, expectations about the size of the benefits — corporate earnings and dividends — rise and fall. Second, the present value of these future benefits is affected by what is known as the equity discount rate. This is equal to the return on safe assets like Treasuries plus what investors expect to be paid for the extra risks — losses, bankruptcy — of owning shares.
Augustin Landier and David Thesmar have presented evidence that suggests only a small part of the US equity price changes were explained by corporate earnings expectations. In the downward leg, a survey of investment analysts’ earnings forecasts for 2020-23 shows that the present value of these earnings fell only a few per cent from the start of the year, assuming the discount rate remained the same as that pre-Covid-19. During the equities recovery, earnings expectations continued to decline, thus explaining none of the bounceback.
Calculations based on dividend expectations paint a similar picture. Niels Gormsen and Ralph Koijen of Chicago Booth School of Business show that dividend futures, which measure expected dividend growth, fell much less in the crisis than stock prices. Fulcrum economists show that while earnings expectations fell in March, those for dividends changed little.
So amid such volatility, “rational” changes in growth expectations do not seem to explain the behaviour of stock prices. Many economists and equity analysts apparently viewed the Covid-19 shock as a fairly short-term event that would last just 12-24 months, thus making little difference to the present value of future company earnings.
What, then, accounted for the dramatic events? If the path for earnings or dividends was not enough, the discount rate applied to that path must have risen to explain the bear market. Professors Landier and Thesmar indicate that the discount rate had grown 3 percentage points by 23 March.
Given that the risk-free rate on US treasury bills dropped by around 1.5 percentage points over that time, the equity risk premium must have jumped by nearly 4.5 points in the market collapse. Furthermore, almost all the recovery in the market was driven by the ERP returning to normal.
That is not particularly unusual. Nobel Prize winner Robert Shiller showed in the 1980s that large equity price swings are usually driven by the discount rate, not by changes in dividend expectations. More recently, he has suggested that changes in the risk premium were exacerbated by “vivid stories” about the spread of the virus, and the dramatic easing of US fiscal and monetary policy within a day of the market trough.
These narratives certainly played a part, but Prof Shiller may be underestimating another factor. The structure of the financial system, in particular new rules that cut the risk capital available to bond market makers, led to disruptive behaviour there, undermining confidence in the financial system. Goldman Sachs’ financial conditions indicator tightened by more than 3 percentage points during this period, and equities were severely damaged.
The Federal Reserve’s March 23 emergency easing may have been the key factor in addressing these problems, allowing risk appetite to be restored in equities, credit and long-duration government bonds.
Now the Fed has demonstrated its willingness to take unlimited action to prevent further illiquidity and disruption, the equity risk premium should be much less volatile, even in the event of a further major wave of economic damage from the virus.