Brace For A Stock Market Accident
Profits and leverage are locked in a deadly embrace
There is a time-honoured tradition in statistics: whipping the data until
they confess. Bullish and bearish equity analysts are equally guilty of this
It would seem that statistical conclusions are merely an ex-post
justification of a long-held prior belief about equity markets being cheap or
overpriced. Clearly, consensus, notably among sellside analysts, is bullish. I
present the bullish view before discussing a bearish counterpoint.
Who can blame the equity bullish consensus? Earnings yields – a proxy for
real equity yields – stand at comfortably high levels. For example, the forward
earnings yield on the S&P 500 is 8.3 per cent.
Contrast real equity yields with real bond yields: with the US Consumer Price
Index at 1.7 per cent and the nominal Federal Reserve funds rate at 15 basis
points, real bond yields are at -1.55 per cent.
The difference between equity and bond yields – also known as the equity risk
premium – is therefore close to 10 per cent. This is way above the 4-5 per cent
premium required by investors to own equity, and therefore indicative of an
ultra-cheap equity market.
There are two reasons why this consensus is misguided. First, because it uses
dubious metrics. It is wiser to use a long-dated real bond yield because equity
is a long-dated asset.
And forward earnings yields are misleading for well-documented reasons:
analysts’ earnings consensus forecasts are known to be wildly optimistic; in a
bid for juicier equity and call option compensations, managers encourage their
accountants to inflate earnings numbers; and earnings are partially squandered
by managements as they seek to prioritise growth over profitability.
So it is probably a good idea to use dividend-based – as opposed to
earnings-based – equity valuation models. Unlike earnings, dividends do not
Second, because consensus disregards leverage. Profits and leverage are
linked (in a deadly embrace, it turns out). If deleveraging is yet to happen,
then earnings growth can only be headed south.
So what if you trust dividends more than forward earnings? In a simple
dividend discount model, the real equity yield is the sum of dividend yield and
real dividend growth. The S&P dividend yield is 2.15 per cent. The real
dividend growth has been historically 1.25 per cent.
The real 30-year yield is 0.4 per cent. Using these numbers, the equity risk
premium is now 3 per cent, less than the premium level deemed acceptable. But we
are not done yet, as we have not factored leverage into our equation.
Enter Michal Kalecki, a neo-Marxist economist who specialised in the study of
business cycles and effective demand. Mr Kalecki showed that profits were the
sum of investments and the change in leverage.
In the current environment, the implications of this equation are clear: in
G7 economies, total debt is at a record 410 per cent of GDP. And this is
excluding the net present value of social entitlements and healthcare
expenditures, which is larger than the total debt.
Because leverage stands at unsustainably high levels in advanced economies,
it should fall substantially over the long term, affecting profits
It can be assumed conservatively that the total-debt-to-GDP ratio needs to
fall by 100 per cent before the debt position becomes sustainable in advanced
economies. This would bring the US back to 1995, when the profit-to-GDP ratio
was 45 per cent lower.
We can value the S&P under the following scenario: dividends fall by 45
per cent over a zero-growth period of 10 years. Then they resume their real
growth of 1.25 per cent per year. Again, assuming a real yield of 0.4 per cent
and a required risk premium of 4.5 per cent, fair market value is only one-third
of current market levels.
Leverage is hence the fly in the ointment, begging the obvious question: when
does the deleveraging take place? Answering this question is tantamount to
timing the next major bear market. It is, of course, futile to predict a date,
but as economist Herbert Stein used to say, if something cannot go on for ever,
it will stop.
It is increasingly obvious that governments will take no active step towards
deleveraging unless they are under the gun. But there are institutions and
mechanisms that will trigger deleveraging, namely: Basel III, the bond market,
default and, rarely, courageous politicians.
Inflation can also help delever, except in economies where social
entitlements are inflation-indexed.
In the short term, it is clear that central banks need to entertain the
illusion of viable stock market valuations by pulling rabbits from a hat. But as
high-powered money reaches ever higher levels, the probability of accidents