US equities are expensive, according to one longstanding measure.
This measure is the ratio of share prices to the net worth of companies – the value of their capital stock and other assets, It’s often called 'Tobin’s q', after the Nobel laureate who described it in 1970. Latest figures from the Federal Reserve estimate this ratio to be over 165 per cent for non-financial firms. That’s twice the average since data began in 1952, and the second-highest level ever – just behind the record reached in early 2000.
Now, Tobin intended this measure to be a theory of capital spending. The idea was that if the value of shares was high relative to a firm’s capital stock it was a signal that investors expected that stock to deliver high profits, which should cause the firm to invest more. And if share prices were low it was a sign that the firm had over-expanded and so should cut back.
But the measure has instead often been used as a guide to equity valuations. There’s a good reason for this. It has predicted bull and bear markets. High levels of Tobin’s q in 1969 and 2000 led to the S&P 500 falling in subsequent years and low ratios in the early 80s and in 2009 led to high returns on equities.
A high q can therefore be a sign of irrational exuberance and therefore a predictor of falling prices.
With the ratio now near an all-time high, this seems an obvious sell signal.
Or is it? It poses the question: why is Tobin’s q so high?
One possibility is that share prices are high relative to the capital stock because investors are discounting future profits at an unusually low rate. This rate comprises two things: the risk-free rate and an equity risk premium. For a given risk premium we’d expect a lower risk-free rate to produce a higher Tobin’s q. Which is what has happened: Tobin’s q has trended upward since the mid-90s as real bond yields have trended down.
Herein, however, lies a danger. If bond yields rise then Tobin’s q and share prices will fall unless there’s an offsetting fall in the risk premium. It’s probably no accident that Tesla’s share price, one of the most highly valued companies, has fallen this year as bond yields have risen. That could be a portent of the fate of other high q stocks.
But of course, risk premia change all the time. And it’s possible that insofar as higher bond yields are a sign of a stronger economy then a falling risk premium would offset the impact on Tobin’s q of those rising yields. This, however, is only one possibility. It might also be that the high Tobin’s q is itself a sign that the risk premium has fallen too far.
There is, however, another issue with Tobin’s q. It’s that the Fed’s measure of companies’ net worth leaves a lot out. Once upon a time, it was reasonable to suppose that the value of most companies could be measured by the machinery and land they owned. Today, though, things aren’t so simple. Increasingly, the value of companies lies not in physical objects but in intangible assets. You can interpret the rise in q over time as being due to the increasing importance of these intangibles.
To infer that this means stocks are fairly valued is, however, too hasty.
To see what I mean, take Tesla (US:TSLA). Most of the patents it has won are in the public domain. Studying these will, however, not enable you to make electric cars because you just don’t have the skills to turn these patents into real products. And nor do the people you could employ. What Tesla has is organizational capital – an expertise in making cars that is embodied not in legal documents but in the experience of its teams of workers: I say teams because most individuals, even highly skilled ones, are more or less replaceable. This organizational capital is what impedes rivals to Tesla springing up. It’s one of Warren Buffett’s economic moats. It’s a massive intangible asset.
It does not follow from this, though, that Tesla is fairly valued. Organisational capital is almost impossible to value precisely. What we do know, though, is that assets that are difficult to value are more vulnerable than others to swings in sentiment.
A similar thing is true of other forms of intangible assets. Apple has huge brand power which allows it to charge hundreds of pounds more for its phones than its rivals can. That’s obviously a great intangible asset for Apple. But it’s a darned nuisance for Apple’s rivals who suffer a more limited market as customers stick with Apple’s products. Apple’s brand power might therefore justify its high q, but only by depressing other firm’s qs. This only justifies a high q for the aggregate US market to the extent that Apple’s rivals such as Samsung are listed elsewhere.
This point broadens. Think of retailers suffering from competition with Amazon, or would-be internet advertisers losing to Facebook and Alphabet. In these cases, what we should be seeing is a large dispersion of qs but not necessarily a high aggregate q.
It’s hard, therefore, to fully justify a high q. Merely invoking low discount rates or abundant intangible assets is not sufficient. This doesn’t of course mean the market will fall immediately: the correlation between Tobin’q q and subsequent returns, whilst significant, is not huge and over-valued markets can stay over-valued for some time. But it is a reason for investors to be very wary.
