Measuring the gauges of US stock valuations

September 1, 2013 3:00 pm

Measuring the gauges of US stock valuations

By John Authers

The key metrics are coming in for re-examination

Cue q. For weeks, an academic debate has spilled into dealing rooms over the merits of cyclically adjusted price/earnings ratios, known as Capes, one of academics’ favoured metrics for gauging whether stock markets are cheap or expensive. But another key gauge is also now coming in for re-examination: Tobin’s q. Before anyone stops reading, this matters, a lot, to those hoping to make money in stock markets. Cape and q both tell the same story, that the US stock market is significantly overvalued and this contradicts much hopeful analysis from Wall Street and the City. As so many people have an interest in selling stocks, it is inevitable that these metrics will be re-examined; and it is vital to understand why academics find them so useful.Cape and q attempt to value a stock in completely different ways. Cape uses the financial approach; when you buy a stock, you are buying a share of the earnings it produces. How much are you paying for them? Comparing the price to the average earnings per share generated over the previous decade is a good way to do this.

But Tobin’s q is an economic concept, devised by the Nobel laureate economist James Tobin to explain trends in investment by companies. Tobin’s definition of the q ratio was: “The numerator is the market valuation: the going price in the market for exchanging existing assets. The denominator is the replacement or reproduction cost: the price in the market for the newly produced commodities.” He called q “the nexus between financial markets and markets for goods and services”.

Q has nothing to do with a company’s reported profits and there is no reason to expect it to give a similar result to Cape. Instead, if a company has a q above 1, then markets think that it is worth more than the sum of its parts – and will finance it to make more investments. But most of the time, q is below 1 and tends to revert to a mean of about 0.7.

Cape and q do, however, share two things in common. First, they are logical ways to tell if stocks are cheap or dear. Second, to a quite stunning extent, they always tell the same story. Graphs for the Cape and q of the US stock market, as measured by the S&P 500, over the past century are almost identical. According to Andrew Smithers of the London consultancy Smithers & Co, who has made several academic contributions to the debate, the latest q figure, of 1.01 as of March 31 this year, suggests that US non-financials are overvalued by 58 per cent – while Cape suggests an overvaluation for the market as a whole of 61 per cent.

But can we be sure that the q numbers, published as part of the flow of funds data by the Federal Reserve, are accurate? Just as academics raise concerns that changing accounting conventions affect the profits used to calculate Cape, so there is also a re-examination of the replacement value of assets used to calculate q.

In particular, how to deal with intangibles? JPMorgan’s Michael Feroli strolled into the academic debate by publishing a paper recalculating q after the Fed data for the first time included spending on research and development and on artistic commercial products as investment. These sums came to $1.4tn and $435bn respectively. Adding these to the official estimate of non-financial corporate assets has the effect of bringing q down from 1.01 to 0.91.

Mr Smithers, when prompted to respond to this by an academic, pointed out that the Federal Reserve includes some $6.8tn of unidentified “miscellaneous assets” in its calculations. These include such items as goodwill, copyrights, patents, franchise fees, trademarks and client lists – all intangible. So the newly identified R&D spending may already have been included.

Mr Smithers adds that the Fed does this to try to bring the value of identifiable assets into line with the values implied by the retained profits declared by companies. “We know that companies overstate their profits because the average earnings yield on US stocks is higher than the real return to investors and real dividends per share have grown too slowly.”

This overstatement affects both q and Cape, and makes it necessary to look at both with respect to their own historic average, not in absolute terms. It also implies that valuing the stock market is even harder than it at first appears, and that standard metrics if anything show stocks to be cheaper than they would be if profits were reported accurately.

Mr Smithers dismisses the attempt to revalue q as “nonsense – very sensible nonsense if you are in pursuit of business and therefore wish to assert that shares are always cheap – but not sensible if you are in pursuit of the truth”.

Again, this matters. Academics drove most of the financial innovations of the last generation. And the conventional academic wisdom that US stocks are overvalued still looks convincing.