Business | Investment
To buy or not to buy is the question
Are stocks cheap? It is an endless, complicated and unanswerable debate, so let me start by short-circuiting some of the first steps.
- By John Authers, Financial Times
- Published: 00:32 February 16, 2008
Are stocks cheap? It is an endless, complicated and unanswerable debate, so let me start by short-circuiting some of the first steps.
It is traditional to value stocks by their price/earnings ratios (the ratio of their current price to the latest year's earnings), and by comparing earnings yields (the proportion of their earnings to their total market value) to bond yields. On either measure, stocks in the developed world look cheap. But there is growing acceptance that looking at only one year's worth of earnings can be misleading.
Earnings are cyclical, and so multiples should take account of this. When we look at cyclically-adjusted or "trend" earnings (calculated by taking the average of the past 10 years' earnings), stocks no longer look so cheap. Earnings, and profit margins, have been very high, and seem overdue for a correction. Some multiple of "trend earnings" is superior when trying to gauge the long-term value in the market.
So far, this is uncontroversial. The concept of trend earnings has now gained widespread acceptance. And in spite of the bear market that followed the insane peaks of 2000, stocks are still more expensive than their long-term norm.
The spread of this thinking helped the market stand back from some of the high valuations that were seen last year. But now, having accepted all of this, the bulls have a response.
First, and most reasonably, the financial sector is currently in a crisis that has not (yet) affected other sectors. Its fate is so divergent from other business sectors that measures of the market as a whole may not be meaningful. Experience with earnings makes this case. Entering this week, in the US, financials were expected to show severe losses for the fourth quarter of last year, according to Thomson Financial. This fed into an expected decline in S&P 500 earnings of 20.7 per cent.
But non-financials were on course for an impressive growth rate of 11 per cent. Only the consumer discretionary sector had suffered significant disappointments compared with expectations. Narrow the field to non-financial sectors, many of which have suffered double-digit falls in share prices so far this year, and valuations look more appealing.
Rule of 20
Next, multiples can be higher when inflation is low. The "rule of 20" - that the p/e should be obtained by subtracting the inflation rate from 20 - is not a bad rule of thumb. Lower interest rates, a fact in the US and widely expected in Europe, might also justify a higher valuation than usual.
Further, not all regions are uniformly expensive. Ian Scott of Lehman Brothers says the multiple of European stocks to their trend-adjusted earnings is 13.9 - actually below the long-term average of 16.8. Another bullish argument is that earnings fluctuations are not solely cyclical. They may also, in Europe, be affected by a one-off secular shift.
Then we can dust off a deep measure of value - dividend yield. It almost never makes sense for the yield a stock pays out in dividends to be more than the yield on a bond, because the coupon on a bond does not have the potential to rise over time. The only time when the dividend yield on the S&P 500 moved above Treasury bond yields in the last two decades was in early 2003 - a good signal of the multi-year rally in share prices about to commence.
According to Lehmans, the pan-European non-financial dividend yield is now 2.9 per cent. Those dividends are amply covered by cash on corporate balance sheets. In the US, the S&P's dividend yield is almost equal to the yield on two-year government bonds - and the dividend yield on financials, is actually higher than banks' overnight rates.
The bears have answers too. James Montier, equity strategist at Societe Generale in London, says that the average trend p/e, when the tech bubble is excluded, is about 14. It currently stands at 21. Stocks historically overshoot in both directions, and bear markets do not end until the trend p/e hits 10.
Earnings forecasts
Also, while the rest of the corporate sector may not suffer as much as financials, current earnings forecasts suggest that it will not suffer at all. For 2008, the market expects S&P 500 earnings to grow 15.1 per cent compared with 2007. This is partly thanks to an expected 26.6 per cent increase for the financials (aided by a low base of comparison). But all sectors are forecast to grow at least 7.2 per cent.
Analysts, it appears, do not expect even a cyclical slowdown in profits, let alone any negative impact from the financial crisis. But with surveys from the Federal Reserve and the European Central Bank showing that banks are tightening lending standards, it is impossible to take this point of view seriously.
As for dividend yield, it is flawed. Some large companies do not pay any dividends. And arguments about secular improvements in profit margins sound like the discredited "new paradigm" of the 1990s.
There are points of agreement. Stocks are not as overvalued as they were in 2000. If all the bad news about the credit crisis and its second-order effects is already out there, stocks are not expensive, and financials are a screaming "buy".
The problem, of course, is that that is a very big "if".
Share this article
More from Investment
More from Business
Popular in Business
-
XPRESS
Way to go this DSF
A fun-filled route to guide you to all the happening dos in town
Business Editor's choice
-
The brains behind campaigns
Chief executives likely to be guided more by bankers and consultants
-
Leaders blast Britain's green departure tax
UK government pulls out of UN body ahead of Copenhagen summit
-
Emirates could buy more aircraft
Airline plans new routes, expects demand to increase
