Financial Times (Insight Column), 24th July 2007
Investors do not like to be told that markets are expensive, and they will not readily forgive the messenger unless he can time the coming fall.
But shares would not become expensive, as they clearly are today, if their falls could be timed. Market analysis is thus an inherently unpopular business and has poor economic prospects.
Reassurance is preferred. The acceptable messages are those of an 18th-century watchman: either “Midnight and all’s well” or “The fire has started”.
While stock markets are overpriced, they are well short of the excesses reached in March 2000. While nominal prices are similar, seven years of inflation and some ploughing back of retained earnings have substantially improved underlying values.
But profit margins are high and both theory and experience make it extremely likely that they will fall back. To compensate for this, price/earnings multiples need to be well below average rather than, as they are, well above.
Falling profits are among the most obvious threats to the stock market. It is therefore interesting that US national accounts show that profits have fallen in each of the past two quarters. And it is doubly interesting that this information has received so little publicity. It will probably be a surprise to many readers.
Profit margins are most likely to fall when they are wide and have begun to narrow. Both these conditions are currently satisfied. The profit outlook is thus poor, but investors should not be too concerned about this in the short term.
Profits in the national accounts are different from those published by companies, primarily due to differences in the way assets are valued. Changes in both are strongly related, but the relationship is weak in the short term.
This is potentially helpful for investors who are seeking, vainly perhaps, to time the market. It will be interesting to see if profit margins narrow again for a third quarter in a row, when the data are published in early September.
The longer profit margins contract in the national accounts, the more likely it becomes that the profits of listed companies will fall in 2008 or 2009.
Another reason for taking a relaxed attitude is that, even if the profits of listed companies start to fall, it is far from clear that this will drive the stock market down. Markets and profits only move in the same direction about60 per cent of the time. As both are on a long-term rising trend, this is little more than a random relationship.
Even when markets are seriously overvalued, it takes a fall in earnings of 10 per cent or more to have a strongly predictable impact.
There are, nonetheless, some indicators that can be shown to have a statistically significant relationship with short-term market movements. They are not, of course, 100 per cent reliable. Probably the best single indicator is that of the “cash flow dividend”, which defines “dividends” as all cash distributed by companies to shareholders, including buy-backs. The current “cash flow dividend yield” is exceptionally high and genuine research has shown that this is significantly positive for the stock market over the next few months.*
Other helpful signs that lead me to expect the market to remain robust, for the time being, are the low levels of volatility and credit spreads. The former is mildly helpful by itself, as high volatility is associated with negative returns, but also because it is a leading indicator of credit spreads.
Many commentators see credit spreads as crucial to the stock market and I share this view. Companies are almost the only buyers of shares; without them the market is without visible means of support. Easy credit is the essential foundation for this to continue and is vulnerable to further falls in profits.
Andrew Smithers is the chairman of Smithers & Co
* “Dividends, Total Cash Flow to Shareholders, and Predictive Return Regressions” by Donald Robertson and Stephen Wright, Review of Economics and Statistics, February 2006