share prices having fallen so far, the s&p500 continues to boast a p/e of almost 50 on the back of proportionally low earnings. nick dewhirst asks which statistics to believe
After the market crash, are shares now cheap enough? The trite answer is that this depends on what one means by cheap. The sensible answer is to decide what measure of value is most meaningful.
I base my answer on Wall Street, because this is where most investors look for leadership and because that is where the data is most controversial, but not because the conclusions are necessarily true for the rest of the world.
There is a wide choice of valuation criteria, some of which are based on balance sheets, while other are based on income statements.
While widely quoted by others, Price to Book Value has little appeal to me, apart from investment companies whose business is investment in liquid assets like securities or real estate.
The intellectual difficulty in using book value as a meaningful measure of value is that it is a poor proxy for market value. Some tangible assets may have appreciated massively as a result of inflation, such as real estate, while other intangibles may have depreciated massively due to obsolescence, such as steel mills.
The problem also affects intangible assets. On the one hand, consumer companies generally have understated assets because investment in building brand names is written off against income. On the other hand, technology companies may record high priced acquisitions that are now worthless.
Measures based on the income statement have greater appeal, because an asset that cannot easily be sold has limited realisable value, except as a capitalised income stream, which brings the discussion back to the income statement.
I was always sceptical of the new era valuation measures dreamt up by analysts during the bull market.
• Enterprise value (EV), included debt as well as equity, but what use is that to an equity investor, unless they are bidding for the whole company?
• Enterprise value added, (EVA), might be a fine measure of corporate efficiency, but how does that enable one to value shares?
• Cash flow might be a good way to compare companies within the same industry, but how does that help one to compare one industry that needs massive investment to stay in business, with another industry stock, that can generate high profitability without much capital spending?
• Earnings before interest, tax, depreciation and amortisation (EBITDA), may similarly be a good way to compare companies in some industries, but isn't that twice as complicated as Cash Flow and twice as bad?
Of the traditional valuation measures, P/E ratios seem better than dividend yield, because the former includes all distributable profits, while the latter merely represents the director's preference for returning funds to shareholders on a regular basis.
Clearly earnings can be manipulated and the Enron scandal shows that this has been all too easy in recent years.
If the P/E ratio is intellectually superior, but faces practical difficulties, the answer is to remedy those practical difficulties by tightening the rules that define profits. The accounting authorities have been working intensively on the specific issues, and the two global players, US-led GAAP and UK-led IAS are moving towards a single global standard.
The other invaluable strength of this traditional measure is the long history of back-data available for analysis. While the precise definition of profits may have changed over time, what has not changed is that it represented what investors believed they were buying at the time.
In crude terms, the market's P/E ratio is not cheap based on past history. The ratio for the broadly based S&P 500 index is 43.9. That compares with a peak of 62.9 in 2002 after the millennium bubble and a low of 6.8 in 1980, at the height of the commodities boom. The present valuation compares poorly with historical average, and suggests the market could still halve from these depressed levels unless corporate profits double.
The P/E ratio is shown by the thick line on the chart above and displayed on the right hand vertical axis. For comparative purposes with other assets, its reciprocal, the Earnings Yield, is shown on the same basis as other assets on the left-hand axis.
However, this crude historical comparison takes no account of inflation. The current return on all assets should fluctuate with inflation, in order to provide the same real return.
Over the broad sweep of history that is indeed what this chart shows. Inflation peaked in 1975 and again in 1980. The first upsurge caught markets unaware, but the second one was fully discounted. Since then inflation has steadily declined, with cyclical interruptions, and so too have current returns on most asset classes.
Unfortunately, even after inflation adjustment, shares are not cheap compared to past experience over three decades. In terms of earnings yield, the current return is -0.7%, which compares with a peak of 7.4%, a low of -4.1% and an average of 2.5%.
This may disappoint today's investors, but ultimately that is irrelevant, for they must make their decisions in the light of alternatives available today.
Therefore it is preferable to consider valuation against other assets. For the dominant category of institutional investors these alternatives are long-term bonds, cash and commercial real estate, and they are shown as the other series in the chart.
Sadly, US equities do not seem outstandingly cheap on this basis. At a P/E ratio of 43.8, the earnings yield is 2.3% before inflation. This beats Cash at 1.2% in the form of T-Bills, but it does not beat long-term bonds at 3.8% in the form of 10-Year Treasuries. Real estate is today's bargain, at yields averaging 7.2% on large commercial properties.
This is most disappointing to investors who have seen blue chips portfolios halve and technology hopefuls become almost worthless, so how is it possible that shares are still not cheap after the worst market setback in quarter of a century?
The explanation lies in the performance of corporate profits, which have also collapsed, as shown in the second chart.
According to Standard & Poor's (S&P), reported profits peaked at $56 per share in 2000, but collapsed to only $38 in 2001 and recovered only marginally to $46 last year.
Economic slowdown is only part of the reason for this, as the setback was not particularly sharp, amounting to a reduction in growth from 3.8% to 0.3% ' not even a recession.
The other reasons relate to the pricking of the bubble. Write-downs of over-priced acquisitions are a major cause, but more conservative accounting practices will also be playing a role since the Enron and WorldCom scandals.
In the past there was little difference between reported and operating earnings, but as the chart shows, since the millennium, there has been a much greater setback to reported profits. S&P has also recently introduced an even more conservative definition for Core Profits, after adjustment for other hidden costs such as pension fund profits and employee stock options. On this basis the setback is even greater.
Investors must base their optimism on hopes that the setback in profits will prove temporary, and there are good grounds for doing so, not least economic activity.
Growth recovered to 2.4% last year, with 2.3% more expected this year and acceleration to 3.6% next year, according to Consensus Forecasts. Presumably conservative accounting practices are here to stay, so that will not help, but the flood of write-downs should come to an end.
S&P estimates the net effect will be to boost reported profits by 38% this year and 9% next year. On an ongoing operating basis it expected 17% increase this year and 9% next year.
By way of comparison my own top-down model predicts a rise of 9% this year, accelerating to 15% next year.
It seems entirely reasonable to base valuation on forecast earnings for 2004, as the market nowadays discounts profits two years into the future, and to do so on the basis of operating earnings, assuming special factors are neutral.
Based on prospective earnings, shares look distinctly more attractive and rival real estate. Using S&P operating estimates for 2004, the P/E falls to 15 and the earnings yield rises to 6.7%. Bearing in mind that returns on bonds and cash are fixed, not only do shares and real estate provide the best current return, but they are also the only assets to offer the prospect of rising returns in future.
Before getting carried away with enthusiasm, it is important to note one snag. Whereas S&P estimates reported earnings on the basis of top-down forecasts, as I do, its more optimistic estimates for operating earnings are based on aggregating bottom-up company forecasts, as the brokerage houses do.
Even without any pressure from investment banking towards optimism, bottom-up estimates remain as fallible as the corporate spokesmen on whose guidance they are based.
Using S&P reported profits, the prospective earnings yield is 4.5%, equivalent to a P/E ratio of 22.3. This is okay, but not fantastic.
This inconclusive analysis may well leave investors confused at a higher level of sophistication, so gives little cause to expect another big bull market in Wall Street, whatever lesser rally may be overdue.
However, my research indicates that the same analysis elsewhere reaches different conclusions, but that is another subject.
Nick Dewhirst is chief executive of www.investorsroutemap.co.uk
DB and a lack of alignment
Encouraging better use of tech
Win one of three £20 Amazon vouchers
Vanguard's multi-asset range