Company valuations are highly sensitive to perceived risk

Citation

Robinson, B. (2003), "Company valuations are highly sensitive to perceived risk", Balance Sheet, Vol. 11 No. 2. https://doi.org/10.1108/bs.2003.26511bab.002

Company valuations are highly sensitive to perceived risk

Bill RobinsonBill Robinson is Head UK Business Economist in Financial Advisory Services at PricewaterhouseCoopers. He is a former Special Adviser to the Chancellor of the Exchequer.

At the start of 1995 the FTSE 100 index was hovering just above 3000. By December 1999 it was approaching 7000. Since then it has halved in value. These gyrations of the UK stock market stand in stark contrast to the behavior of the underlying UK economy, which embarked in the 1990s on its longest period of steady growth since the Second World War, followed in 2001 by a deceleration that has so far proved remarkably mild. In consequence the profits and dividends which underpin share prices have suffered no more than a slight slowdown. The variations in share prices are thus almost entirely due to variations in valuation ratios or PE multiples, which have gone from 15.6 in December 1995 up to 30.5 at the end of 1999 and back down again to 17.7 in March 2003.

The strain on valuations

Do these events prove that markets are inherently irrational? They certainly make life difficult for those who make a living valuing companies, since the owners become understandably upset if they are told one moment that their company is worth 30 times annual earnings, only to learn a year or two later that it is actually valued at only 15 times. However, the main lesson of the boom and bust is not that the entire financial community (world wide) went collectively mad. It is that the price placed by markets on a stream of (future) income can vary widely depending on economic conditions and expectations, for good reasons explained below.

The value of a stock is simply the value of the future stream of dividends (broadly defined to include all cash flows to shareholders), and the first uncertainty is how fast those dividends will grow. A sensible assumption for the UK market as a whole is that they will grow at 2.5 percent in real terms, the same as the underlying growth of the economy. But it would not be ridiculously optimistic to assume 3 percent nor unduly pessimistic to assume 2 percent. So there is uncertainty about the future rate of growth.

Next, in order to calculate a net present value of this dividend stream, we have to apply a discount rate which reflects the risky nature of those dividends – the so-called cost of equity. The cost of equity is generally calculated by adding an equity risk premium to a risk-free rate of interest (e.g. the return on ten year index linked government bonds held to maturity). The real rate of interest is known with certainty – it fell from around 3.5 percent in 1995 to 2 percent in 1999 and has since hovered somewhere above 2 percent. The equity risk premium by contrast cannot be observed directly. It is a subjective measure of shareholders' aversion to risk and current estimates of the equity market risk premium range from 2.5 to 5 percent or higher.

Going by the textbook

There is a text-book formula[1] that enables us to calculate the PE ratio from the cost of equity and the expected growth rate. Table I shows different values of the ratio associated with different values of the risk free rate, the equity risk premium and expected growth (assuming a constant dividend payout ratio). In December 1995, when the risk free rate was close to 3.5 percent, then if the expected growth of dividends was 3 percent (because the economy was enjoying a vigorous recovery), the valuation multiple of 15 was consistent with an equity risk premium of around 3. At the end December 1999 peak, when the risk free rate had fallen to 2 percent, the valuation multiple of 30 can be explained by dividend growth of 2.85 percent, and an equity risk premium of 2.5 percent. By end December 2002, with the risk free rate still at 2 percent, the multiple of 17.8 is consistent with expected growth of 2 percent and an equity market risk premium of 3.5 percent.

Table I

A closer examination of Table I reveals that changes in the risk free rate, and in expected growth, have a bigger effect on PE ratios the lower is the equity risk premium. This suggests that the key factor in the boom could have been the fall in the risk premium, which magnified the effect of the falling risk free rate. Unrealistic growth expectations, on which the bubble is often blamed, may on this view account for a relatively small part of the rise in PE multiples. Investors did undoubtedly become over-bullish about future growth, especially in the growth sectors (telecoms, media, and technology). But as Table I illustrates, the doubling of the PE ratio between 1995 and 1999 requires no increase in expected growth, if we assume that the equity market risk premium fell by 0.5 percent.

The collapse of the market and of PE ratios since early 2000 is a somewhat different story, because there has been no significant change in the risk free rate over this period. But we can still account for the dramatic fall in PE ratios by, for example, a 1 percent rise in the equity risk premium and a downward revision of growth expectations of less than 0.5 percent.

Table I and the valuation formula which underpins it, is not just a piece of textbook theory. It is a description of the way real-world analysts value shares. What it suggests is that:

• •

  the main driver of the boom was the fall in the risk free rate of interest combined with a lower perceived equity risk premium;

• •

  the main driver of the slump has been the rise in the perceived equity risk premium; and

• •

  none of the implicit changes in growth expectations or discount rates has been particularly large or (except with the considerable benefit of hindsight) irrational.

Be gloomy about risk premiums

Does this framework for analysis have anything to tell us about the future? It does, and the message is that a great deal hangs on the equity risk premium, estimates of which vary widely. Historically UK equities have yielded 6 percent per annum more than gilts on a total return basis, and one school of thought holds that the premium actually paid is the best possible estimate of the premium required. On balance however, estimates of the equity risk premium have been falling. The most authoritative recent study, by London Business School professors Dimson, Marsh and Staunton, puts the UK equity risk premium as low as 2.5 percent, though the same study estimates the global risk premium at 4 percent.

These higher risk premium estimates have quite gloomy implications. If the market were to discount future cash flows using a 5 percent equity risk premium, the PE ratio would fall from 16 to 11. Moreover it would only take a 1 percent increase in the real rate of interest (returning it to more normal levels on the basis of the experience of the past 20 years), and a downward revision of growth expectations to 2 percent to bring the PE ratio down to around half its present level.

By contrast, if the LBS view of the UK market is correct and UK investors eventually revert to a 2.5 percent equity risk premium, then PE ratios should revert to 24, implying a rise of 50 percent in the market. That would be another triumph for the optimists.

Note

1. 1.

   P=d/(rf+e–g) from which it follows that P/E=(d/E)/(rf+e–g) where P=share price, E=earnings, d=dividend, rf=risk free rate of interest, e=equity risk premium, and g=expected growth.