Markets with a gilt complex

Click to follow
Could the equity market suffer another crash if yields on gilts, the British government's debt, continue to rise? Or, to put it less starkly, could gilts outperform equities?

An increasing number of investors and analysts are beginning to question whether the equity market has moved fundamentally out of line with gilts.

Their concern is that the stock market can be sustained at current levels only if gilt yields fall.

The traditional mechanism for analysing the value of equities is the dividend discount model - DDM. It is based on the principle that rational investment decisions should be made by discounting expected future cash returns back to the present and comparing the sum of the resulting net present values with the cost of the investment.

The DDM links current share price valuations with interest rates and expectations regarding future dividend payments, providing a simple framework to assess the relative value of bonds and equities. In its simplest form, it can be written: Risk-free rate + risk premium = dividend yield + dividend growth rate.

Adjusting for inflation, the expression becomes: Real risk-free rate + risk premium = dividend yield + real dividend growth rate.

The main shorthand methods for valuing bonds and equities - such as the yield ratio, yield gap and reverse yield gap - are simple rearrangements of these expressions.

A number of problems need to be solved before the DDM can be applied. For example, dividend growth at this stage of the business can be expected to be above its long-run trend.

Secondly, the risk premium - the return that the average investor requires to compensate for the additional risk associated with equities - cannot be observed directly and must be deduced. It is normally calculated by examining historical returns on gilts and equities and assuming that the difference between the two measures the risk premium.

However, it is difficult to believe that the very low real returns on gilts in the post-war years are an accurate measure of the return anticipated by investors in advance of events. This is because the post-war period contains two inflationary periods - the Korean war in the 1950s and the mid-1970s oil price shock - which were almost certainly unanticipated.

It is likely therefore that the risk premium is rather below the 7 per cent inflation-adjusted figure that emerges from comparing returns on equities and gilts over the past 30 years (see table 1). A more plausible range is 2.5-4.5 per cent.

Comparing equities with index- linked gilts, which provide a fixed real return, simplifies things. The yields on equities and index-linked are currently similar, which means that the real risk-free rate equals the dividend yield. The DDM then implies that equities can sustain their current valuation only if the real growth rate of dividends equals the required risk premium.

There are three ways in which the corporate sector could boost dividend payments to satisfy this requirement. First, it could pay out a higher proportion of its profits as dividends, increasing the so-called payout ratio. Secondly, they could boost the share of profits in national output.

Finally, they could use the proceeds from real economic growth.

The payout ratio, the proportion of earnings paid out as dividends, has been the subject of considerable debate recently. The substantial rise in recent years has taken it to levels well above those of the 1970s and 1980s. It is in line with typical payout ratios in the 1960s, when inflation was much lower. A continued rise in the payout ratio is unlikely.

The profit share has also risen in the past 10 years, notwithstanding the normal cyclical slump during the recession. It can be expected to recover further as long as the economy continues to grow above trend, but only to a limited degree. Given our expectation that economic growth will slow next year while margins remain under pressure, particularly in the retail sector, the scope for a further rise in the profit share beyond the end of this year is limited.

Dividends are likely to grow above trend this year and next, but beyond that it is unlikely that they will exceed the trend growth in real GDP, which is 2-2.5 per cent. Given that this is below the lower end of the range of plausible risk premia, there must be real concern about the sustainability of the equity market at current levels.

The dilemma for investors is that moving out of equities into gilts could reduce their average returns. With increasing pressure on performance measurement for fund managers, the temptation to stick with equities is intense.

Indeed, as Table 2 reveals, British pension funds have reduced the share of bonds in their portfolios from roughly 20 per cent in the 1980s to only 11 per cent at the end of last year. UK bonds occupy just one-25th of pension fund portfolios.

Pension funds might have to reduce the share of equities at some stage over the next five to 10 years as more and more of their members reach retirement and they come under more pressure from their actuaries following the Goode report.

The risk is that as more and more pension fund managers move away from equities in favour of bonds of one form or another, the combination of peer pressure and benchmark performance measurement will prove self-feeding.

The question for fund managers is, should they make this switch now when bonds look cheap relative to equities? The clue lies in the crucial role of the dividend yield in the DDM. Table 1 reports the average returns on equities, cash (treasury bills) and gilts over the past 30 years. It displays the familiar result that equities substantially outperform the other investments.

However, nearly all of this occurred in years when the dividend yield was above average in the previous year. In other words, if a fund manager had simply held an excessive weight of equities only when the dividend yield in the previous year was high he would have captured nearly all the outperformance of gilts versus equities.

Suppose a nave fund manager avoids equities only when the dividend yield is 1 per cent or more below average. In this case he would have held equities for all but five of the past 30 years. He would have avoided an average return on equities of minus 5 per cent per annum on those years.

More remarkably, had he switched his portfolio into gilts for those five years he would have outperformed his 'equity-only' competitors by 2 per cent a year in real terms. The old adage that gilts only outperform equities in a bear market may be true, but the evidence here suggests that these markets can be anticipated simply by looking at the dividend yield.

The dividend yield is currently more than 1 per cent below its historical average, suggesting that equities are likely to underperform in 1995. It was also more than 1 per cent below average in 1993, suggesting that equities will perform badly in 1994. We shall soon see.

Steven Bell is the chief economist of Morgan Grenfell.

(Table and graph omitted)