COMMENTARY.

Pain, Nigel ; Weale, Martin

Introduction

Since last summer the British economy has suffered from three adverse shocks that have acted to slow economic growth. First of all, in September the fuel crisis depressed output sharply. Secondly, in the Autumn and Winter the chaos on the railways disrupted normal patterns of business. Thirdly, the outbreak of foot and mouth disease has proved extraordinarily expensive to bring under control. Estimates of the cost to the economy vary but it is possible that the damage done to the service sector by discouraging people from visiting the country from abroad will outweigh the permanent cost to agriculture had the disease simply become endemic. It may well turn out that the response to the disease has been designed to protect farmers' interests rather than the national interest. A reasonable estimate would be that the outbreak of foot and mouth disease will depress output by just over 1/4 per cent this year.

Adjustments to financial markets

These shocks are events whose effects should pass fairly quickly, although the losses to the tourism industry may last for much of the year if the crisis has had the effect of discouraging holidays in the UK. However, superimposed on these temporary shocks, and on the damping effect of monetary tightening last year, have come falls in the major stock markets. In March the Financial Times All Share Index was 10 per cent below its average for 2000 and despite some recovery it remains markedly below its trading range of the previous 18 months.

While the National Institute has been among the more optimistic forecasters in recent quarters, our forecasts have also indicated the consequences of falls to share prices. Our analysis in the January Review (National Institute Economic Review, 175, pp. 36-8) of the effect of stock market weakness suggested that a fall of 20 per cent in the United States combined with a 10 per cent fall in Europe would be likely to lead to a fall in the UK growth rate of 0.8 per cent in the first year. This assumed that interest rates were reduced world-wide in response to the fall; the appropriate reduction in the UK was 1.4 percentage points. In our model simulation, however, the shock was associated with a weakening of the US$ against the European currencies. This mitigated the effects in the United States but augmented the slowdown in Europe. While the dollar remains strong, the appropriate European interest rate reductions are lower than this simulation suggests. On the other hand the dollar strength has meant that US in terest cuts much larger than our simulation identified are needed.

This stock market weakness comes after a period in which the stock market behaved rather strangely. Rising share prices over a five-year period had generated high historic returns on equities. Quite possibly many investors were tempted into the market in the belief that these high returns were normal. On the other hand, high long-run returns could be maintained only with high dividend yields or rapidly growing dividends. High dividend yields can be delivered only in economies in which capital is unusually scarce or in which companies are extremely highly geared. Rapidly growing dividends require, in the end, rapidly growing national income. These seem unlikely explanations of the stock market boom. Some people argued that that labour productivity growth had accelerated. But unless capital--output ratios start to fall, an increase in labour productivity will not increase the growth rate of earnings per share.

Thus an alternative explanation of the rise in share prices in the 1990s was that real rates of return had fallen. There is some justification for this view. In the 1980s government debts were in many cases large and increasing. Since the mid-1990s debt in most countries apart from Japan has been falling as a share of income. For any given amount of private savings, lower government debt means that relatively more can be devoted to productive investment leading to a higher capital-output ratio and a lower marginal return on capital. In the longer term savings rates might be expected to fall, but in the interim real interest rates would be lower. This process may have been helped by reduced corporate taxation in the major economies.

Superimposed on these there is an argument that the risk premium associated with equity investment may have declined. A famous study by Mehra and Prescott (1985) showed that it was difficult to provide a coherent explanation of why the return on equity had historically been so much higher than the return on assets such as government debt. It is perfectly true that equities are much more risky, but they showed that the degree of risk aversion need to explain the yield differential was far higher than economists believe to be plausible. Since then there have been many explanations of why the equity risk premium should be so large. Wadhwani (1999) provides a summary of the various arguments. Other authors have suggested that observation of the risk premium anomaly has eliminated it, so that returns on equity have fallen permanently.

Nevertheless Wadhwani (1999) argued that, unless this was the case, equity markets were overvalued; Bond and Cummins (2000) came to a similar conclusion. Despite the falls of recent weeks, Wadhwani's calculations still suggest that share prices are high rather than low. On the other hand this sort of observation provides no guide as to the timing of an adjustment. For this reason we have tended to produce our forecasts on the assumption that current share price levels are the best guide to future levels. This approach is probably the best overall method for forecasting economic developments arising from the current level of the stock market, but it does mean that the consequences of sharp adjustments to share prices have to be presented as alternatives to the main scenario.

Implications for the real economy and the profile of the US slowdown

If stock markets did no more than represent current moods of optimism and pessimism among investors they would be of little consequence for economic management and economic policy. However they are important in two respects; first because they influence investment and secondly because they influence consumption.

The influence on investment arises because high share prices provide firms with a cheap source of capital and therefore allow them to carry out real investment cheaply. This means that more investment is undertaken than would be the case if share prices were not unusually high. Conversely, a fall in share prices leads to lower levels of investment demand and is therefore a contractionary influence on the economy.

Share prices also affect consumption levels. The assumption that much saving is motivated by the wish to provide for old age provides a reasonably coherent explanation of savings levels in advanced economies. Steadily-rising share prices reduce the pressure on people to save out of their income; conversely a fall in share prices is likely to lead to people cutting back on their consumption in order to rebuild their wealth.

Both of these variables are components of aggregate demand, and movements in them are likely to affect inflationary pressures. Central banks face the difficult task of adjusting interest rates to maintain an appropriate balance between supply and demand without giving the impression that they have any view as to an appropriate level for stock markets. Both these factors are present in developments in the United States and elsewhere. In its recent statement (Federal Reserve Release, 18 April, 2001), the US Federal Open Market Committee referred to weakness in capital investment and concerns about effects on consumption as reasons for reducing the federal funds rate. We expect a further reduction to 4 per cent per annum by the summer, and that this will be sufficient to maintain growth in the United States and to prevent a world recession.

In 2000 in the United States the output of technology industries rose very sharply. Output of computers in 2001Q1 was 28.8 per cent higher than a year earlier and output of high technology products, which includes communications equipment and semi-conductors as well as computers, rose by 36.9 per cent over the same period. Output of this industry was 1.3 per cent higher in 2001Q1 than it had been in 2000Q4. Conversely, non-energy non-high technology industrial production fell by 1.7 per cent in 2001Q1. It was 2.7 per cent lower than it had been a year earlier. This suggests that the slowdown in the United States comprises two elements. The rise in the volume of expenditure on high technology products almost certainly masks a fall in the value of demand leading to lower reported profits. Growth in the volume of output at a rate much slower than regarded as normal has led to the build-up of excess stocks. Staff who were recruited to meet demand which has not materialised are being laid off. The reduction in the growth rate of the volume of output of these industries is the most important factor behind the sharp reduction in the US growth rate.

At the same time the volume of output from the other industries has actually fallen. The biggest single cause of the output falls in the rest of industry is probably the high level of the US$ which has risen sharply against all major currencies since the start of 2000. It is possible also that the depressed demand in this sector has reduced enthusiasm for new investment in computer equipment compared to what might have been planned last year; the recent fall in share prices and consequent increase in the cost of capital to these companies has led to recent declines in new orders for investment goods.

We do not see any return to annual growth rates of 4 per cent or more in the United States. In the past it has had occasional years in which output and productivity grew rapidly as they did between 1996 and 2000. The period 1996-2000 was unusual in that a run of such years came together. Our projection assumes that this run of good luck has now come to an end and that growth in the future will be more like pre-1996 experience but the effects of the productivity boom on the level of output remain locked in.

Growth prospects in the United Kingdom and the rest of Europe

Despite the widespread concerns about possible effects of weakening financial markets on the real economy, our view of the European economies remains reasonably buoyant. We expect a growth rate of 2.2 per cent in Germany, 2.5 per cent in France and 2.4 per cent in the United Kingdom. The United Kingdom remains supported by a sharp increase in government spending. Slower growth of world trade is taking its toll on the external position of the UK, although fortunately from the UK's perspective, international trade is more buoyant in Europe than in other parts of the world. We now expect exports to grow at 5.5 per cent in 2001 as compared with the figure of 7.4 per cent in our previous forecast. Import growth, is forecast to be 6.3 per cent per annum, reflecting buoyant domestic demand and the balance of payments deficit is expected to widen to [pound]l6bn from a figure of [pound]l4bn last year.

The Euro Area cannot remain insulated from the factors affecting the rest of the world economy. We have assumed that the European Central Bank cuts the euro interest rate by 1/2 percentage point this year. Even with this we expect Euro Area growth to be slightly slower next year than it has been this year. Without such a cut the deceleration will be more marked.

Productivity and unemployment in the UK

Our earlier expectations of productivity growth of close to 3 per cent per annum in the UK are damped by the slower expansion of demand, particularly this year. The outlook has also been complicated by a substantial statistical revision raising the estimate of current employment by almost one million jobs. However, averaged over the next three years, we continue to expect GDP per employee in the UK to rise by more than its long-term rate, with rises of 2.6 per cent next year and in 2003. Manufacturing productivity appears to be extremely buoyant, although this increase in productivity is likely to lead to reduced employment rather than increased output.

One adverse consequence of the slowdown in UK growth is its effect on unemployment. The February unemployment figures were revised upwards, indicating that they remained above one million then. The March figures, showing unemployment below one million may suffer the same fate. In any case we take the view that the level of unemployment is set to rise modestly from current levels; we anticipate the claimant count rising to 1.1 million next year.

The UK fiscal position

Our projection of the public finances is more optimistic than the government's. This is an outcome of continuing tax buoyancy shown in the most recent data (which has a counterpart of slower growth in disposable household incomes and thus lower household saving than we had previously expected) combined, in the short term, with the view that the government is unable to meet its spending targets and in particular its investment targets. In the fiscal year 2000/01, public sector net investment was only [pound]4.3bn as compared with an estimate of [pound]7.4bn shown in the March Budget. Looking to the future we assume that by 2005/6 net investment reaches its target of 1.8 per cent of GDP, but that it rises to that level more slowly than the government currently intends. We also assume that the under-spend is not made good in future years.

These assumptions deliver a budget surplus which disappears by 2003/4; the current budget of course remains in considerable surplus. It is no real surprise that this position is more optimistic than the government's and there are two factors behind this. The first is the buoyancy of tax revenues in the first quarter of this year. We have assumed that much of this buoyancy persists. Secondly, we are projecting an average growth rate of 2.7 per cent per annum between 2000 and 2005, compared with the long-run growth rate of 2 1/4 per cent per annum assumed by the Treasury. This means that revenue rises faster than the government has assumed. However, given the inevitable uncertainty associated with projections this far in the future, it would be imprudent to say that there is bound to be room for extra public spending. On page 14 we discuss the implications for the budget of output growth being much slower than our main projection shows.

Interest rate prospects

So far this year the Monetary Policy Committee has reduced interest rates by half a percentage point. The rate of inflation is currently well below its target value. Tax changes announced in the Budget have the effect of depressing the inflation rate below what the Monetary Policy Committee would have expected 12--24 months ago when it was setting interest rates in order to influence today's inflation rate. Escalation of motor fuel duty has ended and the Chancellor of the Exchequer has not indexed alcohol duties. Tobacco duty has not been increased in real terms. The fact that, partly in consequence, the inflation rate may fall below 1 1/2 per cent, the lower limit of the inflation target range, should not be a major influence on current interest rate setting because current interest rates do not have a strong influence on the current inflation rate. However, current circumstances are bound to make the MPC more willing to cut interest rates than they would be if the inflation rate were at or above the centre of the target range. Thus we expect a further cut to interest rates to come relatively soon; without further disturbances in financial markets, we expect that to be adequate to keep the UK economy growing at a respectable rate. If equity markets recover to the levels seen in 2000, then the risk of medium-term inflation could start to outweigh the risk of below-trend output growth. Indeed one of the concerns we have is that, with a very low rate of inflation in the short term, it will be difficult for the MPC to raise the interest rate again even if conditions improve rapidly and the concerns which have led to it cutting interest rates begin to recede. Given the Government's plans for public spending the case for further interest rate reductions is weaker here than elsewhere.

Looking further ahead, the prospect of a General Election and a new parliament opens the question of the level of the inflation target. It was fixed at 2 1/2 per cent per annum and the Chancellor suggested it would be held at that level for the life of the existing parliament. Our forecast is constructed on the assumption that the current inflation target remains unchanged, but there are very good reasons for taking advantage of the current low rate of inflation to reduce the inflation target to 2 per cent per annum. In the short term it reduces the pressure on the MPC to react to the inflation-reducing effect of the budget. In the longer term it gives the prospect of lower interest rates and facilitates the process of convergence with the European Monetary Union. There, inflation is meant to be held in a range of 0-2 per cent per annum as measured by the Harmonised Index of Consumer Prices (HICP). The construction of this index means that it is likely to rise by 0.5-1 per cent less than the Retail Price Inde x excluding mortgages (O'Donoghue and Wilkie, 1998). Thus a reduction of the UK target to 2 per cent per annum (equivalent to 1-1.5 per cent per annum as measured by the HICP) would come very close to aligning our inflation target with that of the Euro Area.

Membership of the European Monetary Union seems less likely than it has for a considerable time. However, conditions are such that there is relatively little difference between monetary policy in the UK and Euro Area and projections show UK interest rates converging with those in the Euro Area after 2005. Since we assume that exchange rate movements reflect interest rate differentials, this implies that the euro stabilises against the pound from then on. The exchange rate at which we show the euro stabilising is [pound]1 = [epsilon]1.53.

REFERENCES

Bond, S.R. and Cummins, J.G. (2000), 'The stack market and investment in the New Economy: some tangible facts and intangible fictions', Brookings Papers in Economic Activity, 1, pp.61 - 108.

O'Donoghue, J. and Wilkie, C. (1998), 'Harmonised Index of Consumer Prices', Economic Trends, 532, pp. 34-44.

Wadhwani, S. (1999), 'The US stock market and the global economic crisis', National Institute Economic Review, 167, pp. 86-106.

 Summary of the forecast
 Probabilities [a]
 inflation Real gross
 target Output national Real
 met [b] falling [c] income [d] GDP [d]
2000 - - 2.9 3.0
2001 80 - 2.3 2.4
2002 53 1 2.6 2.6
 UK economy
 Retail price index [f]
 Manufact-
 uring Unemploy- Excl
 output [d] ment [e] All items mortgages
2000 1.6 1.60 2.9 2.1
2001 1.3 1.59 1.6 1.7
2002 2.1 1.68 2.3 2.3
 World economy
 Current Real Consumer World
 balance [g] PSNB [h] GDP prices [i] trade [j]
2000 -14.2 -18.6 4.8 1.7 12.9
2001 -15.9 -11.1 2.9 1.7 7.3
2002 -19.3 -4.9 3.2 1.4 7.1

(a.)In percentage terms.

(b.)Inflation excluding mortgages below 2 1/2 per cent per annum at the end of the year.

(c.)A fall in annual output.

(d.)Percentage change, year-on-year.

(e.)ILO definition, fourth quarter, millions.

(f.)Percentage change, fourth quarter on fourth quarter.

(g.)Year, [pound] billion.

(h.)Public sector net borrowing, fiscal year, [pound] billion.

(i.)OECD countries, percentage change, year-on-year.

(j.)Volume of total world trade, percentage change, year-on-year.