THE UK ECONOMY.

Blake, Andrew P. ; Young, Garry

Garry Young [*]

Section I. Recent developments and summary of the forecast

By the third quarter of this year, the economy had grown for thirty-three consecutive quarters since its trough in the second quarter of 1992. Such a long period of expansion inevitably raises questions about its sustainability. Chart 1 compares the paths for the level of output from the last two business cycle troughs in 1981 and 1992.

By this time in the 1980s recovery (roughly corresponding to the first quarter of 1989), output was 31 per cent higher than at its trough. This is substantially more than the 26 per cent increase recorded so far in this recovery, roughly equivalent to two years' growth. But it is also the case that by this time the 1980s expansion was only a year away from its conclusion. It is possible that the slower rate of growth in this expansion, especially over the last two years, may enable it to go on for much longer. Certainly, there are relatively few signs of the need for a slowdown now to ward off inflationary pressure, whereas interest rates had already risen to 13 per cent by the beginning of 1989. Chart 2 shows that the combined rates of unemployment and inflation (added together in a 'misery index') are now much lower than had been the case in the late 1980s.

The most recent news on inflation has been mixed with increases in the world price of oil and a fall in the value of sterling against the dollar having adverse implications for future UK inflation. In addition, the headline RPI inflation rate has risen to 3.3 per cent in the year to September, more than one percentage point higher than the 2.2 per cent increase in RPIX, the government's target measure of inflation.

Clearly the impact on inflation of a higher price of oil depends on whether it is sustained. Present indications, built into our main forecast, are that prices will peak in the current quarter before declining through next year. Consequently the impact on global inflationary pressure is relatively slight. The implications of a sustained increase in the price of oil are discussed in Box 1. This suggests that even a permanent 50 per cent increase will have a relatively small effect on general inflation as it would be largely offset by an appreciation in the sterling exchange rate.

Sterling has depreciated against the dollar by around 10 per cent since the end of last year. Typically this would add to inflationary pressure, but world prices denominated in dollar terms have also fallen over the same period, thereby lessening the impact. In addition, sterling has appreciated against some of the other major currencies, most notably the euro. In effective terms, it is currently around 2 per cent higher than it was at the end of last year and this is helping to damp inflationary pressure.

While higher oil prices have attracted a great deal of attention, inflation in other areas of the economy has been very muted. For example, the internationally harmonised index of consumer prices was only 1 per cent higher in September than it had been a year earlier. At the same time, the goods component of the RPI was only 0.5 per cent higher than it had been a year earlier, with much of the increase accounted for by petrol and oil which was up by 13 per cent and alcohol and tobacco, which was up by 4 per cent.

More significant has been the continued slow growth in average earnings, despite the large falls in unemployment seen in recent years. This makes one of the starkest comparisons between the current situation and that in the late 1980s. By the end of 1988, average earnings growth had accelerated to over 9 per cent on the national accounts measure of wages and salaries per employee. But in the second quarter of this year average earnings were only 3.7 per cent higher than they had been a year earlier, a slowdown in the growth rate since the second half of last year. In part this reflects a reduction in bonuses paid out this year, but excluding bonuses whole economy average earnings in August were 4.3 per cent higher than a year earlier.

The continued lack of inflationary pressure in the economy as a whole owes much to the fact that there has been little push for large pay increases despite the sustained fall in unemployment over the past seven or eight years. Indeed some measures of the state of the labour market continue to suggest that it is not particularly stretched at present, contrary to appearances given by claimant unemployment. For example, chart 3 compares the employment rate as measured by the Labour Force Survey against the number of workforce jobs per head of the working age population. The difference in the two series is accounted for by people who work beyond the statutory retirement age (about half a million women and 300,000 men) and those with second jobs who are counted twice in the workforce jobs series. The former, measuring the proportion of working age people in employment, is at an historically high level, exceeded in the last forty years only by the employment rate attained in the years 1973-5. By contrast, the latte r measure, while higher than through most of the 1980s and 1990s, is lower than it was in 1989 and 1990 and throughout most of the 1960s and 1970s. On this basis, the number of jobs in the economy does not appear particularly high by historical standards in relation to the population available to do them.

Our overall assessment is that the labour market is now broadly in equilibrium in contrast to the situation in the late 1980s when it had become too tight. The main basis for this view comes from the lack of any present inflationary pay pressure, but is supported by a consideration of the factors likely to affect the labour market. As we discussed in the April Review, the geographical distribution of jobs and vacancies is much more even now than it has been in the past. By the same token, the general increase in skills and qualifications in the labour force has meant that the lack of opportunities for unskilled workers is impacting on fewer people. While the employment rate of men without qualifications was low at around 60 per cent in Spring 1998, the number in this category was less than 2 1/2 million.

In addition, work incentives for the less well paid have probably improved as a consequence of the introduction of the Working Families Tax Credit and the National Minimum Wage. Further the opportunities for people to remain out of work while claiming benefits is being limited by benefit stringency associated with the Welfare to Work programme. On top of these factors adding to the availability of labour, the decline in the power of trade unions and greater job insecurity among professional workers are curtailing pay pressure.

Against this generally favourable background we believe that the economic recovery has a number of years to run in the absence of any significant adverse shocks or policy errors. However, the nature of the expansion is liable to change, with productivity growth picking up sharply, especially in the manufacturing sector, benefiting real wages and household incomes.

Monetary conditions (Table I)

The expected pick-up in productivity growth means that the economy can sustain a relatively fast rate of growth without provoking a large rise in interest rates. It also means that average earnings can grow for a while by a little more than the 4 1/2 per cent annual rate that is believed to be consistent with the inflation target in the long run. However, while we do not believe that there is a pressing need for an immediate rise in interest rates, the balance of risks suggests that some tightening of domestic monetary policy is desirable.

The decision of the Monetary Policy Committee (MPC) on 5 October to vote unanimously for no change in interest rates is surprising. Preceding decisions had been determined by the Governor's casting vote with the rest of the Committee split evenly between those who thought rates should rise and those who thought they should be unchanged. It is likely that the news of unexpected weakness in earnings was the main factor in causing a change of heart among some of the members.

One of the difficulties facing the MPC in setting interest rates is in deciding whether the economy has fundamentally changed in such a way that it can now sustain an above trend rate of growth for a number of years. In a recent speech, one of the MPG members stated that 'the main reason for believing that the UK economy might be behaving differently is that economic forecasters have, on average, significantly under-predicted GDP growth while simultaneously over-predicting inflation since 1992'. [1] This point is well-taken and supports the generally accepted view that the sustainable rate of unemployment in the economy is now substantially lower than previously thought. Chart 5 illustrates this point.

The chart shows the extent to which the out-turns for RPI inflation and output growth differ from National Institute forecasts made in October of the previous year. By historical standards, these forecasts have been quite accurate over this period, although it is true that we have tended to over-predict inflation and under-predict growth. When the inflation over-prediction is added to the growth under-prediction, illustrated by the line in the chart, it is apparent that 1997 was the only year out of the last six that turned out to be a disappointment.

However, this does not necessarily mean that there is a bias in the forecasting process and the economy will continue to perform better than forecast. This again points to one of the lessons of the 1980s. The chart also shows the degree to which expectations were exceeded in the 1980s at a similar stage in the recovery (so that 1984 is compared with 1995 and so on). In the five-year period from 1984 to 1988, the degree to which inflation and growth turned out to be better than expected was much greater than it was in the 1990s. It is not obvious that we would be right to interpret those errors as evidence of an improved economy at that time. This is borne out by the fact that forecasts for 1989 turned out to be too optimistic as inflation was significantly worse than expected. As is well known, economic performance in 1990 and 1991 was also extremely disappointing.

Clearly, macroeconomic policy runs the risk of repeating the errors of the 1980s. As noted above, we do not believe that there is a pressing need for an immediate increase in interest rates, but it appears that the damage that might be done to the economy by having rates too high is less than that which would come from having them too low.

In view of this, the MPC should raise interest rates to 6 1/2 per cent by the beginning of next year and we have assumed this in our forecast. We assume that rates then stay at this level for three years, before gradually converging on European levels of around 5 1/4 per cent at the beginning of 2005.

Our exchange rate projection gives a path consistent with euro entry in 2005 at a rate of [epsilon]1.65, although following the Danish referendum and the Prime Minister's change of heart, it is increasingly doubtful that this will happen. This rate is only slightly below the rate of [epsilon]1.68 that is assumed for the average of the current quarter in the forecast. This projection is based on market exchange rates and yield curves as of 12 October (although our forecast implies UK rates will be higher than those given by the yield curve). Thus we are not forecasting any substantial change in the exchange rate in the medium term, although exchange rate projections are highly uncertain. The forecast exchange rate for 2001 is about 6 per cent higher against the euro and 2 per cent lower against the dollar than in our last forecast.

Fiscal policy (Table 2)

The substantial improvement in the public finances seen over the past few years looks set to continue despite the increases in government spending confirmed in the Comprehensive Spending Review (CSR) in the early part of the Summer. The public sector recorded a surplus on its current budget of [pound]19.9 billion in 1999-2000 and looks well set to follow this with a similar surplus in the current fiscal year. This now looks like beating the Budget forecast of a current account surplus of [pound]13 billion.

Our forecasts for the public finances are shown in Table 2. The outlook for public spending is consistent with that shown by the Treasury in the CSR. Revenue now looks like being much stronger than was expected. As a consequence, both the current balance and the overall deficit are likely to be more favourable than previously thought.

Our forecasts of a continuing surplus on the current balance indicate that the government could afford to increase spending or cut taxes substantially and still meet its fiscal rules. Including the current year, the average surplus on current account over the next six-year period is projected at around [pound]18 billion. Put in the context of current disputes about fuel taxes and state pensions, this is only slightly less than the [pound]22.5 billion raised by the taxation of all hydro-carbon oils, including petrol, in 1999-2000. Alternatively, it represents about 50 per cent of spending on the state pension. In view of recent interest, Box B discusses the costs of increasing the basic state pension.

Summary of the forecast

Provisional ONS estimates indicate that output grew by 0.7 per cent in the third quarter of this year to a level 2.9 per cent higher than a year earlier. We are expecting growth at a slightly faster rate in the fourth quarter as a rebound from the fuel crisis to give a rate of just over 3 per cent for the year as a whole.

Domestic demand is set to expand by around 3 1/2 per cent this year, similar to the rate seen in 1999. The negative contribution to growth from net trade is forecast to fall from 1.5 per cent in 1999 to around 1/2 per cent this year as export growth of around 8 per cent is only slightly exceeded by import growth of 8 1/2 per cent.

It appears that the service sectors are continuing to grow more quickly than manufacturing in 2000. Nevertheless, manufacturing is expected to grow by about 1 1/2 per cent in the year as a whole. Growth in public sector output in the first half of the year has not been as rapid as we previously expected as a consequence of weaker spending and is now set to grow by about 2 per cent in the year as a whole. Stronger growth is apparent in the services and construction sectors.

As we have previously discussed, the economy must now be approaching the point at which further employment growth will generate excessive inflationary pressure. Instead, with little slack in the labour market, much of the expected growth in output has to be met by increases in productivity. Productivity growth is expected to be particularly strong in manufacturing where it reached 3.5 per cent in 1999 in difficult trading conditions. We expect to see productivity growth in manufacturing of about 4 per cent this year, in line with its average over the 1980s and first half of the 1990s. This will contribute to a rate of productivity growth in the whole economy of around 2 1/2 per cent, the fastest rate since 1994.

The combination of increasing aggregate demand and a relatively tight labour market is expected to put some upward pressure on wages. But slower earnings rises in the middle of the year will limit growth over this year to about 4 per cent. With consumer price inflation remaining subdued, this represents an increase in real wages of around 2 1/2 per cent.

Real household disposable income is forecast to grow by about 3 1/2 per cent in 2000, similar to last year. With household consumption forecast to grow at a similar rate, the saving ratio is projected to remain quite low at just over 4 per cent down from 5.1 per cent in 1999.

The various influences on household spending, including financial and housing wealth, interest rates and unemployment are all supportive of a relatively low saving ratio, but it has now reached an unusually low level and we are expecting some modest increase over the next few years. We are expecting the growth in household consumption to slow from 3 1/2 per cent in 2000 to 3 per cent in 2001.

Although household consumption growth is slowing, the growth in domestic demand is forecast to remain at around 3 1/2 per cent per annum over the coming three years. This is partly driven by increased public spending. Government consumption is expected to grow by 4 per cent in 2001, and government investment is set to rise by almost 20 per cent. With the negative contribution from net trade waning, the overall growth rate is forecast to rise to 3 1/2 per cent in 2001.

On the basis of this central forecast, we calculate that there is only a slim chance of output declining in 2001. Indeed, we estimate that the chances of growth being below 2 per cent by the end of the year are only about 20 per cent. Most of the risks are for growth being very strong with a probability of it exceeding 4 per cent being put at about 30 per cent.

We are now expecting some slight shift in the sectoral pattern of output growth in 2001. Manufacturing output is expected to grow by about 3 per cent, and continued pressure on profit margins will encourage further productivity growth of about 5 per cent in 2001. Job losses in this sector will be more than made up for by employment gains in the rest of the economy. With productivity growth rising to around 3 per cent per annum, aggregate employment is projected to rise at around the same rate as the workforce and this will reduce unemployment only modestly from current levels to about 5 1/4 per cent on the ILO definition by the end of next year.

Against this background, inflationary pressures are expected to begin to build up, although from a very weak position. We are expecting RPIX inflation to remain below 2 1/2 per cent until around the middle of 2002, with an expectation that it will fall below 2 per cent in the short term. The probability of inflation being above target at the end of 2001 is put at 38 per cent, with an 18 per cent chance that it is above 3 1/2 per cent. Our central estimate is that inflation will be at 2 1/2 per cent by the end of 2002.

NOTE

(1.) 'Monetary Challenges in a New Economy', speech by Sushil Wadhwani to HSBC Global Investment Seminar, 12 October 2000. Emphasis in original.

The effects of higher oil prices on the UK economy

Garry Young

The sudden rise in the price of oil over the past year has raised a number of concerns about its economic impact. Most immediately, consumers and industrial users of petrol in a number of countries have complained bitterly about the effect of the high price on their living standards. At home, the government has been under intense pressure to reduce duties on petrol and diesel to mitigate the effect of high prices. Some calculations have even suggested that the government can afford to do this because higher crude oil prices raise its receipts of both petroleum revenue tax (PRT), reflecting higher profits on oil production, and VAT on petrol.

However, this takes a rather narrow view of the overall, macroeconomic impact of higher oil prices. The experience of the oil shocks of the 1970s and 1980s suggests that the effect on living standards extends beyond the expense of higher petrol prices. Once the macroeconomic impact is taken into account, it is clear that there are few winners in this country from higher oil prices. While the government may be one of the winners in the short term, in the longer term it too loses as tax receipts suffer in a weaker economy.

Our overall assessment of the world economy and the oil market suggests that prices should moderate from the high level reached recently. However, there is a risk that this is too optimistic a view and that prices will be driven much higher. In the World Economy chapter of this Review, we discuss the effect of a permanent increase in the price of oil to a level 50 per cent higher than in our main forecast giving a figure of $40 in 2001. According to our analysis, this would reduce the level of output in the world economy while forcing up interest rates to prevent the higher price of oil feeding through into higher general inflation.

The induced slowdown in the world economy combined with higher interest rates and a higher price of oil would quickly impact on the UK. Given the inflation target, interest rates would also rise here and sterling would probably rise against other currencies, partly reflecting the UK's position as an oil producer. In addition to the effect that these changes would have on aggregate demand, there would also be an impact on aggregate supply. In particular, wages would have to fall relative to prices to accommodate the higher price of petrol. In the medium term, some rise in unemployment would be necessary to sustain a lower real wage. The extent of the rise in unemployment would depend on the extent of real wage resistance. Profits outside of the oil producing sector would also fall, possibly having a detrimental effect on investment. Table A summarises the quantitative impact on the UK economy.

Given the large rise in the oil price, the simulated impact on the UK economy is relatively small, with output about 1 per cent lower than it would otherwise have been after ten years. This change is in line with that estimated for the major 7 economies. There is relatively little effect on retail prices, reflecting the assumption that monetary policy pursues an inflation target. The rise in interest rates is also small. This is partly due to the fact that a 4 per cent exchange rate appreciation does much of the work that higher interest rates would otherwise need to do. Although the government budget benefits from higher oil prices in the short term, after ten years the surplus would be smaller by about [pound]3.5 billion per annum as extra oil taxes fail to compensate from lost revenue due to lower activity. This contradicts the claim made by some commentators that the government can afford to cut fuel duties in response to higher oil prices.

The true cost of increasing old-age pensions

James Sefton and Martin Weale

The cost of old-age pensions, like other benefits, has to be assessed taking account of changes to the age structure of the population. A pay-as-you go pension scheme which is affordable to a population with relatively short life expectancy and thus relatively few old people can become insupportable if increased longevity results in the proportion of retired people to people of working age rising markedly. Of course the situation is made even worse if low fertility rates lead to the population of working age declining.

This means that, in discussing the cost of any pension increase, it is prudent to consider not what increase in taxes is needed to meet that increase in the current year, but what increase in tax rates is needed to ensure that the capitalised value of the extra tax revenue matches the capitalised cost of the increased pensions. Both the tax and the pension payments are capitalised taking the effects of demographic change on tax revenues and pension payments into account. The calculations are most helpfully done in the context of generational accounts which show the capitalised value of tax payments and receipts of benefits for men and women at each point in their lives. Taxes are typically paid during working life while benefits are received much more by children and old people. For a young adult the discounted value of tax liabilities is considerably larger than the discounted value of benefits received, while for someone over 65 the opposite is true. The accounts are described in considerable detail on the National Institute's Generational Accounting web site (www.generationalaccounting.com) and the results we present here are calculated from the models which can be downloaded from there.

We explore two changes to current pension arrangements. First of all we look at a 10 per cent increase to old-age pensions, while secondly we look at the implications of the restoration of the earnings link. We find that, if a 10 per cent pension increase is to be financed by an increase in income tax, then the base rate of tax needs to rise by about 0.75p in the pound. On the other hand, if indexation is to be re-introduced, then the basic rate of income tax needs to rise by about 5p in the pound. These figures take into account an allowance for a fall in payments of means-tested benefits to old people which would follow from an increase to the basic pension.

We show in Table B the impact of these changes on the capitalised income tax burdens and pension benefits for men currently aged 30 and 65. The table shows that, for the young man, indexation results in a doubling of pension benefits but an even larger absolute increase in the tax bill. The 10 per cent one-off increase results in smaller increases to both numbers, but the increased tax bill is still larger than the present value of the increased pension. For the 65-year old the situation is reversed. Indexation is not worth a great deal more than a 10 per cent increase and in either case the increase to pensions is much larger than the increased tax bill. These effects arise because, obviously enough, the 65-year old is reaping the benefit of an increased pension without having to pay for it. Less obviously, but as a direct counterpart, the 30-year old is paying more out than he expects to receive back because he has to finance not only his own increased pension but also those of the older people who will not be working for long enough to pay for the cost of their own old age.

Section II The forecast in detail

The components of expenditure (Table 3)

It now seems likely that the economy will grow by just over 3 per cent in 2000, although the fuel crisis has obscured the picture to some extent. The latest retail sales figures indicate strong growth in high street spending, with the three months to September 1.6 per cent higher than the previous three months and 4.3 per cent higher than the same period in 1999. Second quarter government consumption increased by 1.9 per cent to recover from the fall in the first quarter. This accelerated growth is needed to meet spending targets. Our forecast is for an overall 2 per cent growth rate in the current year rising to 4 per cent next year. Gross fixed investment was some 6 per cent higher in 1999 over the previous year, and we are forecasting this to slow to 2 1/2 per cent for the current year, rising to 5 per cent for the next two years. Inventory accumulation should recover from the de-stocking evident last year.

The high growth in domestic demand next year of 3 3/4 per cent will maintain the current high growth rate, suggesting some monetary tightening would be prudent. We forecast domestic demand to continue to grow strongly in 2002. Although the rate of growth in world trade is expected to slow from this year's very high rate, we still expect the negative contribution of net trade to GDP growth to fall to almost zero by 2002.

Household sector (Table 4)

Some recovery of the saving ratio from the very low second quarter figure can be expected, with continued strong growth in real disposable income. Total compensation of employees grew by 6.2 per cent in 1999, and is projected to grow by 5 1/4 per cent this year. This slowdown reflects the slower than expected growth in earnings despite the tightness of the labour market. In Chart 6 we plot the savings ratio in the UK and by comparison that for the US. The 3 per cent figure for the second quarter is by no means as low as recent US experience but it is unusually so for the UK. The surge in domestic housing wealth has presumably contributed to reduced saving, but we are forecasting a modest recovery in the savings ratio as the household balance sheet weakens with slower growth in the value of equities and a considerably cooler housing market. Despite low interest rates, we are forecasting house price inflation to have peaked and to be a (by recent standards) moderate 51/2 per cent in 2001. However, overall real disposable income should grow by about 31/2 per cent this year with a slightly higher growth rate for household expenditure leading to the falling savings ratio.

Durables expenditure by households was largely static between quarter one and quarter two. However, non-vehicles expenditure was actually up. Household spending on vehicles fell by 1.1 per cent in the second quarter with the continuing uncertainty over the expected price of new cars. The general anticipation of falls in prices earlier this year has to a large extent been confirmed. Both foreign and domestic producers have lowered their UK prices, with Ford, Rover and Vauxhall all announcing price reductions in October alone. The CBI Distributive Trades survey confirms a continuing overall weak picture for car sales. Competition among manufacturers seems set to increase, with several offering further discounts to Internet buyers. It remains to be seen if the trend for lower prices will continue and how consumers will respond. We are forecasting an increase in overall durables expenditure with an annual growth rate of nearly 7 per cent for 2000. This is despite a fall in vehicle prices, which has yet to bring a bout a strong recovery in that sector, perhaps in the expectation of even lower prices. We therefore expect stronger durables expenditure growth next year of 7 1/2 per cent to offset this.

Fixed investment and stockbuilding (Tables 5 and 6)

Total fixed investment for the second quarter was 1.9 per cent higher than a year previously, the smallest such growth rate for some five years. The national accounts breakdown indicates that increases were recorded for investment in dwellings, other machinery and equipment and transport equipment. Without the 6 per cent growth in this relatively small last category overall investment would have remained static.

Although general government investment rose strongly in the second quarter, it was less than enough to sustain the reasonably bullish investment expectations we had in our last forecast. With half the data for the year available, public sector investment is lower than we would have expected. This illustrates the considerable uncertainty about the timing of public sector investment, which has large scope for growth within set spending limits.

Manufacturing investment is forecast to increase by 5 per cent in 2000. By contrast, non-manufacturing business investment is forecast to remain comparatively static, with growth falling from the very high 13 per cent in 1999 to a projected 2 per cent this year. Total investment should therefore grow by 2 1/2 per cent in 2000, and we are projecting some recovery next year and the year after, with 4 per cent growth in each.

We illustrate this in Chart 7, where we plot the annual growth rates for both general government investment and private sector non-manufacturing investment. This illustrates nicely the marked difference between the two. Public sector investment growth rises strongly from the low of 1996. By contrast, private sector non-manufacturing investment growth has slowed markedly to leave the overall growth rate for 2000 below that of GDP.

Balance of payments (Tables 7 and 8)

World trade in 1999 grew by 6 per cent. It has since accelerated to a forecast 11 1/2 per cent growth rate for 2000. Whilst we forecast a subsequent slowdown from this unsustainably high level, the still buoyant world economy should affect UK exports favourably. As we have commented before, a persistently high value for sterling has caused a loss of market share for UK exporters, which would have been greater without sharp export price cuts. Even the recent devaluation of the exchange rate against the dollar has been partially reversed, with an overall 2 per cent appreciation in the effective rate since last year, due, in large part, to movements against the euro. But export price competitiveness has improved by nearly 3 per cent since the beginning of the year as a consequence of lower export prices.

Export volumes grew by 2 per cent in the second quarter, helped by buoyant sales to non-EU countries. However, the net trade position has worsened slightly with import volumes rising by 2.2 per cent. Imports of capital and intermediate goods from non-EU countries were particularly strong.

Our forecast for manufactured export growth this year is now particularly buoyant at some 12 1/2 per cent. This is in marked contrast to the weakness of two years ago. Growth of 10 per cent is expected next year despite an improvement in export profit margins.

The overall balance of payments is expected to deteriorate slightly, with the current account deficit as a percentage of GDP rising to 1 1/2 per cent by 2001 from 1.15 per cent in 1999. We are now expecting the deficit on the balance of trade in manufactures to be [pound]24 1/2 billion. This would be a record value, but we forecast it to subsequently to decline.

In Chart 8 we show annual growth rates for both exports and imports of goods. It is clear that import growth has considerably exceeded export growth since 1996, but the considerable growth in exports in 2000 seems likely to turn that around, and we forecast export growth to exceed import growth both next year and in 2002.

Output and employment (Tables 9 and 10)

Service sector output rose by 0.9 per cent in the second quarter, with every category contributing positively.

This was outstripped by the 1.4 per cent rise in production. Although this last increase is composed of less than a 1/2 per cent growth in manufacturing output, electricity, gas and water supply was up 6 per cent, and mining and quarrying, which includes oil and gas extraction, was up 4.9 per cent. Output in the oil sector is forecast to rise further, reflecting in part a response to the increase in the oil price. The strong possibility is that oil output will increase still further in response to recent global shortages.

The August figure for the index for the output of production is 1.1 per cent higher than three months earlier, although the third quarter GDP figures point to a pause in September. The so-called 'new economy' sectors led the increase in manufacturing output. As we commented at the time of the last forecast, businesses are believed to have spent considerably on infrastructure to enable the use of new information technology. Some part of this spending may have been to pay for upgrading associated with concerns about year 2000 effects pointing to weaker demand in the future. Also some manufacturers of mobile telephones, a key component of IT spending, have warned that growth in demand for their product worldwide is expected to slow sharply. There is no reason to discount this happening in the UK. But despite these concerns we expect overall manufacturing output to grow by more in 2001 than it has in 2000.

The labour market currently provides mixed signals. We comment on the surprisingly sluggish behaviour of average earnings below, while employment growth remains quite strong. The most recent labour market data show an ILO unemployment rate for June to August of 5.3 per cent, falling from 5.6 per cent in the previous quarter and the almost 6 per cent a year before. We forecast the unemployment rate as defined by the ILO to fall still further to 5.2 per cent by 2002. The seasonally adjusted claimant count for September was 1.04 million, falling by 16,400 from the previous month. We expect this to stabilise.

Earnings and prices (Tables 4 and 11)

As noted above, skill shortages are widely reported and unemployment continues to fall. Conversely, wage growth remains comparatively modest. Average earnings fell slightly in the second quarter, with the annual growth rate 3.7 per cent. In 1999 average earnings grew by 5 per cent, and a large part of the slowdown in growth is attributable to the aftermath of large bonuses paid at the end of last year. As Chart 9 illustrates, this pattern of quarterly growth is unusual, and we are forecasting a recovery of earnings growth. This does not necessarily indicate an increase in inflationary pressure. Forecast productivity growth can easily sustain the 5 1/4 per cent average earnings growth predicted for each of the next two years, without pushing the inflation rate above its target value.

Whole economy productivity growth was 1.4 per cent in 1999, and is forecast to be 2 1/2 per cent in 2000 with a continued strong performance by manufacturing. Unit labour costs grew by 4 per cent in 1999, but we expect this growth rate to halve in the current year with increased productivity and low earnings growth. Thus taken as a whole, it seems inevitable that the labour market must translate some of its tightness into higher wage inflation, but we suggest that a modest interest rate rise will be enough to contain this.

In fact, consumer price inflation jumped unexpectedly in the most recent figures. A large part of this is attributable to the increased cost of motoring, which remains a considerable source of uncertainty. RPI growth remains above consumer price inflation, largely reflecting the rise in base rates that happened at the end of last year. The Bank of England's target inflation measure, RPIX, is forecast to rise by 2 per cent in the current year, and 2 1/4 per cent next year, slightly below the central target of 2 1/2 per cent. This forecast is, of course, predicated on our assumption of slightly higher interest rates over the next year. We persist in assuming that the UK will join the Euro Area around 2005, but this assumption is of little importance for our forecast.

National and sectoral saving (Table 12)

Saving and investment in the UK economy are expected to rise gradually over the medium term to reach 18 1/2 per cent and 19 1/2 per cent of GDP respectively by 2004. Most of the increase in investment spending is due to a projected rise in government investment. The forecast increase in saving mainly reflects more saving by households, partly offset by lower government saving. Because the rise in saving and investment are broadly similar, the current account deficit will remain at around 1 to 1 1/2 per cent of GDP. The current account deficit of the economy as a whole is a reflection, subject to a statistical residual, of the financial position of the individual sectors in the economy. Table 12 shows how the imbalances between the saving and investment of the individual sectors is resolved. Ultimately, any investment that cannot be financed by domestic saving needs to be financed abroad.

Household sector saving is expected to be about 3 per cent of GDP this year, but picking up in the years ahead. At present saving by households is less than investment resulting in an unusual financial deficit. Company sector saving is expected to remain below investment such that the shortfall of saving relative to investment is of the order of 2 per cent of GDP. The government sector is now meeting the Golden Rule so that its saving is positive and is forecast to remain above its investment over the next three years.

The current account moved into deficit last year and is expected to remain at a little over 1 per cent of GDP in the coming four or five years. It is fairly clear that the movement into deficit of the current account of the balance of payments is associated with the increased deficit of the private sector. Household saving fell sharply in 1998 and has remained low since. For companies, the move into deficit can be partly traced back to the strength of the pound which has encouraged spending to be switched away from domestic goods and towards foreign goods, thereby reducing profits and company saving. If this situation were to continue then companies would at some stage need to raise their saving or reduce investment to prevent their indebtedness rising too quickly. This would feed through either directly, through lower spending on foreign investment goods, or indirectly through the effects of lower dividend payments on household spending on imports, to the current account. The process of adjustment would also affect prices in such a way that competitiveness would be eventually restored. Through these channels, an excessive current account deficit would be ultimately corrected.

The economy in the medium term (Table 13)

The way in which the economy behaves over the medium term is determined partly by a range of shocks that are inherently unpredictable. But there are other important influences on its development that can be foreseen. These include trends in the size and composition of the population, forthcoming changes in the policy framework as well as adjustments to existing disequilibria. As we have noted, the economy as a whole is close to an equilibrium position at present, but there are a few imbalances that will be adjusted in the medium term. Among these is the current weakness of manufacturing due largely to the overvalued exchange rate.

We continue to assume that sterling's rate against the euro will stabilise in the fairly near future. This is consistent with the market estimates of future interest rates. We have assumed that sterling will be stable at [pound]1.65 (equivalent to DM3.23) from 2003 onwards, a picture broadly consistent with market expectations. Many would regard this rate as being too high and it certainly implies a higher real exchange rate than the UK has been able to sustain historically. But since the end of 1996 the economy does appear to have adjusted to a high exchange rate. At first exports appeared not to be affected by the strong pound, but they then weakened sharply both in 1998 and in the first half of 1999. The contraction in the manufacturing sector was also a consequence of the high pound. But we now see grounds for expecting the growth of exports and manufacturing output to recover without a substantial depreciation of the nominal exchange rate. However, we are not expecting the ground lost over the last few y ears to be made up; exports will remain lower than they would have been if the euro were stronger.

The increases in public spending announced in Budget 2000, especially with respect to investment, will have an expansionary impact on aggregate demand over the medium term. This will put some upward pressure on domestic inflation and the current account deficit which will however decline over the coming decade as households and firms bring saving and investment into line.

Inflation is forecast to remain low over this period. With the sterling exchange rate fixed, there will not be room for large persistent differences in inflation across the Euro Area. So long as this is clearly understood, expectations, which are crucially important in the inflationary process, should help anchor inflation itself.

The outlook for interest rates and inflation is consistent with long-term real interest rates of around 2 per cent. This is much lower than has been normal in the UK over the past twenty years, but is consistent with real yields on UK government index-linked debt. Short-term interest rates are expected to rise slightly, reaching 6 1/2 per cent in 2001. They are then projected to fall, converging on euro rates by the beginning of 2005.

The government has identified slow productivity growth as one of the important problems of the UK economy and it is possible that policy action over the coming years will be successful in raising it. We have made no special allowance for this. Nevertheless, we are forecasting some above trend growth in the early part of the decade as productivity makes up some of the ground lost in the second half of the 1990s. This, should the forecast prove correct, will allow a period of rapid income growth. At the same time the forecast is a warning that, without this rapid productivity growth, the economy risks an unsustainable boom.

Forecast errors and probability distribution (Tables 14 and 15)

Table 14 provides a set of summary information as regards the accuracy of forecasts that have been published in the October Review. The latest complete National Accounts information available when these forecasts are constructed is for the second quarter of the current year.

A rule of thumb is that a 70 per cent confidence interval for a variable of interest can be obtained by adding a range of one absolute average error around our central forecast. Thus we can now be 70 per cent sure that GDP growth in 2000 will be between 2.7 and 3.7 per cent. The size of the average errors indicates that some variables are easier to forecast than others. For example, the errors in forecasting consumers' expenditure are smaller than those in forecasting fixed investment. It is also the case that the error in forecasting GDP growth is smaller than that made in forecasting its components. This arises because of offsetting movements among the components.

The probability distributions around the growth and inflation forecasts have been calculated assuming that the distributions are normal and are shown in table 15. The standard errors are based on the historical forecast errors underlying table 14. These estimates of the probability distribution around our central forecasts provide a quantitative assessment of the limits of our forecasts. Our estimates suggest that there is now a 78 per cent chance that the four-quarter growth rate at the end of this year will turn out to be in the range of 2 to 4 per cent, with a negligible chance of growth below 2 per cent. For next year, growth prospects are much more uncertain. However, it is now thought very unlikely that it will be below 1 per cent. There is a 67 per cent chance that it will be between 1 and 4 per cent and a 28 per cent chance that it will be above that.

For inflation, there is a 51 per cent chance that it will end the year between 1.5 and 2.5 per cent with a significant risk that it will be below 1.5 per cent. The risks are clearly larger looking ahead. In two years time, there is an evens chance that inflation will be above 2 1/2 per cent, consistent with the inflation target. However, the imprecision of the forecast is shown by the fact that there is almost a 30 per cent chance that it will be below 1 1/2 per cent. The chance that it will be more than 1 per cent away from target in either direction is put at 56 per cent.

(*.) The forecast was compiled using the latest version of the National Institute Domestic Econometric Model. We are grateful to Ray Barrell, Richard Kneller, Nigel Pain, Rebecca Riley and Martin Weale for comment and discussion.

AT A GLANCE...

The UK economy

* The Bank of England needs to raise the base rate to 6.5 per cent by the start of 2001.

* There will be a [pound]15bn budget surplus this financial year, [pound]10bn more than projected by the Treasury in the March Budget. The current budget will be in surplus by [pound]19bn.

* Restoring earnings indexation for the basic state pension would require around a 5p increase in the standard rate of income tax, according to the National Institute's generational accounting model.

* The economy will grow at over 3 per cent a year in 2001 and 2002, continuing the second longest sustained expansion since the war.

* While there is little room for further declines in unemployment, the labour market is in broad balance, and strong productivity growth will offset an increase in the growth of average earnings to above 5 per cent a year.

Contrary to expectations in financial markets, interest rates should rise further, by half a point by the beginning of next year. The continuing economic expansion and decline in unemployment now mean that the risks of monetary policy being too loose outweigh the risks of it being too tight. In addition, the rise in oil prices adds a modest inflationary impulse which should be countered through monetary tightening.

The public finances continue to be bolstered by strong economic growth. For 2000/01, we are forecasting a surplus of a similar magnitude to last year's record. The run of surpluses will continue until 2003/04 when the deficit will be much lower than the Treasury's projection. However, the boost to tax revenues from North Sea oil production from a rise in oil prices does not give scope for a cut in fuel duties. A simulation of the impact of a sustained increase in the oil price shows that short-term budgetary gains are followed by longer-term revenue losses caused by lower economic activity.

The present strength of the public finances appears to make more generous state pension provision eminently affordable. However, a proper assessment should take into account the effects of population ageing on a pay-as-you-go pension system. The National Institute's generational accounting model shows that the basic rate of income tax would need to be 5p higher if pensions are indexed to earnings than if they remain indexed to prices.

The sustained economic recovery since 1992 will continue and strengthen over the next two years. Compared with a similar point in the recovery of the 1980s, overall output growth in the current expansion has been less vigorous. GDP has risen by 26 per cent since the 1992 trough, compared with 31 per cent from the 1981 trough to 1989 -- a shortfall roughly equivalent to two years' growth. However, there is no sign as yet of the severe imbalances that brought the 1980s recovery to an abrupt end.

In particular, the labour market is in much better shape. Average earnings growth is running at less than half the rate of the late 1980s. However, there is little scope for further falls in unemployment. There will be some modest increase in wage pressures over the next two years -- earnings growth is forecast to grow above 5 per cent a year -- but this should be offset by a rise in annual productivity growth to around 3 per cent.