The UK economy.

Kneller, Richard ; Riley, Rebecca ; Young, Garry 等

Section I. Recent developments and summary of the forecast

There has been a very sharp deterioration in confidence about the economic situation in the UK over the past three months. This has been highlighted in a range of business and consumer surveys. It is also evident in profit warnings from the corporate sector, falling asset prices, high profile job losses, downward revisions in economic forecasts and an easing of monetary policy. This pessimism is not now confined to the manufacturing sector as appeared to be the case in the earlier part of the year, but seems to pervade all sectors of the economy.

Partly the gloom reflects new information about the state of the world economy. This is discussed in detail in the World Economy chapter of this Review. Problems which arose outside of the major North American and European economies are likely to have a substantial impact there through their effect on trade and asset prices. More uncertain, but possibly more important, is the effect of these distant troubles on the world financial system. The fear of many is that the losses made by banks and other financial institutions in emerging markets will lead to a rise in the cost of financial intermediation and a possible absence of new credit. This could threaten the smooth expansion of demand in the major economies. At present, there is considerable uncertainty about how significant these effects might be, contributing to substantial volatility in world asset prices in recent weeks. We have revised down our forecast of GDP growth in the OECD economies in 1999 by 1/2 per cent to 1 3/4 per cent.

However, it needs to be stressed that the UK economy was already slowing down in response to domestic policy tightening. Much of the pessimism now evident is a direct consequence of deliberate policy action to slow the economy from an above trend rate of growth and increase the chances of meeting the inflation target. There is now a very clear risk that the momentum behind the slowdown will be supplemented by contractionary forces from the rest of the world that could drive the economy into a recession.

In order to give an indication of the quantitative importance of the current pessimism, Chart 1 plots the four-quarter growth rate of GDP against a measure of business optimism from the CBI survey. The chart also shows a projection of the growth rate based on a simple model of the relationship between these two series. This [TABULAR DATA FOR TABLE 1 OMITTED] would predict falling output from the fourth quarter of this year to the third quarter of next, a technical recession, and a zero growth rate in 1999 as a whole.

In our view, this prediction is too pessimistic. While there are a range of contractionary pressures that raise substantial risks for the domestic and world economies, it is probable that a recession can be avoided.

It is useful first to set out the main possible sources of declining demand. In addition to the risks to the global economy affecting the UK through trade, there are three main other potential problems.

First, there is the danger that equity and other asset prices decline substantially. After falling by 25 per cent from a peak in late July to a low of 4649 in early October, the FT-SE 100 had made up over a third of the losses by 23 October. This volatility makes it difficult to assess whether the recent rise is a genuine recovery or an upward blip on an otherwise downward correction; although we note that with reference to most fundamental measures, equities are still grossly overvalued. The reduction in the level of equity prices so far is likely to have had only a modest impact on spending. We have assumed that this partly reflects a rise in the required rate of return on risky assets, raising the cost of capital for business finance. This, together with lower household wealth, would reduce growth next year by about a quarter of a percentage point. More sustained falls in equity prices would intensify these effects.

Second, there is the danger of a credit squeeze. There is evidence that the losses made on domestic lending by banks and other financial institutions in the early 1990s caused them to act in such a way as to intensify the recession.(1) By raising the cost of credit, usually indirectly by insisting on more demanding terms or closing credit channels altogether, the recession was made worse than it might otherwise have been. Information in this area is so far very patchy, but there is no evidence that the major British banks have sustained losses in any way comparable to those of the early 1990s. It has been suggested unofficially that the losses made by one of the major British banks in emerging markets are of a similar magnitude to their profits elsewhere; losses of this order are easily manageable and would not be expected to impact on the bank's domestic operations. However, if the major British banks have incurred substantial losses on their overseas lending, then there is a risk that this could lead to increases in the cost of intermediation similar to those experienced in the early 1990s.

Third, there is the danger that domestic spending declines further because of a delayed response to earlier policy tightening. In previous forecasts, before the recent turmoil in world financial markets, we have drawn attention to the risks that the domestic economy might go into recession. These risks are still present.

Against these dangers there are a number of grounds for optimism. First, unlike in the early 1990s, private sector balance sheets are very strong. Then, companies and households had borrowed heavily in the expectation of further growth in profits and asset prices. The combination of falling property prices and higher interest rates required a large correction in balance sheets, involving lower spending and more saving. The private sector is not now vulnerable to shocks to underlying credit conditions. Second, and most important, is the awareness of policymakers in the UK and abroad, of the dangers faced by the global economy. The recent cut in interest rates in the UK suggests a willingness on the part of the Monetary Policy Committee (MPC) to act in response to these dangers. This is in marked contrast to the policy reaction in the last two recessions, when there was no easing of policy because of the monetary regime then in force (the ERM in the early 1990s and monetary targets in the early 1980s). Additionally, the new public spending plans announced in July, allowing a fiscal expansion, provide a counterpart to any fall in private demand. Thus, the combination of healthy private sector balance sheets and a willingness to ease monetary and fiscal policy reduces the risks of a very hard landing for the UK economy.

Monetary conditions (table 1)

As with equity prices, exchange rates have been particularly volatile in recent months. Since our last forecast the exchange rate has fallen by about 5 per cent in effective terms, rising 3 per cent against the dollar and falling by 6 per cent against the DM. By our estimates, the pound is still substantially over-valued, although not by as much as in July. We continue to assume that for the UK to join EMU in 2003, sterling will be fixed at around DM 2.70 at the beginning of 2001. This is now broadly consistent with the rate implied by market interest differentials.

With underlying inflation now at 2 1/2 per cent, and with an increasing likelihood of even lower inflation, there appears to be ample scope for large cuts in interest rates. This forecast is based on there being a further quarter point cut in interest rates to 7 per cent in November, with more cuts throughout 1999 so that rates end the year at 5 3/4 per cent. Further reductions to 5 per cent are then assumed to take place in 2000, bringing UK interest rates broadly into line with those in Euroland.

This slow reduction seems consistent with the gradualist approach adopted by the MPC over the past year or more. However, given the very clear downside risks to the economy apparent at present, a somewhat more rapid adjustment of rates might be appropriate.

Fiscal policy (table 2)

The government's Economic and Fiscal Strategy Report (EFSR), set out its fiscal strategy and its overall spending plans for the years ahead. Strategically, the government stressed the distinction between current and capital spending and committed itself to borrowing only what is necessary to finance capital investment over the economic cycle. It also committed itself to maintaining debt at prudent levels, thereby limiting the amount that it was prepared to spend itself on capital items.

Table 2 sets out our assessment of the public finances. The outlook for public sector spending is broadly that shown in the EFSR. Total Managed Expenditure (TME), the sum of current and capital spending, is set to grow by an annual average rate of about 4 per cent over the next three fiscal years. Within this total, net capital spending is due to more than double, but from a very low base.

It is worth noting that there is likely to be a sharp rise in the share of current spending devoted to goods and services by the government. At present this amounts to roughly half of TME, with the remaining half including transfers, debt interest and capital depreciation. To the extent that the 'welfare to work' strategy is successful in reining back social benefits, and interest rates remain relatively low, then these items are unlikely to show very fast growth. This means that spending on goods and services has scope to grow quickly without upsetting the overall spending plans. We have projected this as rising by an average rate of 3.8 per cent over the next three years.

Our forecast of government receipts is based on announced tax rates, together with our expectations about the development of the tax base. This shows a picture less buoyant than that described by the government. The projections in the EFSR show the 'golden rule', that borrowing is not used to finance current spending, being met throughout the period from 1998 to 2001 and [TABULAR DATA FOR TABLE 2 OMITTED] beyond. Our projections do not show it being met in any of these years. This is despite making similar assumptions about public sector spending and investment.

Partly, this reflects our forecast of weak corporate tax receipts. These have been very strong of late, with UK tax payments of companies and financial institutions almost doubling in three years to reach [pounds]29 billion in 1997. This last year included the first instalment of the windfall tax to pay for the Welfare to Work programme, with the second and final instalment due this year. The reform to the corporate tax system with the move to a quarterly payments system and the abolition of Advance Corporation Tax (ACT) is expected to add to tax payments, although this will be offset by special transitional arrangements. But more fundamentally, the slowdown in profits growth that we are anticipating, together with an accumulation of capital allowances due to relatively strong business investment in recent years and lower rates of corporation tax are likely to reduce the tax yield. Our forecast shows corporate tax payments falling to [pounds]24 billion in 2001-2.

In addition, the general economic outlook has worsened since the EFSR was published and this is likely to lead to revisions in the government's own fiscal projections. Our forecast suggests a significant worsening in the outlook for the public finances over the coming three to four years. The EFSR forecast for public sector borrowing (including the windfall tax) in the current fiscal year was [pounds]2 billion. This would be an improvement of [pounds]3 billion on 1997-8. Our forecast is for borrowing to rise to [pounds]10 billion, a worsening of [pounds]8 billion. Figures for the first three months of this fiscal year do suggest an improvement in the outlook, but there is considerable scope for a deterioration in receipts over the following nine months. Borrowing at this level would imply that the golden rule is just missed.

Our projection is for borrowing to rise further in 1999-2000 to [pounds]18 billion, again in excess of net investment. Given the weaker outlook for the economy in 1999, an increase in borrowing of this magnitude is entirely plausible and appropriate.

Summary of the forecast

Despite the deep pessimism about the extent of the slowdown in the economy, there is as yet little firm evidence that the economy has slowed below trend. Figures for GDP in the third quarter indicate quarterly growth of 0.5 per cent. This means that even with growth of as little as 0.1 per cent in the fourth quarter, the year-on-year growth rate in 1998 will be around 2.8 per cent. This is much greater than looked possible three months ago and reflects the revisions to the National Accounts which raised the estimated growth rate throughout 1997.

Growth is expected to slow in response to a worsening of the external position as well as a continued response to earlier policy tightening. Among the components of GDP, household spending in 1999 is expected to increase by around 1 1/2 per cent, but investment is expected to fall by a little over 1 per cent. The strongest growing component of demand is likely to be general government consumption, which we are forecasting to rise by a little under 4 per cent. Both net trade and inventory accumulation are expected to make negative contributions to GDP in 1999 and the overall growth rate is expected to fall to around 1 per cent. The main area of growth is likely to be in public sector services. Manufacturing output is expected to fall by 1/2 per cent. Private sector services output is expected to grow by about 1 per cent.

With growth slowing, the recent upward trend in employment is expected to abate. This is likely to be associated with further slow growth in productivity. This reflects a shift away from manufacturing as well as the normal response of productivity to a reduction in the rate of growth. In addition, the New Deals for young unemployed people and the long term unemployed are encouraging firms to take on more workers than would usually be the case. This too is expected to have a depressing effect on productivity in the short run.

Slow productivity growth means that the rate of wage inflation consistent with the government's inflation target is relatively low. But the slowdown in growth also means that wage pressure should diminish as employment falls. However, unit labour costs are expected to rise by more than target inflation in 1999. This partly reflects the effects of wage increases in the pipeline and the introduction of the National Minimum Wage. But with demand very weak, it is unlikely that firms will be able to pass on these cost increases. Inflation is expected to remain at around the government's target, despite the anticipated loosening of monetary policy. It is expected to fall below 2 1/2 per cent on the target measure in 2000.

We expect some rise in unemployment over the next two years. ILO unemployment is expected to rise to about 2 million by the end of 1999 from its current level of a little over 1.8 million.

As usual, we have attempted to quantify the risks to this forecast, using our past forecasting performance as a measure of the possible uncertainties faced by the economy. This suggests that there is a one-in-four chance of output in 1999 being lower than that recorded for this year. This outcome could involve falls in output from the level reached in the third quarter of this year. A better outcome would be for output to end next year at the level reached in the third quarter. The chances of output being lower than this, and therefore falling for some of the time, are put at one-in-three. On the upside, there remains a one-in-four chance of output exceeding 2 per cent in 1999. The inflation outlook is benign with there being an evens chance that it will remain below 2 1/2 per cent at the end of next year.

Section II. The forecast in detail

The components of expenditure (table 3)

Growth of the UK economy has slowed steadily throughout 1998. The latest complete data, for the second quarter of 1998, shows a slowing of year-on-year growth to 3.0 per cent from 3.7 per cent in the first quarter. A further fall to 2.6 per cent in the third quarter is shown in the first estimate. This slowdown is expected to continue into 1999 with the yearly growth rate forecast to be 1.1 per cent as compared to 3.5 per cent in 1997 and 2.8 per cent in 1998.

During 1997 and the start of 1998 the growth of GDP was fuelled by fast growth of household expenditure and held in check only by the deterioration of the trade balance. Through the latter half of 1998 and into 1999 the growth rate of GDP is being pulled downwards by slow growth of household expenditure and investment and held up by the promised increases in government expenditures in the latter half of this year [TABULAR DATA FOR TABLE 3 OMITTED] and into next.

Household expenditure growth averaged some 4.3 per cent in 1997 driven by increases in disposable income and positive wealth effects. In the first two quarters of 1998 this rate of growth has slowed and is expected to continue to fall through the latter half of this year and into 1999. Growth of consumption is therefore expected to be 3.0 per cent in 1998 and 1.5 per cent in 1999. Strong growth of domestic demand (3.9 per cent in 1997 and 3.8 per cent in 1998) has outstripped output growth through this period and has led to a worsening trade balance. Export growth is expected to slow from 8.4 per cent in 1997 to about 3 1/2 per cent in 1998, import growth is set to decline more sedately from 9.5 per cent in 1997 to 6 1/2 per cent this year. This is consistent with a further fall in the contribution of net trade to GDP growth from -1/2 per cent in 1997 to -1 per cent in 1998.

The slowdown in consumer expenditure and the depreciation of sterling is expected to lead to a less negative net trade contribution in 1999. Forecast output growth of about 1 per cent is sustained by domestic demand growth of 1.2 per cent and a negative trade contribution of 0.2 per cent.

Investment grew strongly in 1997 and in the first quarter of 1998. In the first quarter, investment was 11.6 per cent higher than in the same quarter a year before. The rest of 1998 and 1999 portray a very different picture however. In the second quarter, investment dropped by 1.8 per cent. Declining confidence regarding future output is expected to see significant reductions in the amount firms are willing to invest. Investment is forecast to fall by a little more than 1 per cent in 1999.

Inventory accumulation has been strong so far in 1998, although the figures partly reflect the alignment adjustment in the National Accounts. With final demand slowing, some involuntary inventory accumulation is likely. Our forecast of a [pounds]6.4 billion increase in inventories accounts for 0.4 per cent of the growth in GDP this year. The forecast slowdown in inventory accumulation, to [pounds]4.3 billion in 1999, contributes 1/4 per cent to the slowdown in the growth rate.

Although many of the components of GDP are forecast to grow slowly through the latter half of 1998 and into 1999 the exception to this is government expenditure. The increase in government expenditure is acting as an automatic stabiliser upon the economy. Government consumption expenditure is forecast to contribute 3/4 per cent to GDP growth in 1999.

Household sector (table 4)

The pick-up in average earnings growth seen in 1997 and the first half of this year contributed to an improvement in the spending power of the household sector. Real household disposable income (RHDI) grew by 3.5 per cent in 1997, the fastest rate since 1993. Within this total, real growth in wages and salaries was over 4 per cent as real average earnings and employment grew strongly. However, the real growth in available income (intended to measure 'take home' incomes) was a more subdued 1.8 per cent in 1997, as the rise in earned income was partially offset by low growth in social benefits received and a rise in social contributions. The pattern in the first half of this year has been distorted by unusually high tax payments and bonuses in the first quarter, but real available income appears to have been growing at an underlying rate of about 3 per cent.

In the short term, household income is set to benefit from continued growth in real earnings, reflecting existing pay settlements and the introduction of the National Minimum Wage next year. There is also liable to be an improvement in net property income as mortgage interest rates fall. However, employment is expected to fall from the third quarter of this year as the economy slows down. Overall, RHDI is expected to grow by 3.8 per cent in 1999. Available income, which excludes net property income, is expected to grow by about 2 1/4 per cent.

The value of the financial assets of the household sector has clearly been reduced by the fall in equity prices. In the second quarter of 1998, households had net financial assets of [pounds]2,100 billion, enough to support about 4 years' expenditure. With their holdings of currency and deposits broadly offsetting their loans, this corresponds almost exactly to their net holdings of equity, of which about two-thirds is held indirectly through life assurance and pension funds. Assuming that the average value of these holdings is consistent with that of the All- Share index, net financial wealth is likely to have fallen to about [pounds]1,750 billion in the fourth quarter. This is roughly equivalent to the level of net holdings in the second quarter of 1997.

The estimated loss in the net value of financial assets of [pounds]350 billion is substantial, dwarfing the windfall gains of about [pounds]40 billion received in 1997. But its direct effect on spending is likely to be relatively small. This depends on the extent to which households had taken account of the earlier rise in equity prices in formulating their spending plans. Those who had budgeted on the basis of high equity values will need to cut back, but others, many of whom are unaware of the value of their holdings in pension funds, are unlikely to respond. Of equal importance to financial wealth is the value of housing wealth, estimated to be about [pounds]1.3 billion. Until 1993, housing wealth had been larger than net financial wealth, but the substantial rise in equity valuations at a time of housing market weakness has altered this position. Nevertheless, strong growth in house prices over the past year or more will have partly offset the effects of falling equity prices on household spending.

Overall, the direct effect on household spending of recent falls in equity markets is likely to be fairly small. Our best estimate is that household expenditure in 1999 will be [pounds]1 billion lower than it would have been if equity prices had continued around the levels seen in the second quarter of this year. This is equivalent to a 1/4 percentage point reduction in the growth rate.

Household expenditure is also likely to be adversely affected by the recent spate of bad news concerning the jobs market. Uncertainty about future employment prospects is thought to be an important factor in determining spending and saving patterns. With unemployment [TABULAR DATA FOR TABLE 4 OMITTED] expected to rise over the next year, confidence is expected to fall further.

Household spending grew by over 4 per cent in 1997, driven by fast growth in incomes, against a background of rising asset prices, windfall receipts and falling unemployment. Expenditure on durable goods rose by over 10 per cent. Growth in the first half of this year has slowed somewhat with spending up by 3 1/2 per cent on a year earlier. Retail sales growth in July and August of a little over 3 per cent suggests a continuing slight slowdown in household spending. We expect household spending to grow by 3 per cent this year, with slightly slower growth in the second half of the year than in the first. Growth in 1999 is expected to fall to about 1 1/2 per cent, slightly less than the growth in available income and much slower than the growth in RHDI. This partly reflects the waning of the windfall effect, but is also attributable to lower equity prices and rising unemployment.

This pattern is consistent with a rising saving ratio. The household saving ratio averaged 9 1/2 per cent in 1996 and 1997, but fell to 7.8 per cent of disposable income in the second quarter. This is now expected to rise back to around 9 1/2 per cent in 1999.

The forecast decline in household spending growth is likely to reduce growth in the demand for housing. This is expected to slow the rate of house price inflation from about 9 per cent in 1997 to about 4 1/2 per cent in 1999.

Fixed investment and stockbuilding (tables 5 and 6)

Aggregate fixed investment is estimated to have grown by 6.1 per cent in 1997. Although far from comparable, this is the fastest growth rate since the Lawson boom in 1988. In contrast, the second quarter of 1998 saw investment falling in all sectors. Business investment shrunk by 2.7 per cent from the first to the second quarter this year. Manufacturing investment fell by 2.1 per cent. Here, capital spending was reduced most significantly in the solid & nuclear fuels and oil refining industries and in the textiles, clothing, leather and footwear industries, by 21.8 and 19.1 per cent respectively. The general economic slowdown is not only affecting manufacturing. Non-manufacturing investment fell by 2.8 per cent, led mainly by the 2.9 per cent decrease in service sector investment. General government capital spending fell by 4.6 per cent, but this is less than the 6 per cent drop in the previous quarter.

Our forecast for business investment reflects worsening conditions on both the supply and demand side of the capital market. As indicated by the statement accompanying the US Federal Reserve's second interest [TABULAR DATA FOR TABLE 5 OMITTED] rate cut this autumn, turbulent financial markets and increasingly cautious lenders are expected to raise the [TABULAR DATA FOR TABLE 6 OMITTED] real cost of capital. Chart 3 shows the differential between commercial banks' prime rate and the three month interbank rate. Between August and September the differential rose by 0.4 percentage points, consistent with the view that banks are commanding a higher premium on risk. Although increases of this order are not unprecedented, it is likely that the uncertainty surrounding global financial markets will restrain lending. From the third quarter this year to next, we are expecting a 0.4 percentage point rise in the real cost of capital.

However, several factors continue to be favourable to investors. Despite the further rise in the financial deficit of non-financial corporations in the second quarter, expected to rise from [pounds]7 billion in 1997 to [pounds]22 billion in 1998, companies' balance sheets are still strong. Long-term interest rates are very low in nominal terms by recent historical standards, reflecting both low real interest rates and low inflation expectations. Furthermore, taking into account recent fallbacks in the stock market, the financial valuation of companies remains high in relation to the replacement cost of their capital.

Against this is the outlook for product demand and profitability. Recent business surveys have been very pessimistic about the prospects for further investment. With expectations of a more abrupt slowdown, reflected in our downward revision of growth for 1999, firms are likely to postpone investment plans that are not needed to meet existing demand. Additionally, pressure on profitability is not expected to ease. Corporate profits as a share of GDP are forecast to fall by around 0.4 percentage points in 1998 and by a further 1 percentage point in 1999.

Based on the assumption that firms have similar expectations about future output and costs to those in the rest of the forecast, we expect business investment to rise by roughly 6 1/2 per cent in 1998 and to fall by 2 1/4 per cent in 1999, including further reductions in capital spending by business throughout the rest of this year and into the next. Following rapid growth in private sector housing starts in 1997, we also expect growth in private sector housing investment to slow to less than 3 per cent in 1998. Conversely, with government investment forecast to grow by roughly 3 1/2 per cent in 1999, the recent deterioration in public capital spending is expected to turn.

Stockbuilding was up by [pounds]1.6 billion in the second quarter this year, reflecting increases in inventories in most sectors. Stockbuilding of a little less than [pounds]6 1/2 billion is expected in 1998. Stockbuilding of this order would add 0.4 per cent to this year' growth rate. The high value of inventory accumulation so far this year would appear to suggest that output has been running ahead of demand and is a further indicator of slowing growth.

Balance of payments (tables 7 and 8)

The short-term outlook for the UK economy depends crucially on the ability of domestic producers to increase their sales in an increasingly difficult world market. Prospects for growth in the world economy have deteriorated sharply in a short space of time. World trade growth, which reached almost 10 per cent in 1997, is expected to decline to a little over 6 per cent this year and around 5 per cent in 1999. This will increase competition in both overseas markets and at home.

The tough market environment will be exacerbated by the effects of an overvalued exchange rate. The recent fall in the exchange rate to DM 2.80 has eased the situation, but we estimate that at this level the overvaluation against the DM is about 12 per cent. The effective exchange rate is likely to average about 102.4 in 1998 against 86.3 in 1996, the year in which it suddenly appreciated. The expected decline in exports from such a deterioration in competitiveness failed to materialise [TABULAR DATA FOR TABLE 7 OMITTED] in 1997 but is becoming evident in the data in 1998. Exports of manufactured goods grew by some 9.5 per cent in 1997, the same figure for 1996, but have grown year on year by 3.2 per cent and 2.6 per cent in the first two quarters of 1998 respectively and not at all in the third quarter. Growth of less than 2 per cent in the year as a whole is now likely. This is much less than the growth in world trade, indicating that UK exporters are losing some of their market share.

The forecast for exports beyond 1998 shows some improvement in the fortunes of UK exporters as they broadly maintain their market share. The expected depreciation of sterling and the effective exchange rate, to around DM 2.75 and 97.3 by the end of the year, contributes to forecast export growth of around 4.8 per cent in 1999; an improvement on this year, but much lower than witnessed between 1994 and 1997.

The UK terms of trade have now returned to the levels of 1992 and 1993 due to reductions in import prices resulting from the appreciating exchange rate in the early part of 1998. UK export prices also fell in both 1997 and 1998 but by not enough to offset the effects of the higher exchange rate. Given that UK producers are facing increases in domestic costs, such price reductions are paid for out of profits in order to maintain price competitiveness on world markets. The ability of exporters to absorb such a deterioration in competitiveness is likely to exist only over a reasonably short time horizon and helps to explain their vocal opposition to the interest rate policy of the MPC. The terms of trade are forecast to fall in 1999 as sterling begins to depreciate. Export prices are forecast to stabilise in 1999 in sterling terms whereas import prices are forecast to rise. A similar pattern to the terms of trade is expected for [TABULAR DATA FOR TABLE 8 OMITTED] the index of UK export price competitiveness. Price competitiveness has improved in 1998, but with competitiveness still 10 per cent worse than in 1995, further improvements are expected as sterling falls.

Subdued growth in export volumes through 1998 can be contrasted with the steady increase in import volumes over the same period. Imports are forecast to have grown by 5.9 per cent (6.2 per cent for manufactured goods) in 1998 to [pounds]208.5 billion in 1995 prices. However, this does represent a slight fall from the growth rate of 9.1 per cent (10.4 per cent for manufactured goods) witnessed in 1997 and can be explained by the fall in UK consumption growth relative to 1997. Reductions in the growth of demand combined with the depreciation of sterling are forecast to reduce the growth rate of imports through 1999 to 2.6 per cent (3 per cent for manufactured goods) and beyond.

The trade deficit in 1998 is expected to worsen to around [pounds]18 billion compared with [pounds]13 billion in 1997. The deficit is expected to continue to worsen throughout 1999 rising to [pounds]5.5 billion by the fourth quarter, a forecast deficit for the year of a little over [pounds]20 billion. A faster rate of export growth relative to import growth is expected to reduce the rate of increase of the deficit on goods beyond this date.

The services, transfers and income balance is, like that of trade, expected to worsen through the rest of 1998. The balance is forecast to fall to [pounds]16 billion in 1998 compared with [pounds]21 billion in 1997. Both of these figures however represent upward revisions compared with the old calculations of trade in services by the ONS, the 1997 surplus having increased by some [pounds]3.5 billion. The main reason for these substantial upward revisions is the broadening of the number of services now measured by the ONS and an increase in the number of firms surveyed to give better coverage. In 1998 the reduction in the service trade balance can be attributed to the faster growth of imports (11.2 per cent) relative to exports (8.2 per cent). These relative growth rates are forecast to persist into 1999 and the services balance to fall to [pounds]18 billion.

The new methods for calculating trade data have also led to a substantial upwards revision of the current balance surplus in 1997, from [pounds]4.5 billion under the old system to [pounds]8.0 billion under the new. Our forecasts for the 1998 and 1999 deficits have fallen to [pounds]2.2 billion and [pounds]3.0 billion respectively, relatively small deficits, but a substantial turnaround from the 1997 position.

Output and employment (tables 9 and 10)

The quarterly rate of growth of the UK economy has slowed from 1.2 per cent in the second quarter of 1997 to 0.5 per cent in the third quarter of this year. The decline in growth has not been shared equally among the different industrial sectors. Instead, the traded sector has struggled under the weight of an over-valued exchange rate, while the private service sector has responded positively to the growth in household spending. Although, with recent revisions of the national accounts, manufacturing no longer appears to have been in recession since the end of 1997, growth in [TABULAR DATA FOR TABLE 9 OMITTED] manufacturing output has still been exceptionally low and is expected to grow by around 1/4 per cent this year.

It now seems that the difficult trading conditions apparent in the traded sector are about to spread throughout the private sector economy. Output in the distribution sector fell by 0.2 per cent in the second quarter. With sluggish domestic demand expected next year, growth in distribution and business services output is expected to slow to around 1 per cent in 1999 and manufacturing output is expected to decline by 1/2 per cent, despite the expected depreciation of sterling. The public sector is expected to grow fastest in 1999.

Productivity growth in the last two years has been faster than previously perceived. This is due to the upward revision of GDP growth in the national accounts following changes such as the reclassification of software expenditure as capital investment rather than an intermediate input. Nevertheless, productivity growth has been slow. In the year to the second quarter productivity growth was 1.7 per cent, down from 1.8 per cent in the year to the first quarter.

The low growth in productivity is partly a consequence of the shift away from manufacturing to the service industries, although manufacturing productivity has also fallen as output has slowed down. It is also likely to reflect the fact that labour remains relatively cheap. Real wages have grown very slowly in the last recovery as the market mechanism has operated to price willing workers into jobs. This has encouraged firms to take on workers, possibly substituting labour for capital in the process, thereby reducing productivity growth.

Productivity growth is expected to continue at a slow pace throughout this year and the next. This reflects the typical pro-cyclical response to a slowdown in economic activity. It is also due to the expected effect of the government's New Deal programme for young [TABULAR DATA FOR TABLE 10 OMITTED] unemployed people and the long-term unemployed. Part of the mechanism by which these schemes are expected to work is by encouraging companies to take on additional workers at subsidised wages. In the short run at least, this is likely to reduce average productivity.

Employment fell by 124,000 between March and June (measured by the number of workforce jobs). This is the first fall in employment since the beginning of 1993. The reduction was primarily due to the drop in self-employment, which had been falling steadily throughout 1997 and a large fall in the numbers of government supported trainees. Despite this, the labour market continued to tighten in the second quarter of this year as both ILO unemployment and the claimant count fell further. From the first to the second quarter, ILO unemployment was reduced by 62,000 and the claimant count by 20,000. However, the fall in employment in the second quarter and the latest unemployment statistics, indicate that this trend may now be reversing as the economy slows down. Although the claimant count continues to fall, ILO unemployment was up by 9,000 in the quarter to August.

The economic slowdown is now expected to lead to further increases in unemployment this year and next. ILO and claimant unemployment are expected to rise by around 140,000 in 1999. This represents a significant increase. The increase in unemployment is expected to be concentrated among older workers. One of the effects of the new deal for young people is to reduce their share in overall unemployment.

We have assumed that there is an ex ante increase in whole economy average earnings of 1 per cent from the second quarter of 1999 as a consequence of the NMW. Our model suggests that an increase in pay pressure of this amount is typically associated with a long-run fall in employment of about 1 per cent, amounting to about 250, 000 jobs at today's population. The effect on the claimant count would be much smaller given the likelihood that many of those affected would not be eligible for unemployment related benefits.

Earnings and prices (tables 4 and 11)

The behaviour of average earnings has been a key factor influencing the analysis and decisions of the Monetary Policy Committee (MPC). In June, when interest rates were raised, the headline rate for March was estimated at 5.2 per cent, up from 4.6 per cent in January. The March figure has subsequently been revised down to 3.9 per cent, a fall from the revised 4.2 per cent in January. Given the MPC's view that wage inflation of 4 1/2 per cent is consistent with the inflation target, it is unlikely that interest rates would have been raised in June if they had access to the current rather than the first vintage of data.

On the latest figures, average earnings growth has remained strong. Growth in average earnings for the entire economy was 4.6 per cent in July. This figure is down from its June peak of 5.2 per cent and closer to the 4.5 per cent that is consistent with the government's inflation target. Performance bonuses are still believed to be biasing these figures upwards although the majority of this effect has now passed through the system. This can be seen in the manufacturing and service industries' average earnings indices which rose by 5.0 per cent and 4.5 per cent in July respectively as compared with 4.6 per cent for pay in the public sector. Also, the latest figures show that pay settlements have remained stable over the six months to August, with median pay awards at 3.5 per cent. Growth in whole economy unit labour costs remains strong at 3.7 per cent for the second quarter of this year.

As mentioned above, the expected weakening of aggregate demand (both domestic and foreign) is expected to reduce the inflationary pressures emanating from the labour market. However the average earnings index is still expected to grow strongly next year because of the lagged effects of existing pay pressure and the introduction of the national minimum wage (NMW). Earnings are forecast to grow at around 4.5 per cent for the remainder of this year and into 1999 but then climb to nearly 5 per cent in the second quarter when the minimum wage legislation becomes effective. By the end of 1999, average earnings growth is forecast to have slowed to 4.7 per cent but will still be above what might normally have been forecast for this stage of the cycle.

The predicted effect of the NMW on future earnings growth is dependent on assumptions regarding the timing of wage rises to meet the legislation. The current forecast is based on the assumption that average earnings will rise by 1 per cent by the second quarter of 1999 and that half the adjustment will have taken place before legislation becomes effective.

The most comprehensive measure of cost pressures within the economy is the GDP deflator. In the second quarter this grew by 1.9 per cent on the year and by 0.6 per cent on the previous quarter. This is a slight increase from the first quarter, but still relatively low compared with recent history. Low inflation in costs can be attributed to the strength of the pound keeping the cost of imported raw materials low and encouraging lower profit margins. Manufacturing input prices have again fallen in the previous two quarters (by around 5 per cent) and are expected to continue to fall throughout the rest of this year and into 1999, albeit at a slower rate. In turn this has helped inflation in the wholesale price of manufactured goods to remain low with this measure of inflation rising by only 1 per cent for the last two quarters.

The forecast weakening of the pound relative to [TABULAR DATA FOR TABLE 11 OMITTED] other major European countries (leading to higher import prices) is expected to reverse the trend of the last few years and costs should begin to rise. Indeed within the forecast as a whole import prices are expected to provide one of the few sources of upward pressures on inflation over the short-term forecast horizon. Correspondingly the value of sterling on the foreign exchange, or rather the speed of its adjustment, may become the focus for the 'hawks' in the MPC over the next year. However it is unlikely that the effect of rising import prices will cause the government to miss its inflation target. Import prices did not move downwards by as much as the movement in the exchange rate over the last few years, probably because foreign export firms took the opportunity to capture additional profits. This may be partly explained by a belief that the appreciation of sterling was unlikely to persist far into the future. Correspondingly some slackening of import margins can be expected when sterling begins to depreciate on the foreign exchange markets.

Retail price inflation has fallen from its temporary increase in the second quarter and averages 3.4 per cent (down from 4.0 per cent) for the third quarter. The government's target measure for inflation (RPIX) remains above its target rate of 2.5 per cent for the third quarter at 2.7 per cent but, like RPI, has slowed since the second quarter. In comparison RPIY growth has remained unchanged since the second quarter at 2.2 per cent. This difference reflects the effect of indirect tax changes in RPIX and RPI.

It is interesting to compare expected paths for these three measures of retail price inflation for the next few periods in order to discern inflationary pressures within the economy. RPI inflation is expected to fall continually throughout 1999 whereas RPIX and RPIY inflation are expected to rise slightly and then fall, with RPIX [TABULAR DATA FOR TABLE 12 OMITTED] reaching its target level by the fourth quarter of 1999. Reductions in the interest rate over the next year explain part of the fall in RH growth relative to RPIX growth while the temporary increases in both RPIX and RPIY in the second quarter of 1999 reflect higher import prices.

Inflationary pressures within the rest of the economy are expected to remain weak throughout the rest of 1998 and 1999 and the government is predicted to meet its target inflation rate for RPIX of 2.5 per cent by the end of 1999. Beyond this date deflationary pressures are expected to emerge, with expected RPIX inflation of 2 per cent in 2000.

National and sectoral saving (table 12)

Table 12 shows our forecasts of national and sectoral saving. This puts together parts of the forecast that have already been discussed. This shows that saving and investment in the UK are now broadly balanced: domestic investment is financed by domestic saving.

The current account is expected to worsen slightly this year and then remain in deficit for the next two years, but the level of the deficit is forecast to be less than 1/2 per cent of GDP throughout. Within the domestic economy, the personal sector is in its customary position of saving more than it invests and lending the surplus to other sectors.

Investment by the company sector is expected to continue at about 9 1/2 per cent of GDP, larger than it is able to fund from retained profits alone. As a consequence the company sector is expected to draw in funds from other sectors to finance its investment plans. The deficit is expected to be at its peak this year.

The public sector is in balance at present. But with investment expected to rise and saving to fall, it too is likely to need to borrow from other sectors to finance its need for capital.

The economy in the medium term (table 13)

Over the medium term, the economy needs to adjust to an over-valued exchange rate. On current trends, there is a strong possibility that the exchange rate against the DM will be fixed from the beginning of 2001 at DM 2.70. Since the overvaluation will not be corrected by exchange rate depreciation, it will be necessary for prices to fall relative to those in Europe and the rest of the world. Against the background of low inflation across the world, this is likely to mean a period of very slow domestic inflation.

The anticipated fall in the exchange rate and rise in competitiveness is expected to lead to a change in the influence of net trade on the UK economy from 2000. By then net trade is expected to be making a significant positive contribution to output growth. This is expected to be accompanied by a relative improvement in the position of manufacturing industry.

Among the other components of demand, government consumption is likely to be rising quickly from 1999 until at least 2002, in line with the government's spending plans. Consumer spending is expected to grow relatively slowly over the medium term as a consequence of slow growth in personal sector income. This is due partly to a deterioration in the terms of trade. Productivity growth is expected to follow the cycle, being relatively low until next year and then picking up speed in the early part of the next decade. This also reflects a shift of resources towards manufacturing which tends to enjoy a higher level of productivity than most other sectors.

Our medium term projection shows output growth recovering to about 2 1/2 per cent per annum after slow growth in 1998 and 1999. ILO unemployment settles over these years at around 2 million. This is probably a little higher than the economy's sustainable rate and is partly responsible for the downward pressure on wages and prices shown.

Public borrowing settles at around 2 per cent of GDP. The current account moves back into surplus in 2001.

[TABULAR DATA FOR TABLE 13 OMITTED]

[TABULAR DATA FOR TABLE 14 OMITTED]

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 over the years in the October Review. The latest National Accounts information available when these forecasts are constructed is for the second quarter of the 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 be 70 per cent sure that GDP growth in 1998 will be between 2.3 per cent and 3.3 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 endogenous offsetting movements among the components.

Table 15. Probability distribution of growth and inflation forecasts

Inflation: probability of 12 month RPIX inflation falling in the
following ranges

 1998Q4 1999Q4

less than 1.5 per cent 16 29
1.5 to 2.5 per cent 50 21
2.5 to 3.5 per cent 31 21
more than 3.5 per cent 3 29

 100 100

Growth: probability of annual growth rate falling in the
following ranges

 1998 1999

less than 0 per cent 24
0 to 1 per cent 26
1 to 2 per cent 10 26
2 to 3 per cent 53 16
3 to 4 per cent 34 6
more than 4 per cent 3 2

 100 100

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 have been calculated from 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.

We have typically calculated the probability of recession in the year ahead by evaluating the probability that output at the end of the year is no higher than at the end of the previous year. This suggests that there is a substantial risk, of close to 33 per cent, that GDP will actually fall in 1999.

Box A: Long-run monetary policy and exchange rate assumptions

The objective of monetary policy is to achieve underlying inflation, measured by the RPI excluding mortgage interest payments (RPIX) of 2 1/2 per cent at all times. Since monetary policy is believed to act with a lag, interest rate decisions are based on the prospects for inflation up to two years ahead. If recorded inflation turns out to be a percentage point higher or lower than the target, then the Governor of the Bank of England is expected to write an open letter to the Chancellor explaining why this occurred and what is to be done about it. Evidence from the National Institute model suggests that, with the best will in the world, it is only possible to keep inflation within such a narrow range for 50 per cent of the time.

Our assessment of inflationary conditions suggests that the MPC will continue to reduce interest rates in 1999, eventually converging on German interest rates in 2001. Our forecast for the effective exchange rate is generated from the assumption that the UK will join EMU in the early part of the next decade. Of key importance here is the level of the exchange rate at which EMU entry might take place. Experience in the ERM demonstrated conclusively the importance of choosing the right rate. In our view an appropriate rate for sterling is about 2.50 DM to the pound, or lower. However, we no longer think that sterling will be as low as this by the time the government is otherwise ready to fix the exchange rate. As a consequence, it is expected that sterling will join EMU at a rate of DM 2.70. In order to join EMU at this rate in 2003, it will be necessary for the exchange rate to trade at around this level from early in 2001. As the Maastricht Treaty implied, it would be unwise to consider entering a monetary union without a reasonable prior period of exchange rate stability. This allows the prospective partners to judge whether their economies are sufficiently convergent for there to be a good chance that the union will be a success.

Box B. Changes to the National Accounts

This is the first forecast that we have made on the basis of the new set of national accounts, brought in with this year's Blue Book. The national accounts are now presented on the European System of Accounts 1995 (ESA 95). This represents the most fundamental change since the introduction of national accounts in the 1940s. Changes have been made to the definition of sectors, the production boundary and to the definition of investment, including computer software and milking cows for the first time. The old 'factor cost' concept is no longer an important concept in the new accounts. Attention is now focused on GDP at market prices (the headline measure of GDP) or gross value added at basic prices (similar to GDP at factor cost). In addition, the accounts have been rebased to 1995 prices, use the Inter-Departmental Business Register (IDBR), and measure public sector output in ways which no longer rely solely on the measurement of inputs.

The changes have caused considerable disruption to users of the accounts, including the re-specification and re-estimation of models, once the relevant data series have been identified and collected. The inevitable errors introduced in this process will have made forecasts produced using these data more uncertain than usual at the moment.

Ultimately, the new accounts present a coherent picture of the UK economy, but one not particularly different from that measured on the old basis. One of the most significant changes is in the saving ratio, now measured for households rather than the former personal sector. Its level in 1997 is now put at 9.5 per cent rather than 11 per cent. GDP growth has also been revised up from 1991, with growth 0.2 per cent per annum faster than previously thought.

NOTE

(1) See Young, G. (1996), The Influence of Financial Intermediaries on the Behaviour of the UK Economy, NIESR Occasional Paper No.50.

* The forecast was compiled using the latest version of the National Institute Domestic Econometric Model. We are grateful to Nigel Pain and Martin Weale for comment and discussion.