Chapter I. The home economy.

Anderton, Bob ; Britton, Andrew ; Gregg, Paul 等

The forecasts were prepared by Bob Anderton, Andrew Britton, and Paul Gregg, but they draw on the work of the whole team engaged in macroeconomic analysis and model building at the Institute. Parts One and Two of the chapter were written by Andrew Britton, Part Three by Bob Anderton.

PART ONE. RECENT DEVELOPMENTS AND SUMMARY OF THE MAIN FORECAST

Since last November we have based our forecasts on the assumption that the UK will soon become a full member of the European Monetary System. There has been no change in the official position on entry to confirm or correct that assumption, but recent hints and rumours make clear that it is the right assumption to make. We retain it for this forecast. The actual date of entry is now more likely to be in the first half of next year; a fall in the rate of inflation is seen as a precondition, and it is one which may not now be fulfilled as soon as we had hoped. We assume that some announcement on the ERM is made about the end of the year, permitting an interest-rate cut in the first quarter, followed by formal accession by the second quarter, about the time of the Budget.

The attitude of the British government to economic and monetary union in Europe has been one of disbelief and hostility. It seems increasingly likely however that agreement will be reached to form such a union, with or without British membership. We assume that the process is complete by 1997; we also assume that the UK joins, even though this is not current policy. For the present forecast this affects mainly the last few years of the decade, and makes very little difference to the prospects for this year or next. (A commitment to EMU would add credibility to membership of the ERM, and might hasten the reduction in interest rates.)

As expected, the Budget this year changed little; it was presented along with a sombre account of economic prospects in the FSBR. Meanwhile local authority expenditure has been increasing faster than expected or intended, financed by the high levels at which the new poll tax has been introduced. We assume that control of spending will be restored, and that next year the rise in the poll tax will be no more than in line with inflation.

Since the Budget the tone of Treasury comments on the economy has become even more sombre (as it often does in the approach to the annual review of public spending). It has been suggested that there will be no scope for tax cuts before the next general election. That would be in line with our analysis and our assumptions for this forecast. Indeed we assume no scope for tax cuts even after the next general election (see below).

Our short-term forecast is not as optimistic as it was in February. On both inflation and the balance of payments we have made significant changes. For output the forecast is not much altered.

The more rapid rate of inflation now in prospect reflects mainly the high level of the poll tax and some smaller increases in charges announced about the same time. These should not be included in an assessment of the underlying rate of inflation to which the rise in the deflator for GDP at factor cost might be a better guide. But they will be quoted in wage claims and may for that reason raise the underlying rate later in the year. For the end of this year we are forecasting a 12-month rise in the RPI of 8.8 per cent assuming no cut in interest rates. By the end of 1991 the 'headline figure' could be below 5 per cent.

The run of disappointing trade figures in the first quarter of the year and revisions to the data on the invisible account revived fears of a continuing large deficit on the balance of payments. (Estimates of the stock of UK net assets overseas have also been revised down substantially.) The high level of imports is difficult to reconcile with the evidence that manufacturing output is falling and that stocks are being run down. The view that the trade figures were `erratic' must be accepted for the present, and the outlook is for a sustained improvement for the rest of the year. Even so the current account deficit for the year is now forecast at about 9-1 5 billion. Next year it should be reduced to below 10 billion [pounds].

The level of output probably fell in the first quarter of this year. it is likely that stocks are being run down and fixed investment cut back in response to the continuation of high levels of interest rates. Consumer spending on the other hand seems to show considerable resilience. For the rest of the year output growth should be resumed, but at a slow pace. The rise in GDP year-on-year is forecast at 1.2 per cent, or 0.8 excluding North Sea Oil. This is close to the forecast published by the Treasury in the FSBR.

Our forecast for growth in 1991 has been revised up a little to 2.7 per cent, a higher figure than would be suggested by the Treasury's forecast for the first half year. The main reasons for the relatively strong growth are the reduction in UK interest rates and the buoyancy of world demand.

Our medium- and long-term forecasts reflect our view of the consequence of full ERM membership and the transition to EMU. Initially we assume that the effective exchange-rate index will fall by about 2 per cent a year; this is consistent either with wider margins for Sterling within the ERM, or occasional realignments. By 1997 we assume exchange rates in Europe are fixed; our exchange rate would then appreciate a little as the European currencies (or currency) strengthen against the dollar.

Consistent with that view of exchange rates, we assume a gradual fall in short-term interest rates in the UK, to 5 per cent by the end of the decade. Although underlying inflation also falls, this is mainly a reduction in real interest rates, and should increase investment, adding to the sustainable rate of growth of output. Long-term bond yields are expected to fall consistently and the price of equities to rise.

A striking feature of our medium- and long-term forecasts is the low rate of inflation after 1992. The prices of imports are given by world inflation (see chapter 2) and the path of the exchange rate. Given that path for the exchange rate (and the pressure of demand) the reduction in interest rates actually helps to slow down inflation by reducing the costs of holding stocks. Moreover the very low figures for the increase in the RPI produced by cuts in mortgage interest rates may encourage low wage settlements.

PART TWO. FINANCIAL SURPLUSES AND DEFICITS

Despite the high level of nominal interest rates the personal sector was in large financial deficit in 1988 and 1989. This was associated with a large deficit on the current account of the balance of payments (a financial surplus for the overseas sector) and with a surplus for the public sector. (We assume that this picture is broadly correct, with the residual error mainly attributable to the measurement of the company sector balance.) This is a combination of financial balances without precedent, and doubts have been expressed about its sustainability. Behind the medium-term financial strategy, published by the government at the time of the Budget, there probably lies a calculation that the personal sector will move back into its accustomed surplus, allowing the current account of the balance of payments to improve, whilst at the same time the public sector moves back into balance within a few years. This could be described as a move back towards equilibrium; we compare that pattern with our own forecasts.

As table B shows we expect a much reduced deficit for the personal sector this year, followed by a small surplus in 1991. This is matched by an improvement on the current account of the balance of payments. The surprising feature of the table is the behaviour of the financial balances after 1991. The personal sector moves again into sustained deficit, the current account deficit widens and the public sector is left to repay debt on an even larger scale.

The reopening of the personal sector deficit is a consequence of the fall in interest rates, assumed as part of the transition to economic and monetary union. Had interest rates been held at their current levels the personal sector would have been in increasing surplus. The major stimulus to personal sector spending when interest rates fall is to investment in housing; both the price and the volume of new building rise substantially. Added to that is a powerful effect from the growth of consumer credit, linked to extra demand for durables. These are further reinforced by the effect of lower interest rates on the prices of gilts and equities, which provide capital gains for the personal sector adding to its net financial wealth.

Even in a forecast to the end of the decade the full effects of lower interest rates do not have time to show themselves. In the very long term the stock of credit would be fully adjusted to the lower level of interest rates following monetary union, and the flow of new borrowing would be no more than that needed to keep the debt-to-income ratio (or debt-to-wealth ratio) constant. Similarly after a very long period the stock of personal sector net wealth would adjust to a new equilibrium relative to income. At that equilibrium the ratio of net wealth to income would be maintained by a combination of capital gains (on equities and indexed gilts) and the financial surplus of the sector. The size of the surplus would reflect the allocation of the sector's portfolio as well as its total net wealth. Perhaps such ideas of stock equilibrium are behind the suggestion sometimes made that a personal sector financial deficit is necessarily self-correcting. Whilst the principle seems to be correct, the time-scale over which it can be applied may be very long indeed.

In our consumption function the effect of accumulated wealth on current spending, although statistically significant, is relatively small. Simulations on the model indicate that convergence on the equilibrium wealth-to-income ratio can take decades rather than years. On the other hand we recognise that our model may not capture all aspects of consumer behaviour. Further work is needed to investigate the effects of expectations on spending. We have in fact changed the profile of consumer spending in this forecast to a certain extent to allow for the possibility that the reductions in interest rates associated with EMU are correctly anticipated.

It is perhaps misleading to pay too much attention to the net acquisition of financial assets as a measure of `balance' or equilibrium. its definition is a matter of accounting convenience alone, especially so far as the treatment of capital gains is concerned. The more fundamental issues concern the stocks of assets and their relationships with current or expected income flows. In the present context however the path of these financial balances does seem to point to a real change of behaviour in the 1990s compared with previous periods. Behind the concern with the financial balance of the personal sector is the accounting relationship between sectors involving the current account of the balance of payments and public sector borrowing. These balances are also defined in a way that is open to question, but they remain of concern to policymakers nevertheless.

Variant forecasts

One way of examining the consequences of the renewed deficit for the personal sector in the forecast period is to look at a variant in which savings remain relatively high. In the variant a negative residual is added to the consumption function in the model. As a result consumer spending is lower throughout, the amount of the reduction building up initially but levelling off at 2 per cent after about three years. As table B shows the personal sector now stays close to balance.

In this variant the exchange rate and the rate of interest were both held fixed. The level of output is lower, initially because of weaker demand, but ultimately because the capacity of the economy is reduced by a lower level of fixed investment. In the first few years of the variant forecast, economic activity and the price level are both depressed.

The consequences for the other sectors of the economy are also shown. Taking the year 1993 as an example, the improvement of 8 billion [pounds] in the financial balance of the personal sector, compared with the main case, is matched by reductions of 4 billion [pounds] each in the surpluses of the public and the overseas sectors. The deficit of the company sector is unchanged. The current account of the balance of payments is still in deficit, but heading more rapidly towards balance than it does in the main case. Perhaps the assumption we make about fiscal policy in the main case is tighter than is strictly necessary for the variant with more savings. To make the outcome on the current account comparable a second variant was computed in which tax cuts were made (in addition to the residual adjustment to the consumption function) so as to restore the balance of payments deficits to broadly the path of the base run. The results are shown in the third section of table B.

If we focus again on 1993, only two columns are different between the main case and the second variant. The changes to the personal sector and the public sector match one another. The personal sector is in small surplus, and the public sector is also in surplus-but to a much reduced extent as compared with the base run.

We have not considered a variant in which extra saving results in a lower rate of interest. This might be appropriate under the present exchange-rate regime, if the prospect for a better outcome on the balance of payments caused the exchange rate to appreciate and the rate of inflation to be lower as a consequence. It is not clear, however, how the current exchange rate would react to the prospect of a better outcome for the current account which would be evident only after the UK had joined a monetary union.

In the absence of an exchange-rate rise, the effect of higher savings on inflation is small. That being so, it is not clear how interest rates would now react since the rate of inflation seems to be the main criterion used by the monetary authorities in deciding when interest rates can come down. Once the UK is a full member of the exchange-rate mechanism interest rates are unlikely to respond much to any signals other than those that come from the foreign exchange market. If we treat interest rates as fixed there is no economic mechanism by which a higher level of savings is translated into a higher level of fixed investment. The increment to net wealth, when it occurs, is all in the form of net overseas assets.

From the point of view of the government in power in the mid-1990s the low rate of personal savings we forecast presents a policy problem. Fiscal tightening on the scale indicated, whether it took the form of tax increases or public spending cuts, would be unpopular and might be damaging to the economy. it would therefore be attractive to find a way of increasing personal sector savings. Some tax concessions were announced in the Budget intended to help savers, but their effect on the personal sector as a whole is unlikely to be large. The government has argued that direct controls on personal sector borrowing would not be effective. We conclude that low personal sector savings, and a tight fiscal policy, are the best assumptions to take for the present as the main case of our forecasts.

PART THREE. THE MAIN FORECAST IN DETAIL

Forecasts of expenditure and output (table 1) The revised figures for GDP show an excess of the output measure over the expenditure measure for last year. One reason for this could be that output may actually be overestimated, particularly manufacturing output; our reasons for believing this may be the case are given in Box 1 on page 19.

Compared to our last Review we are forecasting a marginally smaller growth in GDP this year at 1.2 per cent. Substantial de-stocking accounts for much of the slower growth in 1990. We are also less optimistic about the recovery in oil output. The net contribution to output from the balance of payments is less in this forecast which partly reflects the poor trade figures of the first quarter of this year. In contrast, we have higher public sector output growth for 1990 due to the higher levels of public expenditure by local authorities that have already occurred. Our forecast of consumers' expenditure growth is virtually unchanged at just over 1.5 per cent.

Next year we are expecting a strong recovery in output with exports continuing their good performance and consumers' expenditure also showing renewed growth, especially late in 1991 as our assumed interest-rate falls begin to take effect. Oil output should also be stronger by this time and fixed investment growth could be around 2 per cent. This will bring total GDP growth back towards its potential trend rate of approximately 2% per cent.

Wage and price inflation (tables 2, 9 and 10) Our assumption of UK ERM entry (in the second quarter of next year) implies that UK interest rates will eventually converge towards European average levels. We assume that before ERM entry Sterling depreciates-in line with the open arbitrage condition which is equivalent to the interest-rate differential between the UK and overseas less our assumed risk premium for the UK (see table 10). After ERM entry UK interest rates gradually move towards European levels, which is consistent with some continuation of the depreciation of Sterling against the D-Mark, as well as a remaining, now smaller, UK risk premium vis a vis Germany. This also implies a depreciation of the UK effective exchange rate. UK membership of ERM would therefore deliver a similar future exchange-rate path to that of Italy. Once European monetary union takes place (circa 1997) the UK exchange rate is irrevocably fixed against the other European participating states. However, this will probably mean a small appreciation of Sterling against the US Dollar due to the strength of the D-Mark.

As the UK effective exchange rate has already depreciated substantially over the past year this means that there will be mounting pressure for considerable import price rises this year. However, the short-term effects of exchange-rate movements in our manufacturing import price equation are not as large as the effects arising from world price increases in foreign currencies. This implies that importers may initially cut profit margins (in order to maintain their market share) before implementing the full price rise if the exchange-rate fall persists. In addition, weak world basic materials prices may further help limit import price growth to around 4.5 per cent this year. Our forecast of future exchange-rate falls next year will probably keep import price inflation at approximately the same rate in 1991.

Producers will find these import costs rising at the same time that unit labour costs are accelerating due to the combined effect of increased earnings and falling productivity growth. Profit margins must therefore fall if underlying price inflation is to stop rising-this may be happening to a certain extent but the recent upward trend in producer output prices indicates some producers are reluctant to lower profit margins.

As profit margins become tighter, employers may be more reluctant to grant large earnings increases. However, earnings will be affected by the following factors: first, in previous years average earnings were somewhat restrained by income tax cuts but this will not happen this year and looks increasingly unlikely next year (although the reduction in national insurance contributions effective from last October will have some effect this year). Tax cuts would also increase the sustainable level of utilisation (i.e. decrease the NAIRU) thus decreasing the underlying rate of inflation for any given level of actual utilisation. Second, previous housing price booms have been followed by rapidly rising earnings. Third, earnings may now be more strongly linked to interest-rate increases as the personal sector interest-linked debt has substantially increased in recent years. Fourth, wage settlements have been high relative to increases in average earnings over the past year or so and the discrepancy has often been explained in terms of a reduction in wage drift' as the economy slows down (i.e. overtime hours and payments decelerate). However, latest data indicate that over-time and total hours worked in manufacturing have stopped decreasing in the early part of this year. The implication is that more of the recent high wage settlements will be transmitted into future average earnings growth figures compared to the situation where hours continue to fall. Most of the above points suggest upward pressure on earnings growth and this partly explains why we are forecasting no significant fall in earnings growth both this year and next. In the longer run there should be some downward pressure on wage inflation from the extra discipline exerted by UK ERM membership. However in the short term our forecast of underlying inflation, as represented by the deflator for GDP at factor cost, is therefore higher compared to our last forecast.

The RPI grew by just under 9 1/2 per cent during the year to this April. This partly reflected the excess cost of the poll tax relative to domestic rates and also the uprating of excise duties in the last budget. Further utility price rises should bring retail price inflation very close to 10 per cent in the coming months, which has important implications for average earnings. We believe the RPI will be increasing by around 9 per cent per annum in the fourth quarter of this year but our assumed interest-rate falls of next year will bring it down to approximately 5 per cent in the fourth quarter of 1991.

The consumer price index will be subject to a discontinuity from the second quarter of this year because of the treatment of the poll tax in the National Accounts. Domestic rates were previously classified as a tax on housing services and part of nominal consumers' expenditure whereas the poll tax is treated as a deduction from income. As the volume of housing services consumed will be the same this implies that the volume of consumers' expenditure does not change. Consequently, the implied consumers' expenditure deflator will fall but this obviously does not represent any change in underlying inflation. The tables in the Review now reflect these changes.

Personal income and expenditure (table 2)

Although the data are volatile it does seem that a slowdown in the growth of both consumer credit and retail sales is well under way. This obviously largely reflects high interest rate levels as personal sector debt payments to income ratios have virtually doubled since early 1988. But consumer uncertainty concerning future real disposable income, house prices and lower expectations of future tax cuts may also begin to discourage consumers' expenditure. Our short-term forecast for consumers' expenditure is similar to the February issue of this Review. We expect consumption growth to fall to around 1 1/2 per cent this year with a recovery to around 2 per cent next year. Given our forecast for the growth of personal disposable income this implies a slight rise in the savings ratio. We have not made any allowances for a pre-election reduction in income tax as fiscal restraint is required to offset the effects on demand of falling interest rates given our assumption of ERM entry. Much depends on the growth of earnings and inflation (and hence real incomes) and the expectations of consumers. Our consumption equation is not forward looking, but given the life cycle theory that consumers will maximise utility by smoothing their consumption path over their lifetime their expectations of future income and interest rates will affect current consumption. Consequently, our residual adjustments on consumers' expenditure are partly designed to capture this more smoothed behaviour of consumption.

Fixed investment and stock building (tables 3 and 4) Virtually all of the variables which encouraged strong investment growth in the late 1980s seem to be heading for a downturn this year; capacity utilisation will fall as demand declines and reduce the need for new investment; profitability will also fall in line with demand thus reducing both the amount of internal funds available for investment and the return on that investment; the growing financial deficit of the company sector will probably result in less investment as companies attempt to move back towards financial balance. The high cost of finance is obviously another major factor constraining investment and the recent rise in long-term interest rates-probably reflecting more pessimistic expectations concerning inflation-will also discourage investment by companies who take a `long view' including those who raise their own funds by issuing corporate bonds. The respondents to the April CBI industrial Trends Survey tend to agree with all the above points and indicate that investment intentions have recently weakened particularly for smaller firms and those firms in the consumer goods industries. Given these factors and the depressed state of housing investment we expect around a 2 per cent fall in total investment this year. We are forecasting a mild recovery of 2 per cent total investment growth for next year, but manufacturing investment may well remain flat as capacity still seems adequate after the rapid manufacturing investment of recent years. Our forecast of a recovery in total investment next year is largely due to our assumed rapid interest-rate falls resulting from UK ERM entry in the first half of next year.

Latest figures have revised stock building downwards for last year and it now appears that de-stocking began in the last quarter of 1989. We believe that further de-stocking in all sectors will be an important element of the slowdown in the economy this year. Milder de-stocking may well persist into next year as the company sector attempts to continue its move towards a more satisfactory liquidity position and the trend of declining stock/output ratios is maintained. We expect most of the negative stock building to occur in distribution as this sector seems particularly sensitive to liquidity constraints and also experienced a probably unintended increase in its stock/output ratio last year.

Balance of payments (tables 5 and 6)

This year's first quarter current balance deficit of 5 1/2 billion [pounds] now means that the monthly deficit for the rest of the year must average around 1 billion [pounds] if the Treasury forecast of a 15 billion [pounds] current balance deficit for 1990 is to be proved correct. However, these first quarter figures should be treated with caution. Abnormally high imports of erratic items (officially defined as ships, precious stones, North sea installations and aircraft) added around C600 million to the quarterly deficit and the oil trade surplus is also temporarily low due to short-term production problems in the oil sector. Looking back over a longer period it seems that the trend growth rates of imports and exports are moving in the right direction. Our forecast for 1990 of substantial destocking and declines in both investment and capacity utilisation will keep imports subdued. At the same time strong world trade growth will keep exports buoyant and the oil trade balance should recover somewhat. We are now expecting a slightly larger fall in the UK effective exchange rate this year compared to our February Review. (This follows from our assumption that UK interest-rate differentials are now higher through 1990 as UK ERM entry is now assumed to occur next year.) Therefore, this year's trade figures will be a combined result of J-curve effects from our expected Sterling depreciation this year and a better volume performance due to improved competitiveness from the previous year's depreciation. Taking these points into consideration we expect the current balance deficit for this year to be around 15.5 billion [pounds].

This forecast improvement on last year's deficit is all accounted for by a much better performance in visible trade as our forecast surplus of 2 billion [pounds] for invisible trade for 1990-consisting of a 4.5 billion [pounds] surplus on services, a 1.4 billion [pounds] surplus on IPD and a 3.9 billion [pounds] deficit on transfers-is lower than the unusually low surplus for last year. The very low IPD surplus for 1990 largely reflects the high interest-rate returns on the deposits held by overseas residents in UK banks. Secondly, our forecast of net revenue from IPD for this year and next is lower than last time because upward statistical adjustments to inward portfolio investment, combined with new information acquired from securities dealers, has resulted in a substantial fall in the UK's net asset position compared to previously published figures.

Next year should see the current balance deficit continue its decline to around 10 billion [pounds]. This partly follows from an improvement in IPD net revenue as UK interest rate differentials fall and as exports of financial services begin to benefit more from the opening of financial markets in Europe as we approach 1992. Visible trade is also forecast to improve again in 1991 as the effects of the depreciation this year improve the price competitiveness of UK tradeable goods.

Our forecast for the components of the capital account (the counterpart of the current account) is shown in table 6. The table focuses attention on the sustainability of the present balance of payments situation by showing the `basic balance' which is defined as the current account plus net structural flows. The basic balance represents the amount of short-term accommodating flows (after allowing for official government intervention in the foreign exchange market and a large balancing item) required to balance the capital account. A very large basic balance deficit arose last year because the problem of a large current account deficit was exacerbated by large net outflows of portfolio investment. These outflows were encouraged by several factors: returns on overseas equities relative to the UK were higher (this includes the effect of an expected depreciation of Sterling resulting in substantial outflows in search of potential capital gains from revaluations); large cash flows into UK life assurance and pension funds resulted in portfolio investment abroad, as the dwindling availability of UK government gilts and the limited number of new equity issues restricted the range of suitable investments in the UK. However, the size of the implied banking sector inflow was much reduced by official intervention.

This year should see a reduction in the size of the basic balance deficit largely because our forecast of a decline in corporate profitability and the probable movement of the company sector back towards financial balance should constrain the amount of finance available for UK investment overseas (especially direct investment). Slower UK equity price growth relative to the world average should encourage the continuation of net portfolio outflows, but the effect on the basic balance deficit is partially offset by our forecast of a substantial improvement in the current balance. However, without the help of the large official intervention of last year the required accommodating net banking sector inflow should remain quite substantial this year. Given that we are forecasting 15 per cent base rates for the rest of this year the high returns on foreigners' bank deposits in the UK will depress our net earnings on IPD. Next year we will probably also see a large basic balance deficit; the improvement of the current balance deficit to under 10 billion [pounds] and reduced net portfolio outflows (as the abolition of stamp duty improves the competitiveness of the UK stock exchange and the scope for currency revaluation related capital gains is reduced as the UK enters the ERM) will decrease the basic balance deficit but the renewal of direct investment outflows as company profitability recovers will have the opposite effect. Nevertheless, IPD revenue should recover somewhat next year as the returns on a similar sized banking sector inflow as in 1990 are much reduced by the assumed rapid interest-rate falls in the UK.

Output and the labour market (tables 7 and 8) According to the latest data manufacturing output is now failing but some other economic indicators suggest that the downturn in this sector may actually be stronger (see Box 1 on capacity utilisation). The output of the energy sector is still adversely affected by production problems in the oil sector and is far weaker than we expected. Within manufacturing, output of the non-durables industries (such as textiles, clothing, footwear and leather etc) is not looking as weak as the durable goods industries, which is consistent with the view that high interest rates are slowing credit-led demand. We expect non-oil GDP growth to slow down to around 3/4 of a per cent this year with output falls in both the manufacturing and construction industries partially offset by increased public sector output arising from higher local government spending. Total GDP growth should be a little higher as oil production begins to recover in the second half of this year. Next year will probably see a renewal of GDP growth to approximately 2% per cent helped by both lower interest rates and a better net contribution to GDP from the balance of payments partly arising from the improvements in competitiveness implied by previous Sterling depreciation.

Recent employment trends are discussed in Box 2 but a few extra points are worth noting here. Although total unemployment has been registering slight falls, some southern regions have experienced actual rises in unemployment which could reflect a stronger decline in demand in these areas due to the higher mortgage debt/income ratios prevalent in the south. Even though total employment is still rising, sectors such as manufacturing are experiencing employment falls. The fall in employment will consist mainly of job losses for full-time males who will be eligible for benefit and hence satisfy the unemployment count criteria. We would therefore expect measured unemployment to rise in the second half of this year possibly rising to around 1.8 million at the end of next year. This will be largely due to manufacturing employment falls and slower employment growth in most other sectors, particularly distribution, although public sector employment will be less affected due to higher spending.

Public sector finance (table 11)

The PSDR was around 8 billion [pounds] for the last financial year. This is approximately 4.5 billion [pounds] below the Autumn Statement forecast and reflects both higher expenditure and lower receipts. The changeover to the poll tax seems to have added around an extra 3 billion [pounds] to local authority expenditure. Additional capital expenditure has also occurred before new regulations come into force. An increasing take up of personal pension plans has also reduced revenue and the high level of interest rates has also increased the cost of servicing public debt. For this financial year our forecast of rising unemployment in the second half of 1990 will increase spending on grants and accelerating public sector pay deals will boost an already high level of expenditure on government services. Our forecast of a decline in corporate profits will further reduce corporation tax this year but government revenue will be helped by a recovery in the oil sector. Looking further ahead we are assuming that, like this financial year, caution will dictate that tax cuts are not appropriate next year especially if interest rates do fall rapidly as our assumed UK ERM entry suggests. Of course, these estimates are very uncertain as political pressure in a pre-election year may bring forth tax cuts in the next financial year and methods to reduce poll tax payments may also be introduced. We have assumed that poll tax payments in the future rise in line with inflation.

THE MEDIUM-TERM FORECAST

One difference between our current medium-term forecast and the February Review is that the current balance deficit does not disappear by the end of the forecast. This is largely due to the fact that European monetary union reduces interest rates further thereby boosting imports through increased demand. Although we allow for some fiscal tightening the extent of fiscal restraint required for a zero current balance would be implausibly restrictive and the most optimistic outcome is that the current balance deficit begins to trend downwards. Although highly uncertain, our capital account equations indicate that UK net structural flows will make the situation worse and a substantial basic balance deficit will also persist throughout the medium and long term. The major net structural outflows in our long-term capital account forecast seem to be developing within direct investment. This is largely due to our forecast of stronger output growth abroad compared to the UK and also continually rising real unit labour costs at home but falling real unit labour costs overseas.

However, the actual outcome for the balance of payments could be better than forecast as we have not allowed for any specific benefits to inward UK direct investment resulting from the reduction of Sterling exchange-rate uncertainty once UK ERM membership takes place. Furthermore, there is some evidence to suggest that inward direct investment helps the UK current balance. First, our exports of manufactures equation includes a stochastic trend which shows an improvement in underlying UK export performance which could be partly due to the benefits of direct investment as overseas investors use the UK as a base for exporting into Europe. Secondly, if we include the real stock of inward investment in our imports of manufactures equation we find that direct investment is negatively signed suggesting that overseas investment in the UK results in the substitution of UK domestic production for imports of manufactures.

The financial balances of the domestic sectors as well as the overseas sector may also remain in imbalance in the medium term. The personal sector will be encouraged to move back towards deficit as interest rates fall rapidly encouraging less savings and more consumption based on cheaper credit and revaluations of wealth. This implies a continuing surplus for the public sector as tax revenues remain high because of buoyant activity and defence spending is reduced in the long run. However, the behaviour of the personal sector's savings ratio is uncertain and our variant in part two shows an alternative scenario where the personal sector saves more.

Given our forecast of rapidly failing interest rates, investment is likely to show a reasonable performance in the medium term which seems to restrain capacity utilisation thus reducing pressure of demand effects and hence inflation. Consumer price inflation slows down in the medium term partly because the sustainable level of utilisation rises relative to actual utilisation as the cost of stock-holding falls in line with interest rates. GDP is forecast to grow at a rate close to its potential trend of between 2 3/4 to 3 per cent throughout the medium-term period. This produces good productivity growth rates especially in manufacturing where employment growth is particularly subdued. However, the strong output growth will tend to move import growth above that of exports and worsen the balance of payments.