Printer Friendly

The home economy.


The forecasts were prepared by Bob Anderton, Andrew Britton and Soterios Soteri, but they draw on the work of the whole team engaged in macroeconomic analysis and modelbuilding at the Institute. Part One of the chapter was written by Andrew Britton, Part Two by Bob Anderton.

PART ONE. RECENT DEVELOPMENTS AND SUMMARY OF THE FORECAST Recent developments The recession which began in the third quarter of last year is deepening, and there is no sign as yet that the end is in sight. The output-based measure of GDP fell by 1-1 per cent in the third quarter, with the main contraction being in production industries, but with a slight reduction in service activity as well. On the basis of incomplete and provisional data we estimate that GDP fell by about a further 1 1/2 per cent in the fourth quarter.

The severity of the downturn is emphasised by the results of the CBI surveys and by the rise in unemployment which has already been recorded. (The latest data at the time the Review went to press are in the statistical appendix.) Early indications are that this recession will differ from its predecessors at the start of the 1980s in that the South is being hit at the outset, and that the job losses are in most industrial sectors, not concentrated in manufacturing and construction.

The origin of the recession, according to the framework of the institute's macroeconomic model, can largely be explained by the financial situation of the company sector as it developed last year. Non-oil company profits (excluding stock appreciation) are estimated to have fallen by 7 1/2 per cent in 1990, having risen substantially in the late 1980s. Fixed investment however, went on rising last year, financed in large part by borrowing. The financial deficit of industrial and commercial companies in 1990 may have been as large as 924 billion (5-1 per cent of GDP). in our research we have identified several routes by which a shortage of net liquidity will lead to reductions in subsequent company spending-on stockbuilding, on employment and on fixed investment.

The problem has been to identify how much of the bank borrowing that has taken place in recent years represents a 'disequilibrium' in company or bank balance sheets. Some of it can be seen as the counterpart to an increase in net investment abroad and the purchase of securities, and in dividend payments. To that extent it might be interpreted as voluntary'. Before the event it seemed as if the situation might be sustainable, provided only that demand turned up and profit growth was maintained. After the event the extent of the borrowing looks imprudent.

This highlights another feature of our model-the importance of expectations about output growth in determining company expenditure. The reduction in company spending in the latter half of last year is understandable if firms revised their expectations about output growth. We would see self-fulfilling prophecies of this kind as amongst the proximate causes of the recession, but the underlying causes were continuing inflation and uncertainty about political events at home and abroad.

The 12-month rate of inflation (as measured by the rpi) peaked in October at 10.8 per cent. Since then world oil prices have fallen back sharply. Moreover the unit values of both exports and imports of manufactures fell in the fourth quarter. The 12-month rate of increase of producer prices was under 6 per cent at the end of last year, and the prices of materials purchased by manufacturing industry were falling.

Wage settlements nevertheless remain high, with public sector settlements still around 9 or 10 per cent, although much lower figures are common in parts of manufacturing industry. This high rate of wage settlement (reflecting presumably the unexpectedly high figures for the increase in the rpi over the last year) ensures that the growth of average earnings cannot slow down sharply during this year. The current rate of wage increases is considerably more than past experience would suggest at this stage of the cycle. The rise in unemployment seems to have had little impact so far.

Interest rates remain high (as we have been predicting since last summer). interest rates might have been reduced in December and January had it not been for the recurrent weakness of sterling within the exchange-rate mechanism. At some stages that weakness seemed to result from uncertainties related to the Gulf war, at others the recession in the UK itself was a cause of unease, or again the situation in Germany was quoted as an explanation. The half-point reduction in mid-February was a cautious first step in a process which the market had expected to start much earlier. The underlying problem seems to be that sterling was introduced into the ERM at an artificially high rate. Between the fourth quarters of 1989 and 1990 the effective exchange rate rose by 6.8 per cent, and by 11-3 per cent in real terms (calculated from relative growth in producer prices) mainly because of the weakness of the dollar. The suspicion that the exchange rate was overvalued can be confirmed by the continuing large deficit on the current account of the balance of payments. Despite the recession the deficit in the fourth quarter of last year still averaged almost 21 billion a month. The fall in import volume during the latter half of last year is consistent with the other indicators of rapidly weakening domestic demand and does not indicate an underlying improvement in trading performance. The growth in export volume on the other hand is still encouraging at 8.2 per cent up in 1990 when world trade volume probably expanded by about 6 per cent. Comparing the situation now with what it was say five years ago, the main deterioration is in the oil balance and in the invisible account. These changes mean that an underlying improvement in competitiveness is needed, not just the maintenance of the status quo. it is difficult to judge what effects the Gulf war is having on the UK economy, directly or indirectly. Anticipation of war, which was mistakenly expected to result in high oil prices, and perhaps high world interest rates as well, may have contributed to the loss of confidence last year. The war itself may have led to the postponement of some decisions simply by creating general uncertainty. So far at least the direct economic impact of the war seems likely to be small, even on the public sector financial balance and the balance of payments. Policy assumptions We have assumed until now that the general election would take place this year. This still remains a real possibility, although some opinion now seems to favour the early months of 1992. In our forecasts we abstract from election effects, by assuming a continuation of present policies and implicitly therefore no change of government. (We discussed the economic policy proposals of the Labour Party in the Review last November.) It is possible that the approach of an election may unsettle financial markets later this year, or early next year, but we cannot hope to foresee what form this effect will take. Our monetary policy assumptions therefore are not in fact very sensitive to the timing of the election.

We have built into the forecast the gradual reduction in interest rates which the authorities would clearly like to see during this year and next. The calculation starts from the assumption of a risk premium for sterling against the D-Mark of 16 per cent in the first quarter. The present interest differential against Germany is then consistent with a 3 per cent depreciation of sterling at an annual rate. if sterling interest rates fall to 12 per cent by the fourth quarter (which was consistent with one-year rates in the market at the time the forecast was being made) the associated depreciation of sterling against the D-Mark can just be kept within the wide margins of the ERM. (This depends on less weighty currencies like the peseta being pulled down by sterling, so that intervention does not have to hold sterling above its floor against the D-Mark.) We assume that the move to narrow, 2 1/4 per cent, margins takes place next year (after the election). The band is set so that it has the same floor as the present 6 per cent band, so the central rate is 'devalued', although the actual rate is unchanged. During 1992 interest rates continue to fall, reaching about 10 1/2 per cent by the fourth quarter. The implication of this profile is that real interest rates rise (on most definitions) and that the real effective exchange rate stays roughly constant throughout this year and next. For this year's Budget we assume no net changes in taxation, relative to a conventional indexed base. It will be difficult to introduce any net increases in expenditure taxes while inflation is still relatively high. Had it not been for the Gulf war (and the extra spending to moderate the rise in the poll tax) there might have been net reductions in taxes on personal or company incomes. As it is, the PSBR is allowed to increase by about 7 billions pounds in 1991/2 compared with the financial year just ending. For the later years of the forecast we assume that taxes rise in line with incomes. This would involve some tax cuts relative to an indexed base as conventionally described. Public spending on goods and services which rose quite sharply in 1990, and again in 1991, is projected into the medium term at just 1 per cent a year volume increase. Outline of the forecast inflation is expected to slow down markedly during the rest of this year. In our model we identify pressure of demand' effects on both wholesale and consumer prices. With capacity utilisation already sharply down from last year profit margins will be cut back further. The effect of falls already recorded in import prices, and of the lower level of oil prices now established, will reinforce the deceleration in consumer prices. in the case of the rpi an additional factor will be the expected cuts in mortgage interest rates. The rate of increase in the deflator for consumer spending (adjusted for the discontinuity caused by the treatment of the poll tax) was about 7 per cent at the end of last year; by the end of this year it should be about 4 per cent. The corresponding figures for the rpi are 10 per cent and about 4 per cent.

The size of recent wage settlements precludes a reduction in the growth of earnings of the same magnitude over this period. The rise in earnings over the 12 months to the fourth quarter of this year is expected to be about 8 1/2 per cent, implying a substantial recovery of real earnings and a substantial fall in profit margins. The pressure on company sector profits and liquidity which appears to be the main factor behind the recession will actually intensify during this year. This will mean a further substantial rise in company failures, bankruptcies and job losses.

We are predicting a year of heavy destocking. This has been a feature of the last two severe recessions in this country, accounting for much of the fall in expenditure. It is possible that on this occasion it will be more difficult to cut stocks because they have not been allowed to rise so high during the upturn. The data for stockbuilding are especially uncertain. If there were to be less destocking than we forecast however, the pressure on balance sheets could be more acute, with deeper cuts in employment or investment as a consequence.

The latest CBI survey presents a particularly poor outlook for investment by manufacturing industry this year. This would not be out of line with the predictions of our own model. We are indeed forecasting considerable falls in almost all categories of investment. In the case of private dwellings perhaps the worst fall is over, with house prices now static and the prospect of a rather better outlook by the end of the year as mortgage rates come down.

Employment probably fell in the latter part of last year and will go on falling throughout this year and next. The sharpest reductions will be in the most cyclical sectors, construction and manufacturing. But on this occasion we expect the falls to be quite general, including self-employment as well as employees and with even the business services sector not excluded. Claimant unemployment was already over 1.8 million last December some 250 thousand above its low point of last spring. It is expected to exceed 2 million by the spring of this year and perhaps reach 2 1/2 million by next winter.

Although the average real incomes of those still in work will recover significantly during this year, we are forecasting no change in the volume of consumer spending. The real value of personal disposable income year-on-year should rise by about 2/2 per cent, but all of that increase will be absorbed by extra savings. In national accounts terms, of course, savings include debt repayments; in fact the reduced availability of consumer credit may be a major factor influencing household finances. The rise in real incomes is, moreover, a temporary one, and households may for that reason spread their extra spending out over a rather longer period. The implication is that the personal sector, which returned to financial surplus last year after a period of abnormal deficit, may now for a year or two be building up its financial wealth rather rapidly. In accounting terms this is the counterpart of the deficits being scored by companies.

The outlook for the world economy described in Chapter 11 below, suggests a reasonably encouraging outlook for UK exports. World trade will continue to expand this year, sustained by demand from Germany and Japan rather than America or OPEC countries. The rise in the real exchange rate through last year is damaging the immediate prospects, but we do not expect much further loss of relative price competitiveness this year. Indeed price competitiveness may even be improving as exporters cut their margins in an attempt to keep up the volume of sales.) Year-on-year the rise in exports of goods and services may be a modest 1 1/2 per cent this year, but with the prospect of a good year to follow in 1992.

Import volume should continue falling in the first half of this year. Some special factors, including the Gulf war, will tend to disguise the underlying improvement in the current account in the early part of the year. By the end of the year the quarterly rate of deficit should be reduced to below 92 billion.

The overall implication of these forecasts is that GDP will fall again in the current quarter, with little change in the second, and a moderate rise in the latter half of the year. We now expect a significant year-on-year fall in GDP this year, the first since 1981. The fall we are now forecasting, 1 1/2 per cent, is similar to that measured for 1974 and 1981, but less than that in 1975 and 1980.

Looking ahead to 1992 the recovery should be well under way, with output rising again but unemployment not yet quite at its peak. Inflation should be roughly constant at about 4 per cent, with the rise in earnings moderating significantly during the year (as a result of lower settlements made during the recession). Profits should recover sharply and the company sector should be heading back towards a more sustainable financial position.

There are several relationships in our model which generate the upturn: the fall in inflation encourages a return to a lower savings ratio for example, the shakeout of stocks cannot continue for more than a limited period, a shortage of capacity encourages an upturn in fixed investment. The experience of several recessions of varying length and severity suggests that a 'spontaneous' recovery does normally occur, but its speed and timing is difficult to predict. On this occasion the judgement is especially difficult because the reduction in interest rates, which we expect to reinforce the recovery, must depend on the conditions in world financial markets and confidence in sterling. PART TWO. THE FORECAST IN DETAIL

Forecasts of expenditure and output (table 1) Since our last forecast the economy has moved further into recession. Manufacturing capacity utilisation has continued to fall and the savings ratio has risen further. Unemployment has increased rapidly recently and both falling inflation and a reduced current account deficit indicate that demand has decelerated considerably. We are expecting negative real GDP growth of approximately 1 1/2 per cent for this year. This occurs even though we assume that the interest rate is cut to 12 per cent by the end of 1991 (this is fully consistent with market expectations for UK interest rates). Oil production will not add much to growth this year due to widespread shutdowns of UK oil fields as new safety equipment is installed. Next year will probably see a recovery in total GDP growth to around 2 per cent. This renewed growth in 1992 will be driven by a resurgence in consumer spending encouraged by further interest-rate cuts. Export growth will also contribute strongly to output next year largely due to world trade growth of around 6 1/2 percent.

In recent months, the current account deficit has shown a marked improvement suggesting that domestic savings have increased or/and domestic investment has decreased. Our forecast for 1991 and 1992 shows a continued and substantial improvement in the current account and a transfer of resources (at constant prices) towards net exports. This is explained partly by a recovery in savings but owes more to a downturn in investment. Although our forecast of the savings recovery in the personal and company sectors is considerable it is somewhat offset by the move into deficit of the public sector. Consequently, substantial falls in both gross fixed investment and stockbuilding are required if the current account is to improve as sharply as we are forecasting.

We are expecting more than a 10 per cent fall in real investment this year followed by no investment growth next year. Very substantial destocking will probably take place this year and perhaps continue (at a slower pace) in 1992. Consumers' expenditure will probably remain flat in 1991 largely due to a collapse in confidence as unemployment rises. This is consistent with slow credit growth this year even though real disposable income will remain buoyant as earnings growth exceeds inflation. Consumers will begin to respond to falling interest rates next year and the savings ratio will stop rising.

The expected large decline in the highly import intensive expenditure categories of investment and stockbuilding will induce a fall in imports of goods and services this year of around 3 per cent. Exports of goods and services will be restrained in 1991 by the recent deterioration in competitiveness and the reduction in profitability arising from both lower demand and higher unit labour costs. The public sector will provide one source of growth this year with real expenditure growing by around 2 per cent in 1991 but slowing down in 1992 as fiscal policy is tightened. Personal income and expenditure (table 2) The most recent data show that real personal disposable income grew at an annual rate of over 4.2 per cent during the third quarter of 1990. This represents very strong growth considering that both high mortgage costs and the poll tax were exerting considerable downward pressure on disposable income. Rising unemployment and an increase in the poll tax above the rate of inflation will further depress real disposable incomes this year. However, we still expect approximately 2 1/2 percent growth in real disposable income in 1991 even after allowing for no tax cuts and falling dividend income. Our forecast of high nominal wage inflation relative to rapidly falling price inflation is the major factor which keeps real incomes buoyant. The continued increase in unemployment and the deceleration in wage inflation next year will probably bring down the growth of real personal disposable income to around 1 per cent in 1992.

Real consumers' expenditure actually fell by 1/2 per cent between the second and third quarters of last year. During the fourth quarter of 1990 retail sales declined by 3/4 per cent compared to the same period in 1989. Another indication of lower consumption is that new motor car registrations were almost 30 per cent lower at the end of 1990 compared with December of the previous year. Given the recent growth in real incomes, most of this sudden rise in the savings ratio can be attributed to the deterioration in consumer confidence indicated by the Gallup survey. The substantial decline in personal sector real wealth, especially in the housing market, combined with pessimistic expectations concerning unemployment, are probably the main economic factors explaining this change in consumer confidence. We are forecasting a further increase in the savings ratio during 1991 resulting from virtually flat consumers' expenditure for this year (with consumer durables expenditure registering a significant decline). The slowdown in real consumers' expenditure this year could therefore be as great as in 1980 . However, in contrast to 1981 when real consumption remained very depressed, we expect a recovery in consumption growth to just above 1 per cent in 1992. This is largely due to the growth in wealth next year as our assumed interest-rate reductions stimulate further growth in the UK stock market and encourage a recovery in the housing market.

High real interest rates are one factor encouraging a rise in the personal sector savings ratio but the deteriorating financial position of the public sector may be relevant. it seems that tax cuts in the late 1980s were immediately spent in the expectation that these cuts would not be reversed in the future due to the considerable public sector surpluses existing at the time (in fact, some argue that the expectation of continued future tax cuts added to the consumer boom). The position now seems to be one where the public sector is sliding into a deficit which will be financed by the issuing of government debt. Therefore, uncertainty surrounding future public debt financing, which itself is exacerbated bY high real interest rates, may encourage private sector saving. Fixed investment and stockbuilding (tables 3 and 4) According to the CSO survey (conducted in September/October 1990), manufacturers' intend investment, in volume terms, to fall by 7 per cent this year and to remain flat in 1992. However, both actual demand and expectations of future demand have deteriorated dramatically since that survey was undertaken. In addition, trading profits of the company sector have deteriorated further and real interest rates have risen as inflationary expectations have been revised downwards. The latest CBI industrial trends survey shows that the number of respondents indicating a tall in spending on plant and machinery in manufacturing is at its highest level for 10 years. The only optimistic part of the CBI survey concerning investment is that a majority of firms still expect to increase spending on training. Spending on product and process innovation is also expected to fare better than fixed investment. Other CBI surveys show that the investment downturn is not restricted to manufacturing as falls in capital spending in financial services and the distributive trades are also expected in 1991. The CBI survey expects that the major decline in manufacturing investment will be among smaller firms which partly explains why the CSO investment intentions survey is more optimistic as the latter is conducted among larger firms. The CBI survey also shows that business confidence has continued to fall and this may have now deteriorated yet further as the survey was conducted before fighting broke out in the Gulf.

We are forecasting that total real fixed investment will fall by around 1 percent this year and remain flat in 1992. Manufacturing investment will experience the largest decline in 1991 (possibly showing an annual fall of perhaps more than 20 per cent at the trough of the recession) but may recover slightly next year. Both distribution (especially the retail sales sector) and financial and business services will experience large investment declines this year which will probably continue, but at a more moderate pace, next year. Housing starts in the three months to November 1990 were about 15 per cent down on the same period a year earlier and we expect further falls in investment in both the housing and construction sectors in 1991. Investment in these two sectors should recover in 1992 as the interest rate continues to fall. Substantial excess capacity (see box A), expectations of low future demand and the large financial deficit of the company sector are the major reasons for expecting investment to decline so severely. However, it should be noted that the financial deficit of the manufacturing sector is not in such a serious position as non-manufacturing and this partly explains why the investment decline in the latter sector may be more protracted. Our model incorporates a vintage capital model of production which predicts long-term negative supply side effects of this expected investment downturn. For example, productivity will be lower in the future than it would have been (with continued investment) as technological progress incorporated in the production process will now be lower.

A major route for reducing the financial deficit of the company sector in 1974 and 1979 was via large de-stocking. However, the scope for de-stocking may be more limited now as computerisation and just-in-time stock management techniques have resulted in a gradual diminishing of the stock-output ratio for all sectors in the 1980s (the ratio of total stocks to production is only around two thirds of its 1980 peak). The CBI industrial trends survey is consistent with this view as respondents do not report the same extent of surplus stocks as in the early 1980s. However, the current financial deficit of the company sector has been built on the confidence of high expected future demand. Therefore, we are forecasting substantial and prolonged de-stocking in both 1991 and next year, in order to restore the company sector to a healthy financial balance. More obviously, stockbuilding will also fall because of the decline in demand. All sectors will undergo substantial de-stocking but it will probably be especially severe in manufacturing and distribution. Looking further ahead, the expectation of reduced exchange rate uncertainty (via ERM membership) should again put downward pressure on the stock/output ratio as exporters will not be so prone to unpredictable changes in demand arising from fluctuations in the nominal exchange rate. Balance of payments (tables 5 and 6) At the end of last year the current balance deficit improved with monthly deficits falling below el billion. A considerable improvement in underlying visible trade has occurred-chart B1 shows the recent improvement in the manufacturing trade balance-but revenue from invisibles is still deteriorating. Net exports of services remain subdued and the UK's high interest rate differential is largely responsible for the poor balance in interest profits and dividends. The turnaround in visible trade was achieved in spite of a depressed oil trade surplus due to continued oil production problems in the North Sea. Looking at the individual volume categories we see that, although export volumes of manufactures grew strongly through the whole of 1990, export growth slowed down slightly in the final quarter of last year (with the exception of passenger motor cars). However, real import growth through last year decelerated considerably, especially imports of capital goods and passenger motor cars although imports of other consumer goods remained fairly high (see chart 1). The reduced imports obviously reflect the sharp downturn in demand but the slowdown in exports may reflect a fall in supply due to declining profitability.

Because of the pessimistic outlook for investment, stockbuilding and total output growth this year, we expect the current account deficit to be almost halved to around 29 billion in 1991. Most of this occurs in visible trade due to the recession disproportionately hitting the more import-intensive categories of expenditure (i.e. investment and stockbuilding). However, recent strong growth in UK unit labour costs and subdued investment may result in a deterioration in both the price and non-price competitiveness of UK manufacturing exports in 1991. The UK oil trade balance will once again remain subdued this year. We also expect some deterioration in the services balance in the first half of the year as tourists to the UK are discouraged by the Gulf war. Also, expenditure by troops abroad will increase our imports of services (this is only for troops from the UK and not troops previously stationed in Germany being moved to the Gulf). In contrast, we expect net revenue from interest profits and dividends to improve in 1991 as the UK's relative interest-rate declines. Transfers will also benefit from war payments from Germany and Kuwait.

The renewed output growth next year will prevent the visible trade balance improving further but a recovery in invisibles will help to stabilise the current account deficit in 1992.

The high level of sterling and the large current account deficit of last year contributed to a decline in UK net overseas assets during 1990. In addition, UK outward direct investment flows fell last year in response to lower UK company sector profitability whereas inward direct investment was substantial due to continued mergers and acquisitions in the run-up to 1992. This resulted in a considerable reduction in the UK basic balance deficit (Structural flows plus current account) in 1990 which will partly continue into 1991 although portfolio investment will probably register a net outflow as pension fund wealth continues to rise and is partly invested abroad. Output and the labour market (tables 7 and 8) Output of the production industries experienced a considerable decline during the second half of last year. Manufacturing output fell by approximately 4 1/2 per cent over this period. Within manufacturing, the chemicals, metals and engineering industries suffered particularly strong declines. This is consistent with data suggesting that output of investment goods has decelerated more rapidly than the production of consumer goods. However, survey evidence indicates that the services sector is also experiencing a substantial decline in activity (particularly in the South-East of England). Three reasons explain why services may experience a stronger decline-in employment and to a lesser extent output-in this downturn compared to previous recessions: first, many sectors which were previously classified as part of manufacturing industry are now contracted-out to the services sector but are still highly dependent upon manufacturing activity. Second, privatised industries in the services sector may now react more sharply to falling demand compared to when they were in the public sector. Third, the high ratio of mortgage debt relative to income in the South-East combined with high interest rates, will particularly affect services demand.

In consequence we expect declines in virtually all categories of output during this year. We are forecasting particularly large output falls in the manufacturing, construction and distribution sectors. Manufacturing output will probably reach a nadir about 6 per cent above its 1979 level. Oil production will remain at a very subdued level due to shutdowns as new safety equipment is installed. Accordingly, our predicted fall in non-oil GDP is similar to total GDP.

Unemployment has been growing rapidly in recent months and will soon pass the 2 million mark. Vacancies have been falling dramatically since the middle of 1990. Overtime and average hours worked in manufacturing started falling around the fourth quarter of last year. It seems that manufacturing began shedding labour much earlier than the services sector but anecdotal evidence suggests that job losses are also now occurring in the service sector.

We expect substantial employment declines in all categories (except for the public sector) both this year and next. One reason for employment cuts is that they are a quick way of restoring company sector profitability when demand is falling. It may be argued that a leaner and fitter manufacturing industry will not require such large employment cut-backs as in the early 1980s recession, but this effect (if present) may be offset by the inability of weaker trade unions to maintain existing workforces.

We expect the official number of workers without jobs to be almost 2 1/2 million by the end of this year. This increase in unemployment will reduce the effectiveness of restart measures and long-term unemployment will rise. Unemployment will probably continue rising next year and may reach a peak of around 2.6 million in the second half of 1992. Given that we are forecasting a recession in non-manufacturing as well as manufacturing, unemployment may be under-estimated compared to the last recession. This arises from the fact that the non-manufacturing sector contains many part-time female workers who do not qualify for unemployment benefit and will consequently not appear in the official unemployment count. Wage and price inflation (tables 2, 9 and 10) Latest data indicate that at the end of last year average earnings were growing at an underlying rate of 9% per cent. Manufacturing earnings were growing less rapidly at 9 1/4 per cent perhaps reflecting the extent of the recession in this sector. Although earnings growth peaked around the end of last year there is little evidence of a trend towards a decline in wage inflation. Many settlements have taken place in the first two months of 1991 and a first quarter annualised average earnings increase for this year of 9% per cent is almost certain. Given rising unemployment and failing wage drift we expect average earnings growth to fall through 1991 to about 8/2 per cent by the end of the year. We expect a further fall in average earnings growth next year. Inflation will tall much faster than earnings growth as workers are somewhat backward looking and wish to catch up with past unexpectedly high increases in the retail price index. We therefore expect a substantial rise in real wages in 1991.

Retail price inflation is now falling rapidly. Profit margins are also beginning to fall in response to the recessionary environment. Nevertheless, given the large falls in manufacturing input prices through 1990 it is surprising that output prices have not slowed down earlier and more rapidly. We are forecasting a substantial tall in consumer price inflation by the fourth quarter of 1991. Large falls in manufacturing capacity utilisation (pressure of demand effects restraining prices) combined with weak import and oil prices will all put downward pressure on inflation. Apart from this underlying slowdown in inflation the retail price index will be further restrained by falls in mortgage interest (as interest rates come down) and a less inflationary poll tax increase in 1991 compared to last year. Public sector finance (table 11) During the recession of 1980-1, a contractionary fiscal stance reinforced a sharp downturn in demand. In contrast, fiscal policy during the present cyclical downturn will tend towards stimulating the economy. it is unlikely that this will take the form of tax cuts in the forthcoming budget but we do expect that the economy's automatic stabilisers will be allowed to operate in full and that public spending will be robust in the pre-election period. The objective of a balanced budget will therefore be relaxed for a while and the public sector will be allowed to slide into deficit. The last Autumn Statement forecast a public sector debt repayment of 3 billion pounds for the current 1990/1 fiscal year. This now looks optimistic given the 5pounds billion deficit so far this fiscal year and we believe that there will be a small surplus of just over 1 billion pounds this financial year. The combination of the Gulf war, recession and public spending pressures in a pre-election year will probably result in a PSBR of around 5 1/2 billion pounds in the 1991/2 fiscal year. These totals reflect the following factors: (1) An allowance for extra spending due to the Gulf

war of around 3 billion pounds. This figure is obviously

uncertain as the war may last for longer, a substantial

proportion may be allocated out of the

contingency reserve, lost equipment will be

replaced over a long period of time and the size

of contributions from other western powers is

unknown. Our assumption of falling oil prices

this year means that higher oil revenues will not

help the war to pay for itself. (2) The recession will obviously increase expenditure

through higher unemployment benefit and

the downturn in activity will depress taxation

revenue. However, our forecast recession is

largely investment and stockbuilding driven

whereas real incomes (for those not unemployed)

will rise as consumer inflation falls more

rapidly than earnings. Consequently, corporation

tax revenue should deteriorate more

markedly than income tax revenue. Revenues

will also be somewhat lower due to lower

national insurance contributions as people continue

to opt out of the state pension scheme. (3) The extra spending caused by pre-election political

pressures and inflation may well have a

larger effect upon the PSBR than the recession.

Cash limits will probably be passed in order to

offset the effect of inflation pushing down real

spending. Additionally, disinflationary pressures

on both public sector pay and monopolistic

suppliers' prices to the public sector will be

weaker than in the private sector therefore

adding to nominal spending pressures.

Given our forecast of a public sector deficit next year, government borrowing will obviously increase and a substantial amount of gifts will be issued. However, without the continued revenue from privitisation, public sector borrowing would be 10 billion pounds higher over these two fiscal years. We assume economic and monetary union within Europe circa 1997 which implies a substantial decline in the UK base rate in the medium term as interest rates in Germany and the UK are gradually equalised. If we further assume that returns on financial assets are internationally equalised, then sterling will depreciate against the D-Mark in line with the interest-rate differential adjusted for the declining UK risk premium. This determines our forecast for sterling. Combined with our view that the UK's underlying inflation rate will continue to be relatively high (at just above 4 per cent), it implies that the UK's real exchange rate will rise in the medium term. This would, in combination with our forecast of a return to a failing personal sector savings ratio from 1993 onwards (induced by low real interest rates), imply a substantial current account deficit in the medium term. Consequently, the UK seems to face a balance of payments constraint on growth throughout the medium and long-term forecast (see box B). We therefore assume fiscal policy is kept tight mainly via low public expenditure growth as we have assumed further medium-term tax cuts, in order to deliver a more sustainable current balance position. This tight fiscal stance, combined with deteriorating UK competitiveness, only allows average long-term real GDP growth of around 2 per cent. Accordingly, a persistently high rate of unemployment is a dominant feature of our long-term forecast. An unemployment rate of 8 to 9 per cent is similar to that experienced by France in its post exchange rate mechanism entry period. However, even with this low long-term GDP growth rate the UK current balance deficit could on average be around 1 1/2 per cent of GDP throughout the medium term and would probably further deteriorate once exchange rates are irrevocably fixed when EMU takes place.

Nevertheless, given the above outcome the UK's position should be fairly sustainable; EMU would ease the UK's external constraint by partially transferring it to the Community level and our assumption of tight fiscal policy (and lower payments on interest rate linked government debt) would indicate a sustainable UK public sector deficit in the medium and long term. Ostensibly, it seems that within EMU the UK would probably be in a healthy position compared to countries like Greece, italy and Portugal who currently have unsustainable public sector budget deficit to GDP ratios. However, on closer analysis a medium-term budget surplus may be preferable for the UK as there is a case for the public sector sharing current burdens and benefits with future generations. Specifically; (1) The 1980s and 1990s will be characterised by

substantial, but temporary, oil taxation

revenues. (2) By the end of the 1993-94 fiscal year, the present

government's accumulated privatisation

proceeds will amount to approximately 950

billion. (3) The public sector has a substantial future

unfunded state pension obligation.

Given the above factors, the government's objective of a balanced budget in the medium term (which actually looks like turning out to be a deficit) may not be conservative enough. APPENDIX: STERLING DEVALUATION VARIANT by Bob Anderton Our main forecast assumes depreciation of sterling in line the UK's risk premia adjusted interest rate differential. As this rate of depreciation of sterling does not breach the UK's ERM entry bands our forecast is assuming that the UK ERM entry rate is sustainable. Given the forward-looking nature of the wage-price nexus in the NIESR model, wage and price setters also expect the same amount of depreciation and incorporate this into their wage claims. This approach implies that both wage and price setters believe that the UK authorities commitment to the ERM entry rate is credible (i.e. past and future high wage and price increases will not be validated by a sterling ERM devaluation).

Although wages and prices in our model are forward looking and incorporate rational expectations, there still remains some nominal inertia in the model's wage and price behaviour. The inertia in our wage equation specification arises from the assumption of a wage-contracting model where only a proportion of the workforce renegotiates wages in any particular time period. Inertia in our price equations partly arises from the fact that there are reputation costs in changing prices. Therefore, prices adjust slowly as consumers are assumed to be more able to recognise relative price increases when the increase is large.(1)

The point of this variant is to examine the results of a realignment of sterling within the ERM such that the UK's effective exchange rate is 5 per cent lower than our main forecast. This devaluation of sterling is assumed to occur in the second quarter of next year. We simulate two scenarios; one where the devaluation is unanticipated. and another where it is anticipated by wage and price setters. The unanticipated case does not require any change in interest rates compared to our base forecast. However, in the anticipated case the interest rate may have to be raised, or fiscal policy tightened, relative to our forecast base, in order to maintain sterling before the devaluation occurs. However, for expositional convenience, the simulation of the anticipated devaluation shown below assumes that interest rates are the same as in the base case. In effect we are assuming that price and wage setters anticipate the devaluation but that the foreign exchange markets do not.

Although the government will not want to realign sterling within the ERM, it is possible that it will be forced to do so. As our main forecast sees sterling close to the bottom of the effective bands in 1 992 then the second quarter of next year is not an unreasonable hypothetical devaluation date. This reasoning implies that the sterling ERM entry rate was too high and is therefore unsustainable in the long run.

The simulation results of the unanticipated and anticipated devaluations are described below. Unanticipated devaluation In this case the results are the same as in our main case forecast for the period prior to the second quarter of 1992. However, once the devaluation occurs we can see from table A1 that average earnings are substantially higher from 1992 to 1994 compared to the base case. The major cause of the increase in earnings is the devaluation-induced rise in import prices combined with the attempt by workers to avoid a deterioration in their real wage. Import prices and earnings also contribute to the simulated increase in consumer price inflation which, in turn, feeds back into earnings. This wage-price spiral continues until the real exchange rate reverts back to its base level. This only takes about 2-3 years in the NIESR model. Although there are some initial temporary gains in competitiveness, the effects upon UK real output from the devaluation are very small, amounting to an extra/2 per cent growth spread over the three years from 1992 to 1994.(2) The balance of payments deteriorates in the short run due to the 'J' curve effect, but actually improves in the long run. This long run improvement in the current balance occurs because the nominal devaluation improves the UK's net interest, profit and dividends position (via upward revaluations of both overseas assets and returns in sterling terms) and also slows down consumption growth as real wealth is reduced due to higher inflation. Anticipated devaluation The results of the anticipated devaluation simulation are similar to those outlined above with the exception that the inflationary effects are evident both before and after the devaluation. For example, purely on the expectation of a 5 per cent devaluation in the second quarter of 1992, average earnings could be growing at 9 per cent in the first quarter of next year (which is 1/2 percentage points higher than our main forecast). Consumer price inflation could be 1/2 per cent higher in the same quarter at just above 5 per cent. After the anticipated devaluation takes place, earnings and consumer price inflation are, of course, again higher than our main case but the real exchange rate will revert back to its base level earlier than in the unanticipated devaluation. The effects on real output and the balance of payments will be roughly the same as in the unanticipated devaluation case, but with a different time profile.

One implication of the last simulation is that the high wage and price inflation evident in the UK economy now may be the result of expectations of a future devaluation. The situation now is in some ways similar to 1979. Then the Conservatives were elected into office with a manifesto promising a strong anti-inflation policy based upon restricting the growth rate of the money supply. The extent to which this policy affected real variables instead of inflation was partly determined by the speed of (downward) adjustment of inflationary expectations which, in turn, depended upon the credibility of the authorities, anti-inflation policy. The large increase in unemployment during the 1979-81 recession was partly the result of the incompatibility of high inflationary expectations and the extent of the reduction in the growth rate of the money supply. This experience can be explained in terms of the expectations-augmented Phillips' curve which shows that a short-run trade-off between inflation and unemployment will only exist as long as expectations of future inflation are not fulfilled. In the long-run, if the labour market does not exhibit hysteresis, unemployment will revert back to the NAIRU (Non-Accelerating Inflationary Rate of Unemployment) but at a higher or lower rate of inflationary expectations.

Now, as in 1979, inflationary expectations must be revised downwards if the UK's entry into the ERM is to be successful. if wage and price setters do not believe that the UK authorities are serious about maintaining the ERM entry rate, and instead believe that high wage settlements will be validated, then unemployment will once again rise (i.e. higher unemployment than our base case). This would be the result if wage and price setters expected a devaluation (as in simulation 2) but that the devaluation did not actually occur.

Obviously some of the above simulation results may be somewhat unrealistic as they do not include the effects of devaluation upon intangibles such as credibility or innovation. For example, a devaluation of the exchange rate implies a loss of future credibility and encourages an expectation of further devaluations. Therefore, the devaluation itself could promote even more inflation than the simulations show purely because of the expectation of further future devaluations. Secondly, there are possible benefits of not devaluing such as encouraging firms to undertake innovations in product design, quality and new products etc which may occur if firms cannot expect to maintain export share via price competitiveness achieved through devaluation.

NOTES (1) Nominal inertia in wage and price behaviour is described more fully in 'Nominal inertia and Keynesian effects', by Simon Wren-Lewis, National Institute Discussion Paper no. 174. (2) Changes in competitiveness are not the only source of any differences in real output from the base case. Although, in our model, once the devaluation takes place both wage and price setters immediately change their expectations to be fully consistent with future outcomes, differences in the relative degree of inertia between wage and price setting behaviour will move the economy along the short-run Phillips curve.

A PC VERSION OF THE NATIONAL INSTITUTE'S DOMESTIC MACROMODEL The Institute's domestic econometric macromodel is now available for use on a IBM PS/2 (or on a close clone). The model comes with a user friendly menu driven front end' in the same style as that available with the Institute's world economy model GEM. This allows the user to do either forecasts or simulations using the model, using as a starting point the institute's own forecast which is updated every quarter. The package requires a 80386 based PC, such as the IBM PS/2 Model 70 or 80, with a minimum of 4 megabytes of memory. A maths co-processor (80387) is also recommended. The computer should also have at least 1 0 megabytes of hard disk, an EGA or VGA graphics card and it it is part of a network it must be possible for it to be isolated.

The Institute's domestic macromodel can claim to be the most advanced large econometric model of the UK economy. It incorporates rational expectations in a number of areas including the foreign exchange markets. The model has a well developed supply side including a vintage technology in each of the manufacturing, distribution and business service sectors. The model is firmly data based, in the sense of incorporating very few imposed behavioural parameters.

The annual subscription is 4,000 pounds plus VAT, or 3,000 pounds plus VAT for those already subscribing to the world model GEM. There is a reduced rate of 500 pounds for academic users. For details of GEM see insert. BOX A. MANUFACTURING CAPACITY AND OUTPUT The CBI industrial trends survey publishes a series on manufacturing capacity utilisation. Combining this with the CSO's published series for manufacturing output it is possible to derive a series for manufacturing capacity. Chart A1. Manufacturing output, capacity and utilisation These three series are shown below. The chart shows that from the latter part of 1988 onwards there seems to have been a substantial rise in manufacturing capacity. The National Institute's model is based upon a vintage capital model of production where a measure of capacity output is calculated by accumulating the output produced on each vintage of capital back to the maximum age of capital retained. For the past, the vintage model tracks the implied CBI capacity series fairly closely. However, to reconcile the two measures of capacity over the last year or so would require an implausibly high value for the maximum age of the oldest vintage retained.

We are therefore somewhat sceptical about the extent of the decline in capacity utilisation. It may be the case that the CBI survey measure contains some element of labour utilisation as well as capital utilisation. However, if we assume that the CBI measure is correct, it implies that expectations of future output were too high. It seems that previous consensus forecasts for 1991 were indeed too optimistic and that recent past manufacturing investment growth was therefore too rapid. A realisation that substantial extra capacity was coming on-stream as demand slowed could explain the sudden change in business optimism and the severity of the present recession. The financial deficit of the company sector would have been more acceptable given the expectation of strong future demand. However, as soon as output expectations declined the financial deficit would be less sustainable and expenditure cut-backs would then occur. BOX B. STRUCTURAL PROBLEMS IN UK MANUFACTURING TRADE UK trade in manufactured goods was in continued surplus until 1982 but is now in considerable deficit (see chart B1). During the mid-1980s the UK current account was roughly in balance as this manufacturing deficit was offset by substantial surpluses on oil trade and invisibles. It was claimed by some that the decline of UK manufacturing trade was the result of efficient market forces as the UK's comparative advantage in trade had moved away from manufacturing towards oil and perhaps services. The rise in the UK's real exchange rate in the early 1980s was therefore interpreted as partly the result of the required change in the relative price of UK manufactures after the oil price rise of 1979. That is, given the UK's new surplus in oil trade, a smaller contribution from UK manufacturing trade was now sufficient for current account balance which was brought about by a rise in sterling itself partly triggered by the rise in oil prices). Furthermore, it was argued that a shake-out of manufacturing industry, achieved via a recession and a deterioration in competitiveness, would produce a 'leaner and fitter' manufacturing base which would recover its share of world trade in the future when the UK real exchange rate fell in response to diminishing UK oil production.

One problem with this thesis is that when the UK's oil resources run out, how will the UK's real exchange rate fall now that sterling is semi-fixed within the ERM? Either this was taken into account when the sterling ERM entry rate was chosen or UK inflation must fall below that of our competitors (presumably achieved via subdued domestic growth). However, in support of the 'leaner and fitter' argument it does seem that there has been an upward shift in the performance of UK manufacturing exports. The current NIESR model equation for manufacturing exports captures this improved performance by embodying a world demand for manufactures elasticity of unity (previously around 0.6 in the 1970s) and a stochastic trend. The latter term indicates that the past decline in the non-price competitiveness of UK exports of manufactures has been halted.

In contrast, UK import penetration has grown substantially over the 1980s and does not suggest an improvement in UK manufacturing performance. The NIESR imports of manufactures equation is an 'import-share' type specification which allocates UK expenditure of manufactures between imports and domestic production as follows:
M/Z 0.08 + 0.13 [Delta] in Z + 0.52 (M/Z)[sub.1] - 0.04 (M/Z)[sub.2] + 0.34
(M/Z)[sub.3] -0.05 In RP [sub.1] -0.03 [Delta] In RP + 0.08 In SPEC + 0.06 In

where M = imports of manufactures Z = total UK manufacturing expenditure RP = import price of manufactures relative to domestic manufacturing prices CAPU = manufacturing capacity utilization SPEC = international specialisation of production proxied by world trade in manufactures relative to world industrial production. Long-run responses Given M/Z in 1989 = 0.4; 1 per cent rise in Z = 1 per cent rise in M 1 per cent rise in RP = 0.7 per cent fall in M 1 per cent rise in CAPU = 0.8 per cent increase in M 1 per cent rise in SPEC = 1 per cent increase in M As this is an 'import-share' equation there is an implied UK domestic output of manufactures equation embodying symmetric coefficients of opposite sign to those above.

The equation shows that the UK demand elasticity for imports of manufactures is unity in the long run. Therefore, ceteris paribus, given that the UK is currently in deficit on manufacturing trade, if world demand and the UK demand for manufactures grow at the same rate then UK manufacturing trade will be in continual and increasing deficit. It we assume that capacity utilisation and relative prices are unchanged then we are still left with SPEC, the ratio of world trade to world industrial production representing international specialisation of production. A 1 per cent increase in the specialisation term results in a 1 per cent rise in imports implying a 0.6 per cent tall in domestic output. This suggests that the UK is far more vulnerable to the process of international specialisation than other countries as a 1 per cent rise in world trade relative to world production results in a 1.6 per cent rise in the same ratio for UK manufacturing. In our medium-term forecast SPEC grows by 3 per cent a year. This implies a growing deficit for UK manufacturing trade unless the corresponding exports trend term, representing the relative non-price competitiveness of UK manufacturing exports, can substantially improve in the medium term. Our forecast predicts that the more likely outcome is that UK demand grows slowly relative to world demand. TABULAR DATA OMITTED
COPYRIGHT 1991 National Institute of Economic and Social Research
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1991 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:United Kingdom's economy
Author:Britton, Andrew
Publication:National Institute Economic Review
Date:Feb 1, 1991
Previous Article:Summary and appraisal.
Next Article:The world economy.

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters