Printer Friendly

The UK economy.

Section I. Recent Developments and Summary of the Forecast

The latest set of official statistics and survey evidence suggests that activity remained subdued in the first half of this year, with the optimism apparent in post-election business and consumer sentiment proving to be ephemeral. However it is not unusual for there to be some delay before movements in survey responses are reflected in measured expenditure. On balance it does appear as if output has begun to stabilise and we expect a modest recovery over the second half of this year, although it is difficult to be confident about the pace at which the recovery will proceed. Any potential recovery is unlikely to prove sufficient to prevent a further year-on year fall in the level of GDP and the aftermath of the recession, in the form of rising unemployment, repossessions and business failures, can be expected to persist into 1993.

In the period immediately following the general election there was a marked improvement in the level of consumer and business optimism about economic prospects. However the most recent surveys have proved disappointing, with confidence falling back to the levels reported in the early part of this year, although such levels are much improved on those in the early part of 1991. In recent weeks the level of equity prices has also fallen in response to weaker world equity markets and doubts about the possibility of further interest-rate reductions and the expected speed of the recovery. At the start of August the FT-All Share Index was some 15 per cent below the high point achieved in May, reversing the gains made in the early months of the year.

As with last October's CBI Industrial Trends Survey, the April Survey appears to have heralded a false dawn. However, although the balance of business optimism fell back once again in the most recent survey, close to 60 per cent of firms failed to report any change in optimism from the previous survey, and there was little change in either investment or employment intentions. Other surveys with a shorter track record than the Industrial Trends Survey have proved slightly more optimistic, with both the British Chambers of Commerce Survey and the CBI Distributive Trades Survey suggesting that prospects for the service sector have started to improve.

Taken in conjunction with the national accounts figures for the first quarter of the year and the improvement in retail sales volumes in the second quarter, the latest set of survey data are consistent with the economy reaching the bottom of the recession in the early months of this year. (At the time of writing the level of GDP in the second quarter was still not known.) In fact domestic demand has already begun to recover, with small quarterly rises experienced from the second half of last year, with the main upward impetus coming from additional public expenditure (both procurement and investment) and a stabilisation of stock levels typical of the early stages of an economic recovery.

However, as Chart 1 illustrates, the rise in domestic demand has failed to boost GDP due to a decline in the level of net exports. Import volumes of goods and services in the first half of 1992 were some 4.4 per cent higher than in the second half of 1991, whilst export volumes only rose by 0.7 per cent. A rise of this magnitude in imports is not unusual at this stage of the cycle, with similar growth being experienced at comparable points in previous recessions due to the high import content of both stockbuilding and new fixed investment in plant and machinery. The impact on GDP is a little greater this time as the share of domestic final expenditures met by imports has risen steadily throughout the last two decades.

The profile of quarterly movements in output over the present recession and the previous two in 1974/5 and 1980/1 is shown in Chart 2. The fall of 0.4 per cent in the first quarter of this year (a little greater than we had expected) serves to make the present recession longer lasting than the previous two, although the fall in the level of output has not been as severe. Our forecast implies that the profile of growth over the latter half of this year and the early part of 1993 will be close to that in the immediate aftermath of the 1980/1 recession.

A further factor behind the weakness of the economy in the early part of this year was the first quarter fall in the volume of consumers' expenditure, to a level some 3 1/2 per cent below the level of expenditure immediately proceeding the onset of the recession. With personal incomes being boosted by both bonus salary payments prior to the election and a rise in net property incomes, the savings ratio rose further to 11 1/2 per cent, a rise of 3 3/4 percentage points from the middle of 1990. Although there may be some recovery in expenditure over the remainder of this year, we expect little change in the level of the savings ratio itself, implying a year-on-year fall in expenditure of close to 1/2 per cent.

Whilst there is an undoubted need for consumers' expenditure to recover before the economy can consistently approach its long-term growth rate, it is possible that recovery can begin without any upturn in consumption. Charts 2 and 3 together show that in the last recession (although not in 1974/5) consumers' expenditure remained depressed for some little time after the upturn in output began.

It seems likely that the caution of the personal sector stems from the perceived threat of job losses and the depressed state of the housing market, with the fall in house prices having accelerated in the early months of the year. Average prices in June were around 6 per cent below those of a year earlier, implying a fall of over 10 per cent in real terms. Chart 4 shows the real value of both the net housing wealth held by the personal sector and net non-housing wealth. The latter has shown relatively little change over the past few years, with the relative buoyancy of equity and gilt prices maintaining the value of the assets held in pension fund and life assurance fund portfolios, and the value of outstanding credit card debts having been reduced in recent months. In contrast, real net housing wealth (the value of the housing stock less outstanding mortgage debt) has fallen sharply, with the level in the first quarter of 1992 being some 24 per cent below the peak achieved in 1989Q3.

It is possible that fluctuations in the value of net housing wealth have a greater short-term impact on income expectations and the level of consumers' expenditure than fluctuations in other net wealth since much of the latter is held on behalf of the personal sector by financial institutions. Moreover, consumers may feel that large falls in house prices are less likely to be immediately reversed than equivalent falls in the prices of financial assets.

The company sector has also taken action to improve the state of its overall balance sheet, with investment having fallen by close to 13 per cent over the last two years and cutbacks having been made to stockholdings and employment. Financial constraints on industrial and commercial companies have been eased, with the ratio of interest payments to income having fallen over the course of the recession, aided by the fall of 5 per cent in nominal interest rates since entry into the ERM. However it is possible that the interest rates charged to some companies, particularly small businesses, may have fallen less sharply, since there is evidence to suggest that spreads have widened to reflect greater default risk.

There is less evidence that the aggregate cost of finance for new investment has risen significantly, despite the high level of real interest rates. As Box 2 in Section II explains, the cost of debt finance is presently above that of equity finance; to the extent that companies are able to raise funds in the equity market the aggregate cost-of-capital is lower than the current level of real interest rates, and this should, in the medium term, help to support investment.

Policy assumptions

In our previous forecasts this year we took the view that UK base rates could fall to 9 1/2 per cent by the year end following an easing in the monetary stance of the Bundesbank. However the recent rise in the German discount rate, plus the pressures on sterling within the ERM now mean that the scope for any easing in domestic interest rates over the remainder of this year is minimal. Indeed it is possible that the existing interest differential may temporarily widen if the UK authorities are forced to provide a practical demonstration of the firmness of their commitment to the present parities within the ERM. However we do expect that the inflation risks within the German economy will have diminished sufficiently to ensure a cut of 1/2 per cent in both German and UK rates in the early months of next year, with further falls over the course of the year enabling base rates to reach 8 3/4 per cent by the year end. The profile of interest rates is shown in Chart 5.

The yield on benchmark long-term government bonds is presently a little over 9 per cent, suggesting that financial markets still anticipate future reductions in the level of UK short-term interest rates. However with the yield on the equivalent German assets some 1 per cent lower, it appears unlikely that domestic rates are expected to fall below those in Germany.

In recent weeks sterling has come under considerable pressure within the ERM and (at the time of writing) is presently trading close to its effective floor. Much of the short-term pressure on sterling can be traced to the aftermath of the Danish referendum on the Maastricht Treaty and the doubts about the outcome of the French vote in September, with the uncertainty over the likelihood of eventual European Monetary Union appearing to raise the risk premiums on currencies whose anti-inflationary history is relatively poor. Our exchange-rate forecast implies that the current pressures on sterling will abate sufficiently to allow sterling to remain within the wider band around the present central rate in the ERM. The recent profiles of the key sterling exchange rates are shown in Chart 6.

Our fiscal policy assumptions remain as discussed in the May Review, with the planned contingency reserve being fully allocated to expenditure over the coming two years and no new net cuts in taxation. The impact of the recession and the easing of the policy stance in last year's Autumn Statement and the Budget are apparent in the present size of the PSBR, which reached [pounds] 13.9 billion in 1991/2 (2.3 per cent of GDP) and is likely to be close to [pounds] 30 billion (4 1/2 - 5 per cent of GDP) in each of the next two financial years (although as Box 1 on page 12 shows the possible margins of error in forecasting the PSBR are large).

Official concern about the implied medium-term prospects for borrowing has led to the introduction of a new system of public expenditure control, with the size of the planning total excluding unemployment benefit being determined in advance of the annual public expenditure negotiations. It appears that the intention is to reduce the level of nominal expenditure without impinging on the implied volume of expenditure, although the reported assumption of inflation averaging 2 1/4 per cent in the medium term seems to be rather optimistic on our own projections. There seems little economic reason for reducing the volume of expenditure and therefore we have taken the view that the new system will still be adjusted so as to allow higher levels of expenditure in line with those in our previous forecasts.

In the short term the state of the public finances means that it may be difficult for the government to make further policy changes in response to the weak economic environment. However it could prove possible to introduce additional policy measures if such measures were presented as part of a package incorporating some policy tightening once the recovery was underway. In Section III of this analysis we consider the short-term impact of two possible policy options - the bringing forward of previously planned capital expenditures and a temporary reduction in the level of VAT.

The scale of government borrowing is already beginning to impinge on financial markets, with the government recently forced to reduce the rates offerred on National Savings Certificates in order to prevent any rise in building society deposit (and therefore mortgage) interest rates. A further policy option that has recently been suggested as a means of stimulating the economy is that of underfunding the PSBR, that is selling fewer government securities than would be required to cover the amount of borrowing less income from National Savings with the shortfall covered by higher sales of Treasury Bills to the banking sector. We have not allowed for such a measure in the forecast since the extent to which it would have any significant economic impact is unclear. The most likely effect would be to further twist the yield curve, with long-term interest rates falling relative to short-term ones. Whilst this could help those in the private sector who wish to borrow long term, it is less likely to be of any benefit to companies and households who are liquidity constrained and who wish to see a fall in short-term rates.

Summary of the Forecast

The published figures for GDP in the first quarter of the year along with the relatively subdued level of activity in the aftermath of the general election have led us to make a sharp downward revision to our previous forecasts for growth over the course of this year. We now expect that GDP may show a year-on-year decline of 1/2 per cent in 1992, a forecast close to that shown in the latest Consensus forecasts. Although domestic demand may still rise by up to 1 per cent in the year, GDP will decline because of the adverse movements in trade volumes, with import volumes rising by around 6 1/2 per cent over the year. We continue to show some recovery in the second half of the year, although this is less pronounced than before due to the recent weakness of the dollar and the postponement of any interest-rate cuts until next year.

On balance it appears that the level of GDP in the second quarter should be little changed from that in the first quarter, with some recovery in export volumes and a reported pick-up in retail sales and new car registrations suggesting that consumers' expenditure should have risen slightly. However we expect little change in the savings ratio over the course of the year, with income continuing to be used for debt repayment and the build-up of precautionary savings. House prices may stabilise dose to their present level over the second half of the year and this should help to restore consumer confidence.

As with our earlier forecasts, the main impetus to growth over the year comes from the rise in the volume of public expenditure, with general government fixed investment forecast to grow by 1 1 per cent, and a turnaround in stockbuilding typical of this stage of the cycle. There is a risk however that with output expectations proving over-optimistic once more in the first part of the year the build-up of stocks, particularly in the retall sector, could be reversed in the coming months.

The outlook for 1993 is somewhat brighter with growth now projected to be around 1 3/4 per cent, although the momentum of the recovery we show is less than that in earlier forecasts in which growth had been close to 2 1/2 per cent. Such a revision is well within the margins of error shown in Box 1 below. The slower recovery reflects the slightly higher level of real interest rates in the present forecast along with a lower level of growth in average earnings.

The counterpart to the continuing weakness in activity in recent months should be further falls in the rate of inflation. The recent appreciation of sterling against the dollar will generate a further fall in the level of imported raw material prices in the third quarter, and possibly even in manufacturing import prices. Over the year as a whole we expect aggregate import prices to fall by 1.7 per cent, following a fall of 2 per cent in 1991.

Basic pay settlements have now fallen sharply and average around 5 1/2 per cent when weighted by employees covered. Although overtime hours have risen in recent months, implying a pick-up in wage drift, we would expect the annual growth in average earnings to fall below 6 per cent by the year end and remain between 5 1/2 - 6 per cent throughout 1993. Unit labour costs should therefore continue to behave counter-cyclically, with the pick-up in manufacturing productivity implying that the year-on-year change in manufacturing costs could be close to zero by the year end.

With import prices falling and unit labour cost growth continuing to moderate, wholesale price inflation (excluding food, drink and tobacco) should be around 2 1/2 per cent by the year end and remain close to that level throughout much of next year. Inflation in the more sheltered sectors of the economy has proved more persistent, with the annual increase in the consumers' expenditure deflator close to 6 1/2 per cent in the first quarter of the year. This partly reflects the impact of last year's change in VAT; with this dropping out of the annual comparison, consumer price inflation could approach 3 1/2 per cent by the year end.

The rise in claimant unemployment began to slow in the second quarter of this year, with the average monthly increase being 23,000, some 11,000 below the average in the first quarter. However it seems likely that some firms were hoarding labour in the spring in anticipation of a stronger and earlier recovery than is now expected. Thus there may be some acceleration in the rate of increase in unemployment over the coming months, with the total averaging close to 3 million in 1993.

As we have commented in previous Reviews, the rise in unemployment has been accompanied by changes in the duration structure of the unemployed, with a marked growth in the numbers out of work for over six months. At present around half of those claiming benefit have been out of work for over 26 weeks; our projections suggest that this proportion could rise to close to 60 per cent by the middle of next year. This change in the composition of unemployment could limit the extent to which the rise in unemployment moderates wage pressure, particularly in the absence of measures designed to help with retraining and relocation.

Medium-term projections

The medium-term projections shown in Table 12 of this analysis of the UK economy continue to be built around the assumption that a sufficient degree of convergence is achieved between the main European economies over the 1990s to enable a Monetary Union to be formed. Thus nominal interest differentials narrow further with no adjustment in the UK central rate within the ERM. The forecast still shows a small continuing movement of the sterling effective rate as the D-Mark-bloc depreciates against both the dollar and the yen, in line with the assumed interest differentials.


It seems unlikely that such movements in the nominal exchange rate alone would be sufficient to ensure that prices in the UK relative to those elsewhere adjust to a level compatible with a sustainable current account position and eventual monetary union within Europe. The article elsewhere in the Review by Keith Church estimates that the UK equilibrium real exchange rate was some 5-10 per cent below the actual real exchange rate at the time of entry into the ERM, an estimate in line with our own calculations at the time and the estimates published in last August's Review.

Three alternative measures of the UK real exchange rate are shown in Chart 7, along with their forecast profile up to the middle of the decade. The first is defined using UK consumer prices relative to consumer prices in the Major 4 economies, the second uses UK manufacturing export prices relative to a weighted world export price index and the third uses domestic manufacturing prices in the UK relative to the price of imported manufactures. All three suggest that the actual real exchange rate has continued to appreciate subsequent to ERM entry, although the relative consumer price measure may be a poor guide to trade competitiveness since it includes the prices of many non-traded services.

It is of interest to compare the second and third measures. There appears to be little upward drift in the relative export price series. This is to be expected since exporters need to hold their prices down in an open trading environment if they are to maintain market share. The rise in domestic prices relative to import prices suggests that domestic labour costs have risen relative to costs elsewhere and that export prices have only been held down by a substantial cut in profit margins. Such discounting largely reflects the disciplines imposed by membership of a fixed exchange-rate system.

From these measures it seems that some depreciation in the real exchange rate must occur over the coming years in order that UK manufacturers are able to remain competitive. At present the proportion of firms in the CBI Survey regarding relative prices as a constraint on exports is double that regarding the general world economic situation as a constraint. The adjustment in the real exchange rate plays an important role in determining the reported medium-term projections. In the absence of any realignments within the ERM, or indeed a withdrawal from the ERM, the adjustment in the real exchange rate has to occur through adjustments in relative prices, with inflation in the UK falling below that of the average in her main trading partners for some time.

The present trend in domestic prices relative to import prices suggests that this cannot be done by simply cutting profit margins alone. Whilst price competitiveness determines whether firms can sell in particular markets, cost competitiveness ultimately determines whether they remain in business at all.

The forecasts show each measure of the real exchange rate depreciating by close to 4 per cent over the period from their peak in 1992 to 1995. The process of price adjustment may well prove protracted and require the UK economy to grow quite slowly for some period of time. Over the period from 1990 to 1996 the average annual growth rate in GDP is less than 1 1/2 per cent. This points to a continued downward trend in employment, especially in the manufacturing sector, if unit labour cost growth is to be contained. In many ways this experience would mirror that of the French economy in the 1980s with a prolonged period of growth below potential as the counterpart to relative price convergence.

Our medium-term projections show world trade growing over 6 per cent, in line with the projections discussed in the World Economy forecast. With domestic demand growing between 2 1/2 - 3 per cent, this provides a considerable boost to the growth in net trade volumes, even allowing for the trend specialisation terms in our trade equations. In spite of this, the depreciation in the real exchange rate does little more than help to stabilise the current account deficit at a level between 1 1/2 - 2 per cent of GDP.

Section 11. The Forecast In Detail

Personal Sector (Table 2)

The personal sector is presently in a phase of balance sheet adjustment for which there are few recent precedents, with the perceived constraints from the build-up of debt since financial deregulation appearing to have a greater impact on spending than the reductions in savings and mortgage rates over the past two years. Disposable income continues to be used both for debt repayment and to finance precautionary savings in the face of continuing falls in the value of the collateral on which much of the outstanding stock of debt is secured. Mortgage equity withdrawal has been much reduced in recent months and was possibly even negative over the first quarter as a whole. The scope for future equity extraction has also been limited by the fall in the net value of the housing stock.


Recent consumer surveys paint a much weaker picture than in the immediate aftermath of the general election, possibly reflecting the continued downward drift in house prices. In fact the national house price figures hide some significant regional differences, with highly geared households tending to be concentrated in the South. Such households, who tend to have a slightly higher propensity to consume than the national average, have been relatively more exposed to the effects of the present recession which has been more widely dispersed, both regionally and sectorally, than the recession in the early-1980s.

Real personal disposable income (RPDI) rose by 1.2 per cent in the first quarter of the year (to a level some 0.4 per cent above the level in the first quarter of 1991), helped by a rise in pay bonuses in the run-up to the general election and a higher level of dividend payouts by the corporate sector. Over the year as a whole RPDI may grow by close to 1 per cent, with a further rise of 2 per cent in 1993 as the impact of falling wage settlements is offset by the full year impact of the reduction in income tax in the Budget and the lower level of mortgage rates. (There is a risk that the differential between mortgage and base rates may widen over the next year with the increased costs being incurred by insurance companies as a result of repossessions being passed on to borrowers.)

In spite of the rise in disposable incomes in the first quarter, the level of consumption fell by 0.7 per cent, to a level 2.6 per cent below that two years ago. As a consequence, the savings ratio rose further to 11 1/2 per cent. The improvement in sales volumes identified in the CBI Distributive Trade survey along with the pick-up in total retail sales (up some 0.6 per cent in the second quarter) suggest that expenditure may have risen a little in the second quarter of the year. However we retain our earlier view that there will be little change in the savings ratio over the course of the year (especially in the absence of any further reductions in interest rates) implying that year-on-year the level of expenditure may fall by 1/2 per cent. This is the main change accounting for the downward revision to domestic demand from our earlier forecasts.


Investment and Stockbuilding (Table 3 and 4)

Over the course of the recession investment volumes have fallen by some 12 1/2 per cent, although as Chart 8 shows the level of investment remains at a relatively high proportion of overall GDP. As shown in the May Review it seems quite possible that potential output in some sectors has continued to rise during the recession in spite of the slowdown in investment levels. The short-term prospects for investment are limited with companies reluctant to expand capacity any further in the face of continuing uncertainty about demand.


In fact there was a small rise in investment in the first quarter of the year, helped by some additional public sector investment in infrastructure and the relative buoyancy of new investments by the energy and water supply industries. It seems likely that much of the additional expenditure on plant and machinery helped to boost imports of capital goods since there has been little sign of any rise in output from indigenous investment good suppliers. Overall, we expect public sector investment to rise by close to 11 per cent this year, helping to limit the fall in aggregate investment volumes to around 1/2 per cent. This hides considerable falls in some parts of the private sector. Manufacturing investment, where our econometric model continues to successfully track developments during the recession, may decline by 9 per cent over the year, with any upturn in demand being met by raising the utilisation rates of existing equipment.

A typical feature of the early stages of any economic recovery is a pick-up in stockbuilding, with firms starting to rebuild stock levels in anticipation of future demand. There are signs that this began to occur in the first quarter of the year, where there was little overall change in total stock levels, with a continuing rundown of manufacturing stocks being offset by a pick-up of [pounds] 380 million (constant prices) in stocks in the retail sector. With demand proving weaker than expected in the early months of this year, it seems possible that stockholdings could once again be above planned levels, suggesting some possible destocking, particularly in distribution, in the second half of the year. Overall, the level of stocks may be little changed over the year, with the implied turnaround from the destocking in 1991 helping to add 0.9 per cent to the level of GDP.

Our econometric model suggests that attempts by firms to restore their net holding of liquid assets to desired levels have played a key role in constraining expenditure over the course of the recession. Recent data suggest that the extent of |disequilibrium' liquidity may have begun to decline although it remains a constraint on factor demands. The improvement in liquidity is also reflected in the latest figures for the liquidity ratios (the ratio of total current assets to total current liabilities) of large companies (accounting for three quarters of all ICCs) shown in Chart 9.

An additional uncertain factor in assessing the prospects for the company sector is the extent to which it will prove possible to raise external finance with which to finance new expenditures. Excessive exposure to the property market, either through direct loans or through loans for which property is used as collateral, could serve to constrain the supply of bank credit in the future. Although there have been several well publicised cases in recent months, it does not yet appear that bankruptcies have had an undue influence on the state of banks' balance sheets. The recently announced interim results of the four main clearing banks show an improvement in their capital adequacy ratios, although this also reflects a widening of spreads (designed to raise operating profits) which itself may serve to check borrowing.

If inflation remains at its present low level, it may be the case that equity finance is more attractive than debt finance. Chart 10 in Box 2 indicates that the weighted cost of debt and equity finance has remained fairly flat over the past six years and is presently at a level a little below that in the aftermath of the previous recession. It is possible that this may hide restrictions on smaller companies who have limited access to the equity markets. Such firms may need to rely more on retained savings in the future, suggesting the prospect of further redundancies and destocking in order to improve cashflow.

Balance of Payments (Tables 5 and 6)

With the stabilisation of stock levels in the corporate sector and the pick-up in investment in the first quarter of this year, the current account deficit has begun to deteriorate quite sharply, with the deficit in the first half of the year estimated to be some 5.1 billion [pounds], close to the deficit for the whole of 1991 (although these figures may well be subject to substantial revision). Whilst export volumes (on an overseas statistics basis) this year have been little changed from their average level over the second half of last year, imports of goods have risen by over 4 per cent, with imports of both semi and finished manufactures rising by over 5 per cent. Over half of the latter is accounted for by a rise in imports of cars. Our econometric model suggests that the rise in import volumes has been a little higher than that which could be accounted for by the domestic demand for manufactures and price competitiveness alone.


The impact of the rise in import volumes on the visible trade balance has been limited by an improvement in the terms of trade. Whilst overall export prices were flat in the first half of the year, import prices were some 1 1/2 per cent lower than in the second half of last year. We expect a similar fall in import prices in the second half of the year, with materials prices proving especially weak due to the recent appreciation against the dollar. However export prices could also prove weak, with export price expectations in the latest CBI Survey being lower than at any time since 1962, and our model suggesting that firms have continually reduced profit margins during the past two years.

We anticipate some renewed pick-up in export volumes over the second half of the year, with the temporary loss of competitiveness implied by the weakness of the dollar offset by the gradual recovery in domestic demand in North American markets. Overall this year world trade (as measured by the growth in import volumes in the UK's main export markets) is projected to rise by 4 per cent, with trade having continued to rise steadily in spite of the slowdown in the world economy. Next year trade growth could be close to 5 per cent.

Overall, the current account deficit may be close to 10 billion [pounds] this year (1.7 per cent of GDP), with the deficit next year rising to over 12.5 billion [pounds] (1.9 per cent of GDP) as the pace of the recovery begins to pick up. We expect to see some deterioration in the invisible balance this year, with net transfer credits falling substantially. There may be some improvement in net IPD credits, with the fall in domestic profitability depressing earnings on inward investments.

Output and Employment (Tables 7 and 8)

The overall fall of 0.4 per cent in output in the first quarter of the year was concentrated in the energy and water supply sector where production fell in the quarter by 4 1/2 per cent. However there were additional small falls in most sectors with the exception of transport and manufacturing. In contrast to the previous recession of the early-1980s the falls in output have been widely spread, with manufacturing output having fallen by 7 per cent since the first quarter of 1990, construction by 14 per cent and service output (distribution and business services) by 5 per cent.


Recent survey evidence from the British Chambers of Commerce and the CBI suggests that the prospects for the service sector may now be slightly brighter than for the manufacturing sector, where the weakness of domestic and export markets remains a problem. Indeed it is possible that manufacturing output could slip back slightly in the third quarter, with firms reacting to the slow pace of the recovery by further reducing stock levels.

Unemployment has continued to rise as expected over the course of the year and in June stood at 2.72 million, some 450,000 higher than a year earlier. Over the same period longer-term unemployment (over 26 weeks) has risen by 470,000. Recent movements in the duration structure of unemployment are shown in Chart 11.

The rate of growth in the claimant count has begun to moderate in recent months although it seems likely that this was partially due to firms hoarding labour in the expectation of a rather stronger recovery in the second quarter. Manufacturing employment has continued to decline at a pace similar to that observed in 1991, reflecting the weakness of sectoral output and the need to improve competitiveness. In contrast, service sector employment rose slightly in the first quarter of the year. In a reflection of the widespread nature of the recession, regional unemployment rates have begun to converge to a larger extent than seen in the early-1980s, with many of the job losses occurring in the South.

With the recovery from the recession likely to prove slow, it seems probable that unemployment will continue to rise throughout 1993, to over 3 million in the second half of the year. As the labour shake-out has kept pace with the decline in activity (in contrast to the early-1980s recession) it is likely that unemployment will begin to decline a little earlier in the aftermath of the present recession. However the falls may be limited by the continuing need to improve efficiency in industries such as financial services which are likely to be subject to greater competition within the European Market.

Average Earnings (Table 2)

The whole economy median pay settlement recorded by the Industrial Relations Service in the three months to June was 4.2 per cent, a little lower than in the first quarter of the year. The weighted median settlement, taking account of the number of employees covered, was higher at around 5 1/2 per cent, reflecting the relative size of the public sector groups who received 6 per cent plus pay rises in April. Overall, recent settlements imply that there should be further falls in the level of average earnings growth. With the high pay settlements in the first half of 1991 dropping out of the annual figure, growth in the year to May was 6 1/2 per cent, some 2 1/4 per cent lower than a year earlier.

With headline inflation forecast to fall to around 3 1/2 per cent by the year end and unemployment continuing to rise, there is unlikely to be any substantive rise of settlements from their present range, although productivity improvements could help to hold up the level of real wages. To the extent that such improvements are achieved by making lower paid employees redundant, the decline in overall average earnings growth could be smaller than might be expected given the present conditions in the labour market. We expect that earnings growth will fall to around 5 3/4 per cent by the end of the year and remain at that level for much of 1993.

Two additional factors that could help to hold up earnings growth over the next year are the pick up in wage drift and the increasing substitution of pay in place of non-pecuniary benefits (such as company cars) whose tax advantage is steadily being diminished. Overtime working in manufacturing has already begun to pick up, with the numbers of overtime hours per week rising from 9.89 million in March to 11.26 million in May - a little below the level at the onset of the recession. If labour mobility continues to be hampered by the weak housing market, overtime could rise quite sharply once the recovery gets underway.

In the medium term skill shortages might once again serve to hamper the efficient working of the labour market. Although such shortages are not presently cited as a constraint by many firms in the CBI Survey, around a quarter of recruiters in the latest British Chambers of Commerce Survey report difficulties in finding appropriately trained staff.

Costs and Prices (Table 9)

There is at present a clear downward trend in all measures of inflation, although the drop in the more sheltered sectors of the economy continues to lag behind that in the tradeable sectors. For example, the imputed rent component of the overall consumers' expenditure deflator rose by 14 per cent in the year to the first quarter, reflecting the rise in rental charges apparent in the Family Expenditure Survey.


Domestic price expectations in the manufacturing sector remain weak. In the latest CBI Survey more firms said that they intended to reduce prices than raise them, the lowest recorded balance since early 1967. Producer price inflation has fallen quite sharply, with inflation over the year to July being 2.8 per cent, some 2.2 per cent lower than a year earlier. We anticipate that inflation will fall to 2 1/2 per cent by the year end and remain at that level throughout much of 1993.

Cost pressures are presently being eased by a combination of the slowdown in pay settlements, the pick-up in productivity and the continued weakness of raw material and fuel prices, with input prices in manufacturing falling again in the second quarter. Over the year we expect input prices to fall by close to 2 per cent, with prices weakening further in the second half of the year following the recent appreciation against the dollar. Unit labour cost growth in manufacturing may be close to zero by the year end, with manufacturing productivity forecast to rise by over 4 per cent this year, close to the recent average of pay settlements in the sector.

Retail price inflation fell to 3.9 per cent over the year to June, with average prices remaining unchanged between May and June due in part to a seasonal fall in food prices. In contrast to the sharp falls experienced last year, much of the recent drop in inflation stems from falls in the underlying measure of inflation excluding mortgage interest payments. In fact the falls in the underlying measure in the second half of this year could be greater than those in the headline measure, with last year's mortgage-rate cuts dropping out of the annual comparison. We expect RPI inflation to be around 3 1/2 per cent by the year end and to possibly dip below 3 per cent during the course of next year.

Public Sector Finances (Table 11)

The provisional estimates for the PSBR for the first quarter of the fiscal year show that (seasonally unadjusted) borrowing amounted to 10.8 billion [pounds] compared to 7 billion [pounds] in the same period a year ago. Over the last fiscal year government borrowing amounted to some 13.9 billion [pounds], with general government expenditures (excluding privatisation proceeds) rising from 39 3/4 per cent of GDP in 1990/91 to 42 per cent and tax revenues declining a little in real terms.


Our latest forecasts continue to show borrowing in the current year and in 1993/4 close to that forecast in the Financial Statement and Budget Report (FSBR). Thereafter, we show a much slower decline, with borrowing remaining above 30 billion [pounds] in 1994/95. The present forecasts are little changed from our previous forecasts in spite of the slower recovery we now show. There have been a number of minor model changes, which have led us to raise our forecast of receipts from business rates and indirect taxes. However the lower level of inflation has also helped to reduce expenditure a little.

As in our previous forecasts, the slower growth in income tax revenues that might be expected given the slowdown in the growth rate of the wage and salary bill, is partially offset by our assumption of continuing fiscal drag, with tax allowances uprated in line with prices rather than earnings. In addition we estimate that the value of mortgage interest tax relief will continue to decline, possibly to under 5 billion [pounds] per annum in 1993/4 (compared to 7.8 billion [pounds] per annum in 1990/91).

As we explained in Section I, we remain to be convinced that the new system of public expenditure control will be as effective as intended. In fact the model continues to generate a level of medium-term expenditure close to that projected in the FSBR, with our forecasts suggesting that general government expenditure (inclusive of privatisation proceeds) will rise from 236 billion [pounds] in 1991/2 to 292 billion [pounds] in 1994/5. We continue to assume that public sector pay will grow in line with average earnings; at present the growth is slightly higher, with many workers receiving a 6 per cent pay rise in April and the police due to receive a 6 1/2 per cent rise in September. The recent decision by the government to overrule the recommendations of the Top Salaries Review Body and limit rises to 4 per cent, would seem to suggest that controls may become tighter in the future.

Section III. Some Illustrative Fiscal Packages

In recent weeks, with renewed evidence of the weakness of the economy, several commentators have raised the possibility of the government introducing a fiscal package in order to prevent any further downward spiral in the level of activity. With monetary policy directed towards the maintenance of sterling within the ERM any (temporary) policy adjustment has to come via fiscal policy. One example of such an adjustment has recently been provided by the temporary relaxation of stamp duty on housing transactions over the first eight months of this year.

The general argument for interventionary fiscal measures of this kind is that they could be of benefit to private sector decision makers if they can help to make the growth rate of the economy more predictable. On this occasion they could also help to assuage fears of a continuing fall in the level of output. Over the last two years uncertainty about future demand and the timing of recovery has been cited by the majority of respondents to the CBI Survey as the most important constraint on investment decisions.

Of course the timing of any schemes could prove to be misplaced; if activity in the private sector is more robust than expected, then discretionary changes in fiscal policy could add to inflation rather than real output. However at the present time it appears that the economy will remain weak for some time to come, even allowing for the degree of forecast uncertainty reported in Box 1. An additional constraint on the scope for further significant discretionary relaxations in fiscal policy may be provided by the present size of the government borrowing requirement, with our forecasts indicating that the PSBR is likely to exceed 4 1/2 per cent of GDP in the present fiscal year and in 1993/4.

The rest of this section discusses the possible impact of two alternative fiscal measures - a rise in the level of public sector capital expenditure and a temporary reduction in the level of VAT. These measures are illustrative and are not incorporated in our main forecast which is conducted under the conventional assumption of an unchanged policy stance. In both cases we introduce a package that is designed to add 2 billion [pounds] to the level of the PSBR over the second half of the current fiscal year. To ensure that these measures are not seen as having a permanent impact on the level of public sector debt we assume that the packages also include measures that tighten fiscal policy accordingly over the fiscal years 1993/4 and 1994/5 so that the overall impact on borrowing is neutral.

The Alternative Fiscal Packages

The investment package we consider consists of a mixture of higher housing investment, possibly implemented by allowing local authorities to use the funds they have accumulated from council house sales, and other non-housing investment. Because of the relative speed with which the package is introduced, much of the latter is likely to be accelerated purchases of assorted capital equipment. The total extra current price expenditure is some 2 billion [pounds], with some 570 million [pounds] being added to housing investment. The package is offset by cuts of 1 billion [pounds] in the level of planned investment in each of the following two fiscal years.

For VAT we assume that the tax rate is cut by 2 percentage points for a six month period, with an approximate cost of 2 billion [pounds]. We assume that the foregone revenue is recouped by a rise of 1 billion [pounds] in indirect tax receipts in 1993/4 and 1994/5, with a two year delay in the abolition of stamp duty on share transactions raising around 800 [pounds]-850 million [pounds] each year and a rise in other indirect taxes raising some from 150 [pounds-200 million [pounds] per year. (To give one possible example, this sum is equivalent to the extra revenue from 5 [pounds] added to Vehicle Excise Duty.)

The main difficulty with the VAT reduction lies in obtaining an estimate of the extra expenditure that could reasonably be expected to be brought forward as the result of such a move. Fortunately we have some experience of modelling the effects arising from anticipated changes in indirect taxes in 1968, 1973 and 1979. In common with many other macroeconometric models the equations for total consumers' expenditure and expenditure on durable goods in the Institute domestic model certain dummy variables designed to capture forestalling effects. We assume that the proportionate effect on the level of consumers' expenditure in the present variant will be the same as the average observed during the forestalling of the tax changes in 1968, 1973 and 1979.

The estimated coefficients imply that total consumers' expenditure at constant prices could rise by some 2.2 billion [pounds] over the six month period (1.6 per cent of the level of expenditure in our main forecast), of which some 675 million [pounds] would be on durable goods. To the extent that the calculations are unduly influenced by the experience of 1979 when there was a much bigger switch in VAT, our estimates of extra expenditure should be regarded as upper estimates. We make the further assumptions that some two-thirds of the extra expenditure occurs in the final three month period of the VAT reduction and that all of the extra expenditure is brought forward from the fiscal year 1993/4.(1)

The Macroeconomic Impact

The short-term economic impact of the two packages is summarised in Tables A and B. In both cases we assume that the exchange rate and normal interest rates remain fixed at their values in the main forecast. It is apparent that the initial impact of the packages is to boost the level of output, with the impact of the VAT reduction a little larger than that of the investment package (although this conclusion is sensitive to the calculations about the likely degree of forestalling). The impact on GDP is smaller than the impact on domestic demand since much of the additional expenditure is used to purchase imported goods, although the rise in imports may be overstated in the investment package as our model does not allow for any possible differences in the import content of public and private sector capital expenditures. The rise in imports causes the current account deficit to rise by over 1 billion [pounds] during the initial six month period.


Subsequently the level of output is a little below that shown in the main forecast, reflecting the degree to which planned expenditures have been brought forward. However the economy continues to grow steadily as can be seen from Chart 12. In both cases, the fiscal intervention could help to boost the confidence of industry. Table B shows that the earlier recovery in demand has also helped to bring forward some additional investment expenditures. Although the ex-ante initial cost to the public sector of both packages is designed to be 2 billion [pounds], the actual level of extra borrowing is somewhat lower than this, with the higher level of activity helping to bring forward extra tax revenues, particularly from indirect taxes.

A further point of interest with the VAT package is the extent to which retailers choose to use the temporary reduction in VAT as a means of boosting profit margins rather than reducing prices. Here we allow the VAT reduction to be passed on. The effect serves to change the short-term profile of inflation, with inflation lower in 1993 but a little higher in 1994 and 1995. In turn this allows some of the lower activity effects following the restoration of VAT to the existing rate to be temporarily offset by an improvement in competitiveness.

One notable feature of both packages considered here is that they have a limited impact on employment. Even in the VAT example, employment is only some 11,000 higher by the middle of 1993. To some extent this simply reflects the extent to which both schemes simply redistribute the timing of planned expenditures. However it is also the case that the size of the schemes (equivalent to the cost of a 1 penny reduction in the basic rate of income tax) is small in relation to the overall size of the economy. Whilst both measures could send a signal to the private sector that the recession was coming to an end, they are likely to do little to prevent a further rise in unemployment.


(1) We assume that 40 percent of the expenditure would have occurred in 1993Q2, 30 per cent in 1993Q3 and soon.

Box 1. Forecast Error Margins

All forecasts of future events are inevitably given with a degree of uncertainty even though the possible margins of error are rarely emphasised in any great detail. One notable exception in the UK is the Treasury who publish information about past forecast errors in both the FSBR and the Autumn Statement. Such information on the size of risks attached to a forecast provides the basis for an assessment of how well forecasters might be expected to predict future developments.

This Box reports the average absolute errors made in the NIESR forecasts published in the August Review over the period from 1981. The average absolute error, a measure of accuracy, is defined as the latest available outturn less the forecast. Errors are given for both current year and one year ahead forecasts. All figures relate to calendar year forecasts apart from inflation (fourth quarter on fourth quarter) and the PSBR which is for the fiscal year. The errors only provide a broad guide to the size of error that might reasonably be expected since the margin of error itself could change over time.

The figures in Table A1 indicate that even halfway through the year there is still considerable uncertainty about developments over the course of the year. For example, the average error for current year GDP forecasts is around 3/4 per cent. Thus our central forecast of a decline in GDP of 1/2 per cent in 1992 could be consistent, given the average of the errors made in the recent past, with a fall of over 1 per cent or even a small rise. The outlook for the year ahead is yet more uncertain since forecast accuracy deteriorates as the forecast horizon is extended. Thus the average absolute error attached to our forecast of GDP growth in 1993 is over 1 1/2 per cent.

Box 2. Real Interest Rates and the Cost of Finance by Garry Young

One of the key determinants of the user cost of capital that influences investment and factor mix decisions is the cost of finance: the real rate of return demanded by capital markets for the finance they supply to the investing firm. If companies are managed in the interests of their shareholders, then investment projects need to generate a return sufficient to cover this cost.

The real cost of finance in the National Institute's domestic model is determined as a weighted average of the real cost of equity and debt finance. It may be thought of as being measured by the ratio of current company earnings on domestic operations to the financial valuation of the company sector's domestic capital and as such is similar to the measure described by Flemming, Price and Byers (1976).(1) Chart 10 compares the recent behaviour of the weighted cost of finance with an estimate of the short-term real interest rate.(2)

In the past five years the real rate of interest has varied from a low of - 1 per cent in the second quarter of 1988 to a high of 9 1/2 per cent in the third quarter of 1990 and to a rate of about 6 1/4 per cent now. Over the same period the cost of finance has been much more steady and remained within a range of 3 1/2 per cent to 6 per cent and is now at about 4 per cent. The primary reason for the greater steadiness of the cost of finance is that movements in the cost of equity capital, which have a greater weight in determining the cost of finance, have been less marked than movements in the real rate of interest.(3)

There are two main reasons why the cost of finance might differ from the real rate of interest. The first is due to the system of capital taxation in the UK which tends to encourage investors to hold equity rather than debt. This has the effect that investors are prepared to accept a lower gross return on equity investments than on debt holdings. In the absence of risk this would tend to make the cost of finance about two percentage points lower than the real interest rate. The other factor is risk and this tends to raise the gross return that investors require from equity investment. But over the 1980s and early 1990s, the risk premium has been quite small (adding just under 2 per cent on average to the cost of capital) and at present is not sufficient to offset the tax benefits of equity investment.

Thus the evidence of recent years indicates that the cost of finance to companies has not been adversely affected by the sharp movements and high level of the real interest rate. This is largely a reflection of the fact that equity values have remained high in relation to company profits. In interpreting this evidence it is important to bear in mind that company investment decisions can also be influenced by the availability of finance and this might be more severely affected by the affects of high nominal as well as high real interest rates on company cash flows.


(1) In practice the measure is more complex than this, see Young (1992) for details. (2) The short-term real interest rate depicted is defined as the three month interbank rate less future annual inflation defined using the GDP deflator. (3) Similarly, long-term nominal interest rates have been much steadier than short-term rates. Over the period from the first quarter of 1988, short rates have ranged from 8 to 15 per cent, whereas the yield on 20 year gilts has ranged from 9 to 11 1/2 per cent.


Flemming, J.S., L.D.D. Price and S.A. Byers (1976), |The cost of capital, finance and investment', Bank of England Quarterly Bulletin, 16(2), 193-205. Young, Garry (1992), |Industrial Investment and Economic Policy', in Andrew Britton (ed), Industrial Investment as a Policy Objective', National Institute of Economic and Social Research, Report Series No. 3.
COPYRIGHT 1992 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 1992 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:forecast
Author:Pain, Nigel
Publication:National Institute Economic Review
Date:Aug 1, 1992
Previous Article:A forward-looking approach to learning in macroeconomic models.
Next Article:The world economy.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters