The world economy.

Anderton, R. ; Barrell, R. ; Veld, J.W. in't 等

Introduction

Three major features dominate our forecast. First, the collapse of the Japanese stock market has not yet had a major impact on the rest of the world, but it is expected to slow activity down in that country. Second, the continuing public sector deficits in Germany in combination with above target money supply growth have caused the Bundesbank to tighten its monetary policy, and this has been acting as a break on output growth throughout Europe. High short-term and long-term interest rates have had major consequences. They have been a major factor behind the strains that have been re-emerging in the ERM, and the possibility that a realignment has to take place cannot be ignored. Third, high nominal real interest rates in Europe also appear to be a factor behind the weakness of the dollar. The current German-US interest differential is higher now than at any time since the end of the Bretton Woods system, and the continual monetary easing in the US has been a major factor behind the fall in the dollar effective rate.

Growth in the major economies began to slow down at the end of 1989, but the early signs of recession were muted by exceptionally strong growth in Germany and Japan in 1990. We, and most other commentators, expected the recession to be shallow and short lived, with only the English speaking countries amongst the major seven suffering more than a slight slowdown in activity. The downturn in the growth of activity has been longer and deeper than we had anticipated. In 1991 growth in the major seven economies averaged only 0.7 per cent, and OECD industrial production fell by 0.8 per cent. Our forecast, which is summarised in Table 1, is that the recovery in 1992 will be moderate, with growth in the major seven reaching 1.7 per cent.

Our optimism in 1990, and more recently, was based in part on a belief that the financial deregulation of the last decade would leave consumers free to borrow in periods of recession when their incomes were reduced. In the last issue of the Review we discussed the evolution of personal financial wealth and of debt in relation to income, and we demonstrated how the debt income ratio has risen over the last ten years especially in the UK, the US, Japan and France. This increase in debt, which is at least in part the consequence of financial liberalisation has, we believe, been a factor behind the slow recovery in output. Debt service burdens have remained high, in large part because the spread between borrowing and lending rates has widened noticeably in most economies. Consumption has, as a result, remained depressed.

Developments in Japan and in Germany have also extended the period of slow growth, and they have also helped to make growth more synchronised between the major economies. Chart 1 plots the annual rate of growth on a quarterly basis in the major four economies. over the last few years and over the first two years of our forecast. Amongst this group growth slowed first in the US and France, and we are anticipating some recovery in both. More recently growth has slowed in both Germany and Japan, and we expect it to be around the same level in all four of these countries at the end of 1992.

The sources of the recent slowdown are very different between Germany and Japan, and the issues are discussed in greater detail below. Chart 2 plots recent interest-rate developments, and it shows that the Japanese have been cutting interest rates in response to the slowdown in activity, whilst the Germans have not. The boom in activity in Japan had been sustained for some time, and it had been associated with a very strong rise in the stock market. Price to earnings ratios had risen to unprecedented levels, and even though the Japanese tax system in many ways encouraged this, levels of 50 or more were hard to justify. However, the boom appeared to have some self generating elements, and equity prices and property prices rose strongly for some years. The financial system became dependent on a high level of stock market and property prices. However, both stock market and property prices have fallen as the deceleration of growth gained momentum. The fall in asset prices has been a major factor behind the Japanese authorities' decision to cut interest rates. Asset prices have been much firmer in Germany. Chart 3 plots US, Japanese and German equity prices. All three countries have had low inflation over the last decade, and we would expect their equity prices to rise approximately in line. This has not been the case, but the recent change in Japanese equity prices appears to have rectified the discrepancy.

Our model contains a wealth effect in the Japanese consumption equation, and the reduction in personal sector financial wealth that results from the fall in the stock market has a noticeable, depressing, effect on our forecast of Japanese consumption. We have allowed this effect to feed through in full, and as is clear from Table 1, our forecast for Japan is that output growth will as a result remain below capacity for several years. The risks to the liquidity of the financial system that flow from the potential effects of a further stock market fall could make our forecast rather optimistic. Monetary growth is worryingly below its target range, and the banking system is under considerable pressure to achieve adequate levels of capitalisation, and both would worsen if the market falls further.

[TABULAR DATA OMITTED]

Worries about the scale of government debt have been a major factor behind the monetary stance in Germany. German unification has turned out to be an expensive process, and the cost has not been covered by increasing taxes. Output grew very rapidly in 1990, and the combination of high demand and increases in taxes has meant inflation in Germany has risen to levels that have been unacceptable to the Bundesbank. They have also been worried about the burgeoning deficit and its implications for the stock of debt. Although this is not high by international standards (Chart 4 plots the gross debt/income ratio in the US, Germany and Japan) it is growing, and the response has been to raise interest rates. There is also a worry that the money supply target is being overrun (monetary targets in Germany, Italy, France, UK are discussed briefly in Box 1). The sustained high level of German interest rates can be seen as a response to the latter two factors, rather than fears about current inflation. Wage settlements, which are discussed below, have moderated significantly because of the slowdown in activity, but the Bundesbank has not loosened its stance.

Prospects for the US look less bleak than they do for Japan and Germany, despite the poor outturn in the second quarter. Confidence seems to be improving, and the fall in short rates shown in Chart 2 is beginning to be reflected in lower long-term rates. The monetary activism of the Federal Reserve appears to have been successful. We are projecting a return to growth of 2 to 2 1/2 per cent in the near future. A lax fiscal stance has also helped, and we anticipate that the government debt to income ratio will rise for the next few years. The scale of the projected budget deficit (see Table 7 below), along with developments in Japan, will cause a reemergence of matching public sector and balance of payments deficits, although higher levels of domestic saving will keep the current account deficit well below levels seen in the 1980s. Chart 5 plots US public sector, private sector and overseas balances as a per cent of GDP.

Interest rates and exchange rates

Our forecasts for exchange rates are set out in Table 2, and they reflect the interest differentials that we observe between countries. We set our interest-rate forecast in line with market predictions. The first year of our interest-rate profile is based on current three month, six month and twelve month market interest rates. In the longer term we take account of two factors. First, we look at long-term rates, as these can be seen as weighted averages of future short-term rates. Second, we take account of policy commitments such as the decision to form a monetary union in Europe. Our forecast is based on the assumption that the union is formed in 1999(1) and that interest rates in Europe converge before that date.

Our forecast for interest rates is set out in Table 3. Interest rates in the US are well below those in Europe, and if markets' expectations are consistent then we would expect the nominal dollar exchange rate will appreciate continually over the next ten years. This nominal appreciation is also likely to be a real appreciation of the dollar because inflation rates in Europe are not forecast to be as high as that in the US. We calculated in August 1991 that the real dollar exchange rate was probably already below its long-term equilibrium level, or FEER, and it has fallen some 5 per cent since then. The real appreciation in our forecast is in part the result of market pressures that will be pushing the dollar back toward equilibrium.

Our forecast for European exchange rates is based on the same assumption of market efficiency. The dollar exchange rates of the core ERM countries have appreciated by 5 per cent over the last year putting pressure on all the economies.

[TABULAR DATA OMITTED]

German interest rates have been kept high in order that the Bundesbank can achieve German objectives. The French, Italian and British commitment to the ERM has meant that these countries have also had to keep their interest rates high. This has not suited the cyclical position of these countries, and effects on output of high interest rates have been exacerbated by the depreciation of dollar and the consequent loss of competitiveness. If these countries are to form the core of a monetary union then eventually they have to have essentially the same rate of inflation, and they must settle around a sustainable level of competitiveness associated with their equilibrium real exchange rate. This requires that the price level, as well as the rate of inflation, has to adjust. The achievement of these objectives may be costly.

The last major realignment of the ERM took place in 1987. Since then consumer prices in Germany have risen by 15 per cent whilst in Italy and France they have risen by 18 and 35 per cent respectively. Even if the pattern of real exchange rates had been sustainable in 1987, it is not clear that this is still the case. In 1991 we calculated(2) that the Franc and Lira were overvalued by 5 to 10 and 10 to 15 per cent respectively, and that the D-Mark was undervalued by around 10 per cent. These misalignments can be removed either by a re-alignment of the ERM or by a period of, say, 10 years when Italian inflation is 1-1 1/2 per cent below that in Germany. We believe that this could only be achieved by a sustained period of below capacity growth.(3)

The financial markets clearly appreciate this problem. If the exchange rate were seen as irrevocably fixed we would not be seeing three, six and twelve month interest rates some 5 per cent higher in Italy than in Germany. We perceive this differential as indicating that a realignment of the Lira is anticipated, and we have built this into our forecast in the near future. Interest differentials between Italy and Germany have for some time suggested that a realignment has to take place before 1995, and our forecasts have reflected this. We are, in our main forecast, assuming that only the Italians realign, but there are good reasons for persuading the Germans to realign upwards. The note by Anderton, Barrell, In't Veld and Pittis in this Review analyses a systemic adjustment of the whole of the ERM. We would argue that this policy would not lead to such a large loss of credibility for any one country as would result from a unilateral adjustment. Hence it would allow interest rates to fall everywhere in Europe.

World trade and commodity prices

The boom in world activity in the late- 1980s was associated with very rapid world trade growth, and the slow-down in activity has inevitably been associated with a slowdown in the growth of world trade. Trade in all goods between all countries appears to have risen by per cent in 1991, whilst growth in trade in manufactures amongst the OECD countries fell to 3 3/4 in that year. However, both growth rates have been higher than might have been anticipated given the fall in industrial production and the limited growth in output. Chart 6 plots the growth in world trade and in industrial production over the last 25 years. Both growth rates have been falling over the longer term, and their cyclical movements are obviously coherent.

Industrial production is currently well below its trend growth rate whilst world trade growth has fallen, but only to its trend level (both trend rates are indicated on Chart 6). A number of factors have helped support trade growth. The newly industrialising countries in the Far East, along with China have been strong sources of demand, and post-war reconstruction in the Gulf has also been a support to world demand. Imports into the eastern European economies from the west have been stronger than we had anticipated, whilst trade between these countries has collapsed. Much of the finance for these imports has come from increased sales of raw materials. These three factors have between them added about 1-1 1/2 per cent to our forecast world trade growth in 1992, but they are not the only distinguishing features of the current downturn.

Intra OECD trade has been stronger than we might have expected, especially given the slowdown in German import growth in the first half of 1992. Imports grew strongly in the UK over the last year despite the recession in activity, and as soon as the recovery began in the US and Canada, import growth there rose rapidly. Not all of these trends reflect overvaluations of the currency, and moves toward freer trade do appear to be boosting trade volume growth especially in North America.

The downturn in activity has had an effect on commodity prices. The historical relationship between world activity and commodity prices is evident from Table 4. In general, the exceptionally strong GDP growth of 1988 caused commodity prices to rise as a result of increased demand. In contrast, all commodity prices fell sharply in nominal terms during the trough of world activity in 1991, contributing to the general deceleration in world inflation. Prices of metals have fallen sharply over the last two years (particularly aluminium and copper), depressed by low demand in the industrialised countries and by strong metals exports from the former Soviet Union. Metal prices will probably remain subdued this year, as inventories remain high and industrial production is expected to recover only slowly. In the second quarter of 1992, developed countries' food prices fell as wheat and cattle prices declined. Less developed countries' food prices are being pushed down by excess supply particularly in the coffee and cocoa markets, but cocoa prices may receive a boost if agreement is reached at talks of the International Cocoa Organisation in November. Sugar prices have also been declining in response to increased production, particularly by the EC. Declines in cotton and wool prices contributed to the fall in agricultural non-food prices in 1991. Cotton was particularly affected by strong price competition and excessive production but prices have recovered in the first half of this year. The wool price suffered from the ending of Australia's wool floor price support program in February 1991. In NIGEM (our world macro-model), all non-fuel commodity prices in the long run move in line with manufactures export prices. Given our assumption of a rising US exchange rate in the short term, Table 4 shows little overall growth in non-fuel commodity prices in dollar terms for the next two years or so.

[TABULAR DATA OMITTED]

During 1991, oil production declined by 10 per cent in the former Soviet Union, but exports were hardly affected. Over much of the same period OPEC operated at almost full available capacity producing more than 24 million barrels a day. Given the low level of demand due to the world recession, oil inventories rose significantly. The major movements in the oil price last year were dominated by Gulf war effects. The monthly average petroleum spot price rose from $16 a barrel in July 1990 to $33 in October, falling to $17 in February 1991. Hence oil prices fell by around 17 per cent last year. The fragility of the US recovery and the high real interest-rate constraint on growth in Europe have prevented any robust growth in oil prices so far this year. We are forecasting that oil prices will probably be fairly flat during 1992 unless the US decides military action against Iraq is again required in order to carry out UN resolutions.

United States

Disappointing figures for second quarter US GDP suggest that the emerging economic recovery could be stalling. Real GDP grew at 1.4 per cent (seasonally adjusted annual rate). In the first quarter of this year, GDP grew at an annualised rate of 2.9 per cent and this led to optimism about an emerging strong recovery after several years of slow growth and recession. In 1991, real GDP fell by 1.2 per cent, compared with 0.8 per cent growth in 1990 and 2.5 per cent in 1989. Unemployment figures for June were unexpectedly weak and showed a sharp drop in employment. Concerns about a faltering recovery stem largely from disappointing consumption figures. Consumer expenditure fell 0.3 per cent in the second quarter (at a seasonally adjusted annual rate), compared with a rise of 5.1 per cent in the previous quarter. Consumer confidence dropped sharply in July to 61.0 from 72.6 in June, which was a nine month high. If we were observing a robust recovery we should see the index rising to above 100. Although retail sales edged up in the second quarter, the mixed signals of other economic indicators also suggest that any emerging economic recovery will be relatively weak. Industrial output dropped back in June, after four months of increases. Capacity utilisation also fell back. Housing starts fell in June, but new orders for durable goods recovered from their fall in the previous month. The National Association of Purchasing Managers' corporate Index fell back to 52.8 in June, from 56.3 in May which was a four year high. Optimism about a speedy and strong recovery after the first quarter's GDP figures seems to have been premature and current data suggest that the recovery will be slow and modest.

The rate of unemployment rose from 7.5 per cent in May to 7.8 per cent in June, an eight year high. The decline in employment was widespread and more than offset earlier gains in the previous months. The release of these figures prompted the Federal Reserve to cut interest rates to 3 per cent in July in the hope this would aid the recovery. Annual growth in M2 fell back further in June to 1.4 per cent, well below its 2.5-6.5 per cent target range. M3 fell by 0.2 per cent, also below its target range of 1.5 per cent. The recent discount rate cut follows other cuts in December and November and interest rates are now at a 29 year low. The yield on ten year government bonds has also fallen sharply to 7.4 per cent, from 82.8 per cent at the start of this year. This fall is likely to boost consumer spending and investment and will be an important stimulus to economic growth.

Gross private domestic investment grew 12.1 per cent in the second quarter (seasonally adjusted annual rate). Housing investment rose only 8.7 per cent, after a very strong first quarter increase of 20.1 per cent. However, non-residential investment expanded strongly at 13.5 per cent, compared with 3.0 per cent in the previous quarter. For the remainder of this year we expect the record low interest rates to be a significant stimulus for higher business and residential investment. Our forecast for the US is given in Table 5. For 1992, we forecast residential investment to grow by 11.5 per cent and business investment by 3-0 per cent. We expect stronger growth in 1993 and we are forecasting that business investment will grow by as much as 6 per cent in that year.

[TABULAR DATA OMITTED]

In the May 1992 Review, we discussed the high debt levels of US consumers. Although substantial uncertainty surrounds the degree of response of highly indebted consumers to reductions in borrowing costs, we expect the fall in interest rates to produce some recovery of consumers' expenditure this year. For next year we forecast that consumers' spending will grow at around 2.5 per cent, boosted by higher earnings but somewhat depressed by an expected reversal of the recent interest-rates cut. We forecast GDP growth of around 2.0 per cent this year and 3 per cent next year.

We do not expect that exports will give a stimulus to output this year. Exports volumes fell 3.8 per cent in the second quarter compared with a 2.9 per cent rise in the previous quarter, whereas imports strengthened, reflecting the weak recovery earlier this year. The trade deficit has widened again to $35.9 bn in the second quarter, the highest for nearly 2 years. The dollar has fallen significantly over the last six months and the |J' curve effect of this depreciation has worsened the trade balance. We expect a further widening of the trade deficit to just over $90 bn this year, as exports growth is disappointingly low due to depressed world demand. For next year we forecast a further widening of the deficit due to the expected appreciation of the dollar. Our US balance of payments forecast is shown in Table 6. We expect that the services balance will continue to improve, but that the balance on investment income will deteriorate because of both the appreciating dollar and the worsening net overseas assets position that results from cumulating deficits. Transfers have also fallen sharply compared with 1991, as these were last year boosted by transfers from US allies as contribution to Gulf War costs. The current account deficit, which was reduced by the Gulf War payments to 3-7 bn in 1991, is expected to rise again to $25-30 bn for this and next year. Although this is a significant deterioration, it is still less than 1 per cent of GDP and much less than the deficits recorded in the 1980s.

[TABULAR DATA OMITTED]

The fiscal deficit has worsened significantly. In the first nine months of the current fiscal year the deficit widened to $227.7 bn from $178.1 bn in the same period last year, but for the remainder of this fiscal year the deficit is expected to widen further. For the current fiscal year, the deficit is projected to rise to $334 bn. Various tax measures have been introduced to support the faltering economic recovery. This fiscal year will see temporary tax cuts for employees and first-time home buyers and a temporary investment tax allowance. On the expenditure side, a $50 bn cut in defence spending over the next five years is projected, but welfare payments, pensions and expenditure on other social entitlements are expected to grow strongly in the medium term.

In the short term the recession has cut revenues and banking deposit insurance payments have had to be made, and these have contributed to the rising deficit. Rising health and social security expenditures have also worsened the US's fiscal problems. Table 7 sets out our medium-term projection for the US budget deficit in more detail. We use data from the national income and product accounts. These exclude financial transactions, including those related to the rescue of the failed savings and loans institutions. However, in our forecast we make an allowance for higher interest payments on the stock of debt related to the collapse of these institutions. We project a deterioration in the budget deficit to $278 bn on a NIPA basis for the federal government and $250.5 bn for the total public sector. The recent cuts in interest rates will reduce debt-interest payments and aid the process of deficit reduction in the medium term, but unless serious inroads are made into spending plans, significant deficit reduction will be hard to achieve.

Japan

Japanese real GNP rose by just above 1 per cent in the first quarter of this year reflecting a 3/4 per cent increase in consumption and a small upturn in total investment. This strong performance is contrary to the recent general trend over the past few quarters, and the government's target of 3V2per cent growth over the 92/93 fiscal year now seems far too optimistic. Indeed, industrial production fell by almost 2 1/2 per cent in the second quarter and is now around 6 1/2 per cent lower than last year. Industrial production grew by only just over 2.2 per cent last year compared to 4.6 per cent in 1990 and 6.2 per cent in 1989. Although output growth has declined dramatically, it seems that demand growth has fallen even more rapidly as substantial involuntary stockbuilding is taking place. A survey by Nihon Keizai Shimbun in May indicated that companies expect profits will have fallen by 15 per cent over the 91/92 fiscal year. This is consistent with the deceleration in total investment growth, which declined to only 3.5 per cent in 1991 compared with almost 10 per cent in 1990.

In response to this rapid downturn, the Bank of Japan followed the April discount-rate cut of 3/4 per cent with a further 1/2 per cent reduction to 3 1/4 per cent on 27th July. Furthermore, a supplementary budget will be introduced shortly providing a fiscal stimulus of around 2 trillion yen in addition to the April package which |front-loaded' 75 per cent of the 92/93 public works spending into the first half of this fiscal year.

The Japanese stock market (see Chart 3) has continued to fall dramatically even though the central bank has cut the official discount rate five times since July 1991. It seems that companies have now realised that the downturn will be severe, and hence profits will continue to be very weak, and this greater degree of pessimism may explain the recent decline in Japanese equity prices. However, the long-term fall in the stock market, was caused by several structural factors. Low levels of Japanese interest rates in the mid-1980s, combined with the effects of financial liberalisation, boosted asset prices creating a |bubble' effect which became self-perpetuating equity prices grew at an average annual rate of 34 per cent between 1985 and 1987). The price to earnings ratio became very high compared to the dividend ratio, indicating that high corporate profits did not cause the rise in equity prices, suggesting that speculative funds drove the increasing divergence between asset prices and |economic fundamentals'. By 1990 it was widely perceived that the stock market was substantially over-valued and tight monetary policy encouraged readjustment towards a more realistic level. Equity prices subsequently fell by 25 per cent over the two years up until the end of 1991. This downward readjustment has been reinforced in recent months by anxiety concerning the inadequacy of government efforts to revive the economy and worries over the large amounts invested by banks in equity and property markets. Under new rules introduced by the Bank for International Settlements (BIS), all banks must increase their ratio of equity capital to total assets. Japanese banks have previously met this requirement by counting unrealised equity capital gains as part of capital but these are now much diminished. Although it is unlikely that the banking system will not meet the BIS standards, their existence is creating further financial stringency and uncertainty.

One probable effect of the Tokyo stock market crash is a rapid deceleration in capital expenditure that may result in an investment-led growth recession in 1992. Strong Japanese business investment in the 1980s was partly due to the stock market boom encouraging firms to use equities as a major source of finance. This allowed corporations to raise around $350 bn of new capital between 1987 and 1989. Firms raised capital by issuing a combination of stocks and bonds with an option of converting the latter into equities in the future. Expectations of a continually buoyant stock market allowed companies to offer a very small bond yield of only 1 to 3 per cent. The low cost of capital was a major factor behind developments in the late-1980s when the share of investment in GNP increased to the very high level of 25 per cent. The enormous decline in equity prices is likely to discourage capital investment directly by raising the cost of capital to some borrowers significantly. In addition, there will be an indirect negative effect upon investment as revalued equities will constrain bank lending by reducing the capital position of banks. Housing investment, which accounts for a fifth of total investment, has also deteriorated in response to falls in asset prices. The large falls in property prices, aggravated by the recent introduction of a land tax, were the main reason for the 8 per cent decrease in housing investment last year.

Table 8 gives our forecast for Japanese GNP and shows that we expect business investment to show virtually no growth this year and housing investment to fall by another 4 per cent. Some of this downturn will be slightly offset by a small rise in government investment and consumption. Personal sector real consumption will probably remain subdued this year, growing at approximately 2 1/2 per cent, primarily driven by a deceleration in real personal disposable income as activity weakens. The slow growth in personal income follows on from the effects of lower profits on both bonuses and dividend receipts. The large falls in real wealth arising from the stock market crash will also depress consumers' expenditure. However, the rebuilding of wealth stocks by increased saving takes time, with the result that personal expenditure will probably grow quite slowly for the next two years. Some destocking this year will also reduce activity and we do not expect GNP growth to exceed 2 per cent in 1992. A lagged response to the recent easing of monetary policy should encourage a recovery in both business and housing investment next year. However, this will be the only major stimulus to activity and GNP growth may only reach around 3 per cent in 1993. The lagged wealth effects upon consumption suggest that a return to trend GDP growth will not be achieved until the mid-1990s.

[TABULAR DATA OMITTED]

Table 9 shows that the Japanese current account surplus more than doubled from $36 bn in 1990 to $78.1 bn in 1991. The major reason for this improvement was that the visible trade surplus increased from $52 bn to $78 bn over the same period. Although demand was weak in the US during 1991, the growth rate of Japanese export markets was greater than in 1990 largely due to the strong growth of the newly-industrialised countries of South East Asia. In the second quarter of this year, the trade surplus was averaging $8.2 bn per month, a smaller surplus than the $9.3 bn of the first quarter but considerably larger than the $6.1 bn of a year ago. A recovery in the US market, combined with slowing Japanese demand, explains these surpluses. We expect the improvement in visible trade to result in a trade surplus in excess of $90 bn this year. The current balance will improve by a smaller amount, mainly due to a higher services deficit, reaching around 2 1/2 per cent of GNP. This will be reduced slightly next year as imports respond to an increase in activity. However, in the medium term we are now forecasting a persistent Japanese current balance surplus of between 1 and 2 per cent of GNP. The recent dramatic decline in equity prices will encourage individuals to rebuild wealth stocks by saving more. The counterpart to increased saving is a larger current account surplus which will result in part from slower GNP growth.

[TABULAR DATA OMITTED]

Germany

German GNP growth peaked at 4.7 per cent during 1990 as reunification resulted in a sharp increase in demand for West German goods from east Germany. As the initial boost to purchasing power receded, consumption and hence output growth began to decelerate last year. However, after falling for three consecutive quarters, German real GNP grew by almost 2 per cent in the first quarter of this year. A recovery in both consumer spending and particularly investment explains much of this rise. Investment grew by almost 8 per cent in the first quarter, but most of this seems to be a surge in construction investment encouraged by the mild winter weather. Real consumption rose by around 1 per cent in the same quarter but this is not out of line with a lower trend rate of growth. Consumer spending was originally depressed by the July |unification tax' increases but began growing again in the fourth quarter of 1991 fuelled by real income growth from high wages. However, the economy does seem to be generally slowing down and inflationary pressures are abating. Capacity utilisation, consumer and wholesale price inflation and wage settlements are all decelerating whereas unemployment is rising. Annual consumer price inflation slowed to 4.3 per cent in June (from 4.6 per cent in May and the nine year peak of 4.8 per cent in March) and to 3-3 per cent in July as the consumption taxes imposed from last July fell out of the year-on-year calculation. Wholesale prices actually declined in June, when prices were only 1.2 per cent above a year earlier (compared to 2.3 per cent in April).

Wages per person hour rose by 5.3 per cent on an annual basis in May, which is almost a percentage point lower than in March. Although wage settlements have recently been of the order of 5 to 6.5 per cent, they have been considerably below the original wage claims which frequently exceeded 10 per cent. The recent IG Metall wage settlement, consisting of a 5.8 per cent increase this year followed by 3.4 per cent next year, strongly implies that inflationary expectations are declining (although the deal also contains a one hour reduction in the working week).

It therefore seems that inflationary pressures are moderating in response to the high interest-rate policy of the Bundesbank (wholesale price inflation of around 1 per cent implies that German industrialists have to tolerate ex-post real interest rates of approximately 8 per cent). Under these conditions it seems surprising that the Bundesbank thought it necessary to increase the discount rate from 8 to 8 3/4 per cent in July (although the Lombard rate, which has a much greater influence upon money market rates, remained unchanged at 9 3/4 per cent). However, the strategy of the Bundesbank is to set a target growth rate of the money supply consistent with 2 per cent inflation in the medium term. The announced target growth for pan German M3 for 1992 is between 3 1/2 per cent to 5 1/2 per cent whereas this monetary aggregate was growing at an annualised rate of approximately 8 3/4 per cent in June. Rapid domestic private sector credit growth, largely due to government subsidies for construction and investment in eastern Germany combined with demand for shares in state firms sold by the Treuhandanstalt, accounts for much of this recent growth in M3. The July tightening of the discount rate is therefore primarily aimed at reducing credit growth, but the large and increasing public sector deficit is itself putting upward pressure on the cost of borrowing. Slower GNP growth, fewer privatisations and investment subsidies, plus the lagged effects of the earlier tightening of monetary policy should result in a considerable deceleration in credit growth in the second half of this year. However, M3 growth may still not fall within its target range before the year-end and the German authorities will probably wait for a definite downward trend in underlying inflation before reducing interest rates. We do not, therefore, expect a reduction in the Lombard and discount rates until the beginning of next year. Furthermore, any reduction will be small and gradual, probably only amounting to a 1 per cent decline by the end of 1993.

Table 10 contains our forecast for German GNP. After peaking at 4.7 per cent after unification in 1990 and slowing to 3.2 per cent last year, we expect a further decline in GNP growth to approximately 1.5 per cent this year. This is largely due to weaker investment growth and a negative contribution from net exports. The latter is primarily caused by sluggish world demand, whereas high real interest rates are responsible for the subdued capital investment. Consumer spending will probably grow at much the same rate as the 2 1/2 per cent of last year supported by significant real income growth arising from high wage settlements. However, real consumption will recover strongly next year as real incomes continue to benefit from the removal of the |unification tax' in July of this year. In comparison, real GNP should display a slower recovery, growing by around 2 1/2 per cent next year. Again, investment and net exports will remain subdued. The former category will continue to be crowded out by a substantial public deficit and the latter will be dominated by home, rather than overseas, demand.

[TABULAR DATA OMITTED]

Although consumer price inflation is now falling particularly sharply due to the unification tax effect, the inflation rate for the year as a whole will probably be similar to the 3 3/4 per cent of 1991 because of the rapid acceleration of prices in the early part of this year. Furthermore, producer prices will tend to increase as unit labour costs rise in response to slower productivity growth. There should be some respite in underlying inflation pressure next year as unit labour costs grow less rapidly, but consumer price inflation will be pushed upwards by the |harmonisation' increase in VAT from 14 to 15 per cent scheduled for January 1993. We are forecasting that a significant decline in German inflation will be postponed until 1994 and that it will remain at just below 4 per cent in 1993.

Our forecast for German trade is shown in Table 11. Compared to previous forecasts we are now expecting that it will take longer for the German current balance to return to surplus. This is the counterpart to our more pessimistic predicted profile for the public deficit which implies a decline in savings. Our forecast of strong growth in the highly import intensive categories of expenditure, such as consumption, prevents an improvement in the current account in the short term.

[TABULAR DATA OMITTED]

The costs of German re-unification have created a large public sector deficit. This has arisen because of several factors, such as transfer payments to eastern Germany (ie unemployment benefit etc.) and the German Unity Fund' which finances public expenditure in the east (DM20 bn and DM31 bn in 1990 and 1991 respectively). The total public sector deficit for 1991 was approximately DM126 bn compared to DM93 bn in 1990. This is below the target of DM140 bn originally announced in 1990. Robust tax revenues arising from strong growth in west Germany, combined with lower than predicted expenditure in east Germany and the Federal states, partly explain why the deterioration in the deficit was less than expected. However, because expenditure plans have been delayed, not abandoned, a better outturn than forecast for 1991 suggests that 1992 and 1993 will show larger deficits than had been previously forecast. The major difference between the above figures and Table 12, which show the deficit on a National Accounts basis, arises from the inclusion of the social security balance in the latter. In previous years the social security funds registered a substantial surplus approximately DM20 bn in 1991), but the extension of west German pension rights to the east will reduce the net income of the funds by around DM30 bn to produce a deficit of approximately DM10 bn this year. Although the central, regional and local authorities' budget deficits will probably decline slightly in 1992, the deterioration in social security finances, combined with increased expenditure resulting from the expansion in health coverage, will contribute to an increased total deficit this year. It should be noted that these figures exclude the deficit of the Treuhandanstalt, which directs both the privatisation of former state firms and improvements in the infrastructure of east Germany, which will be approximately DM30 bn this year. Next year should see an improvement in the deficit as revenues are boosted by the VAT increase and stronger GNP growth.

[TABULAR DATA OMITTED]

Chart 7 shows that we are not as optimistic as the German authorities in terms of the expected decline in the deficit in the medium term. We believe that expenditure growth cannot be held within the targets stated by the government. At the moment, Germany is following a loose fiscal/tight money policy similar to that of the US administration in the early-1980s. Some expenditure, such as the aforementioned subsidies for investment, increase private sector borrowing and as a result the growth rate of the money supply increases. The Bundesbank will as a result keep interest rates high. Higher interest rates can exacerbate the deficit by increasing debt interest payments and decreasing net government revenues by suppressing GNP growth. Recent attempts to reduce the deficit by raising taxes indicate that this strategy can result in higher wage demands in the short run because of the |tax wedge' effect. Large deficits can, as the US authorities found, be very hard to remove.

France

In the first quarter of this year, French real GDP rose by 1.1 per cent. In the fourth quarter of 1991, zero growth was recorded compared with 1.1 per cent growth in the third quarter. Real GDP rose by just over 1 per cent in 1991, compared with 2 3/4 per cent in 1990. France has been affected by a slowdown in activity like the other G7 countries, but has been able to avoid a serious recession. Its success in bringing inflation under control and the associated gains in competitiveness make it now better placed to take advantage of the recovery in world demand. Domestic demand remained depressed in the first quarter, but exports showed a remarkable upturn. Consumers' expenditure rose by 0.6 per cent, against 0.9 per cent in the final quarter of 1991. Exports grew by 2.9 per cent in the first quarter compared with 1.1 per cent in the previous quarter. Imports grew by 1.4 per cent, compared with a fall of 0.4 per cent in the previous three months. Estimates for the second quarter suggest this trend will continue with a rise in exports of 1.5 per cent and a fall in imports of 0.2 per cent. Any upturn in the French economy this year is likely to come from a stronger trade performance, as France can exploit its improved international trading position.

Our forecast for output and unemployment in the French economy is set out in Table 13. We expect no strong recovery in the near future and GDP is forecast to grow by only 1.5 to 2.0 per cent this and next year. Consumption remains depressed as personal sector financial wealth has fallen significantly and personal disposal income growth remains low.

Business investment continues to decline, while housing investment showed only a very small improvement. Current indicators are not very encouraging. Capacity utilisation fell to 78.5 per cent in the second quarter, compared with 80.1 per cent in the previous quarter, its lowest level since 1985. New company registrations picked up a little in June in most sectors but fell further in construction. Last year total stocks were at their lowest since 1985, and were 12 per cent lower than the previous year. The service sector and industrial sector are both severely hit by the economic malaise. These sectors also showed the sharpest increase in bankruptcies. In addition to these economic indicators, investment intentions in surveys have been revised downward. The monthly INSEE survey in June showed order books quite weak and stocks of finished goods remaining above normal. In the second quarter of 1992 production remained flat in most manufacturing sectors and showed no prospects of any improvement. These were very similar results to the previous month's survey. New car registrations were up 7 per cent in June compared to a year ago and a further boost is expected to follow from the abolition of the top VAT rate in April. A further reduction of the 18.6 per cent VAT rate has been announced for 1993 in order to harmonise rates in the EC.

[TABULAR DATA OMITTED]

Unemployment rose to a record level of 2.92 million in June, an unemployment rate of 10.3 per cent, but it seems that the steady increase has slowed down. Tighter procedures for claiming benefits seem to have helped bring the rise to a halt. The increase in the jobless total has been one of the government's main concerns. The government's tough anti-inflation policies have led to strikes and protests which have caused severe disruptions this summer. Wide ranging measures have been introduced to bring down long-term unemployment and encourage job creation, including substantial training schemes.

The rate of inflation fell back in June to its February level of 3.0 per cent. Inflation averaged 3 per cent in 1990 and 1991. Since June last year, French inflation has been below that in Germany, but the inflation differential has dropped to 1.3 percentage points in June, from 1.5 points in the previous month. We expect inflation to pick up slightly but to remain below German inflation. Growth in M3 was slightly above the target range of 4.6 per cent for 1992 in the Spring, but was only 4.5 per cent (compared to a year earlier) in June. This target range was reduced this year from the 1991 range of 5.7 per cent. In December, the Bank of France followed the increase in the German discount rate and in May the authorities lowered the reserves that commercial banks are required to deposit with the central bank. This reduced commercial banks' costs and helped to cut their prime lending rates by 0.5 per cent.

As inflation has slowed, competitiveness of French exports has improved. The trade account has shown a dramatic improvement so far this year. In the first half year of 1992 it showed a surplus of Fr 16.6 bn, compared with a Fr 7.0 bn deficit in the same period in 1991. Last year, the trade deficit had fallen to Fr 30 bn, from a Fr 50 bn deficit in 1990. The current account showed a surplus of Fr 0.7 bn for the first five months of this year, compared with a Fr 26.6 bn deficit over the same period last year. Our forecast for the current account is given in Table 14. The invisibles balance is expected to remain unchanged, but the sharp improvement in the trade deficit will help to boost the current account and forms an important stimulus to economic recovery. The French international trading position improved significantly last year with a real effective depreciation of more than 7 per cent, but this gain in competitiveness has been partially offset this year by the recent depreciation of the dollar. We expect exports to grow around 6 per cent in 1992. The current account deficit can almost be eliminated this year, but when domestic demand recovers imports are expected to pick up again and with some of the gains in competitiveness eroded, we expect to see a return into deficits of around .7 per cent of GDP for 1993 and 1994.

The fiscal deficit is expected to deteriorate as tax receipts are depressed by the slow economic growth and VAT rates are reduced. The target for the 1992 deficit is Fr 135 bn and for 1993 Fr 150 bn, still only 2 per cent of GDP. The debt stock amounts to only 45 per cent of GDP. These compare very favourably with many other EC countries and are well below the Maastricht criteria. Our forecasts for the deficit and for the public sector debt stock are given in Chart 8.

[TABULAR DATA OMITTED]

Italy

A new government has now been formed in Italy with a 16 seat majority thanks to a fragile partnership between several parties. The new prime minister Sig. Amato is generally perceived as representing movement towards a new regime. As a Treasury minister, Sig. Amato was the author of several detailed plans to reduce the Italian fiscal deficit which failed because of lack of political will. Under this new leadership, a fiscal package raising an extra Lit 30,000 bn has been accorded parliamentary |urgent' status. Tbe package comprises approximately Lit 14 trillion of cuts in expenditure (privatisations account for half of the savings) and Lit 16 trillion in increased revenues. Although the package is seen as ineffective in tackling the long-term structural problems of the deficit (because it mainly consists of one-off measures) it is at least a signal of future intent. Real long-term solutions to the debt problem are unlikely to be implemented until after the anticipated electoral reforms. However as Chart 9 shows we are forecasting that the Italian public sector deficit will begin to fall as taxes are increased, and hence the debt stock as a proportion of GDP will begin to decline.

As Box 2 explains, the financial markets are not certain of the credibility of the new government, and this has resulted in substantial upward pressure on Italian 3 month interest rates. The new government's credibility has been damaged by some of its privatisation policies. During the two years it will take to privatise EFIM (the state holding company) there will be no servicing or repayment of the firm's debt. If the asset sale does not cover EFIM's debts then creditors will receive some payment in government bonds. In addition, privatisation plans for other state enterprises, IRI, ENI, ENEL and INI have been abandoned due to concern over the $10 bn foreign debt of these companies. However, perceptions of the new government's anti-inflation resolve were given a boost by the recent agreement to finally abolish Italy's 47 year old system of indexing wage rises to inflation. The dismantling of the Scala Mobile was seen as vital in achieving the inflation convergence required in order to maintain Italy's exchange-rate parity within the ERM. Chart 10 shows how Italian inflation has decelerated since the Italian commitment to the ERM 'hardened up' in the mid-1980s. This increased commitment began with the gradual elimination of wage indexation combined with fewer and smaller Lira realignments (the last two devaluations were in 1990 and 1987). (The issues surrounding the dismantling of the Scala Mobile and the reform of Italian wage bargaining are discussed in Barrell (1990) and the note on forward looking wages in this Review). Chart 11 shows the relationship between the exchange rate and relative trend unit labour costs. The early-1980s was a period of Lira devaluation which prevented a deterioration in Italy's cost competitiveness even though wages were growing more rapidly than those of her trading partners. In contrast, the stable exchange-rate policy of the late- 1980s, combined with high Italian wage inflation relative to her competitors, resulted in a large decline in cost competitiveness. This has generally resulted in an increasingly negative level of net exports, which has contributed to slow GDP growth, and also produced a recent move into current account deficit. It should be noted that Italian relative export prices have remained far more competitive but the implied decline in export profit margins has depressed export supply.

Financial markets appear to be anticipating a devaluation of the Lira, and we discuss this in Box 2. The precise timing of the devaluation in our forecast is a matter of judgement, but it is clear from market perceptions that some readjustment will be difficult to avoid.

Table 15 gives our forecast for Italian GDP. Italian output growth has been declining since 1988. An overvalued exchange rate sustained by high interest rates, resulting in decelerating investment growth and declining net exports, may be the major factor behind this sluggishness. Total GDP this year will probably display a similar trend and only grow by around 1.5 per cent. The major boost to growth is provided by consumers' expenditure which was particularly strong in the first quarter of 1992. However, weak equity prices and lower bond prices will be reducing wealth. As a result consumption will be more restrained next year as consumers attempt to move back towards their desired wealth-income ratios by increasing savings.

[TABULAR DATA OMITTED]

If the realignment takes place then the improvement in competitiveness will result in increased exports, and decreased imports, and the net trade volumes growth could well boost GDP growth to 2 per cent in 1993. However, a |J' curve effect could actually worsen the current balance/GDP ratio next year but a temporary improvement comes through in 1994. The devaluation could also increase import prices which could cause an increase in inflation in 1993. Slightly higher inflation, in comparison to our previous forecasts, could persist for several years. This reflects in part a change in the perceived anti-inflation resolve of the Italian authorities which, in turn, increases price inflation expectations as there is now an increased perception of further Lira devaluations in the future. A higher interest differential via-a-vis Germany may be the outcome of this loss in credibility as the return on Italian financial assets will embody a higher risk premia, at least in the short term until credibility is regained in the post devaluation period.

Though we expect convergence of Italian and German short-term interest rates in time for EMU circa 1999, we are now assuming the differential will be somewhat higher in the short run compared to previous forecasts. This is justified in terms of a loss in credibility of the Italian authorities in the immediate post-realignment period arising from an increased probability of future devaluations, resulting in a risk premia on Italian financial assets that slowly disappears as credibility is gradually regained. Chart 12 plots French, German and Italian short-term rates over our forecast horizon.

Canada

The Canadian economy is slowly recovering from a recession. National accounts data have recently been revised back to 1988, and it is now suggested that the recession was much deeper than previously estimated. In 1990, real GDP fell by 0.5 per cent and in 1991 by 1.7 per cent. revised down from 0.5 per cent growth and a 1.5 per cent decline). After the UK, Canada suffered the most severe recession of the G7 countries.

In the first quarter of this year, real GDP grew by 1.7 per cent (at an annual rate), the fourth consecutive increase. Domestic confidence remains weak and consumers' expenditure fell 0.4 per cent. Spending on services declined 1.4 per cent and that on non-durables also fell. Highly indebted consumers are reluctant to increase expenditure and the savings ratio remains high. Fear arising from high levels of unemployment may also play a role in keeping consumer confidence low. The main boost to output appears to be coming from exports, which grew by 15.2 per cent in the first quarter, mainly due to a large increase in car exports to the US. Imports fell by 2.9 per cent, reflecting a drop in spending abroad rather then the effects of low domestic demand. The trade surplus with the US improved significantly but the current account deficit remained at more than 4 per cent of GDP, despite the improvement in the first quarter.

Investment growth remains sluggish and stockbuilding fell as a percentage of GDP by 4.5 per cent in the first quarter. Housing investment dropped 0.5 per cent in the first quarter. The weakness of the labour market remains one of Canada's main problems. Employment fell by 0.6 per cent in the second quarter and by 0.5 per cent in the previous three months and the unemployment rate reached 11.6 per cent in June, the highest rate since 1985. Productivity has shown a remarkable increase and improved 2.6 per cent last year. This is believed to reflect a restructuring of production to reduce costs and improve competitiveness relative to Canada's main trading partners.

Inflation has dropped to 1.1 per cent, a 30 year low. High unemployment has eased wage pressure and wage increases averaged 2.9 per cent in the first quarter, continuing their downward trend. With the easing of monetary conditions, interest rates have fallen sharply recently. This will stimulate consumption and investment and we expect to see some recovery of these components of GDP for the remainder of this year. Our forecast for the Canadian economy is given in Table 16. The recovery remains however largely export-driven. We forecast exports to grow by 5.2 per cent this year and the current account deficit to improve to 3.8 per cent of GDP. The recent reduction of interest rates has narrowed the interest-rate differential with the United States and this has helped to stabilise the exchange rate, which has been relatively stable against the US dollar the last few months, after a gradual depreciation during the previous months. The recovery of the US economy will boost Canadian exports but current account deficits are expected to persist, in part because direct investment inflows into Canada will remain at high levels.

[TABULAR DATA OMITTED]

Box 2. Realignment of the Lira

At the time of writing, the Lira is trading near the bottom of its ERM floor at Lit 756 vis-i-vis the D-Mark and a 5 per cent differential exists between Italian and German three month interest rates. The Italian short-term interest rate is currently 15 per cent (the same as the 3 and 6 month rates) with the 12 month rate at 15 per cent. The recent spate of increases in Italian interest rates peaked at over 17 per cent (from 12 per cent in May) probably reflecting a loss in the credibility of the Italian government due to Denmark's |no' to Maastricht combined with the absence of a long-term strategy to reduce the public sector deficit and debt stock. If exchange markets are efficient, then in the absence of controls the interest-rate differential also reflects an expected realignment of the Lira. Since the Danish referendum, financial markets have realised that there is a possibility that EMU may not actually take place. As a result, some of the discipline exerted by EMU,which implied economic convergence and a narrowing of the Italian interest-rate differential, may well disappear. The public sector deficit is a cause for concern, and even with a primary budget balance of zero, the deficit is expected to be around Lit 170 trillion (over 1 1 per cent of GDP) this year pushing the debt stock up to around 1 10 per cent of GDP. These figures allow for the Lit 30 trillion of extra revenue raised in the latest mini-budget, but the financial markets probably perceive these measures as rather ineffective given the scale of the problem. As the government deficit target for 1992 was Lit 128 trillion, financial markets are becoming more convinced that the debt problem is out of control with the implication that more future bond financed borrowing is necessary, putting downward pressure on bond prices which, in turn, increases money market rates. The loss in cost competitiveness over the last 5 years indicates that a Lira devaluation is not inappropriate. The problems of current account balance and the government deficit are linked, as the government deficit will contribute to the overseas deficit by reducing total savings, as long as consumers do not respond to increased budget deficits by saving more to pay expected higher taxes in the future (ie Ricardian equivalence does not hold). However, the market perception may be wrong, and the new government may be able to bring the deficit under control. If they make their resolve clear in the near future then the market's presumption that a realignment is imminent could prove to be wrong.

NOTES

(1) The Danish referendum outcome and the associated discussion have led us to move this date further into the future than we had assumed in May. (2) See Barrell and In't Veld,|FEERS and the path to EMU', National Institute Economic Review, August 1991, No. 137. (3) Work reported in a note in this Review suggests that the sacrifice ratio in Italy (the points rise in unemployment that are required to reduce inflation by 1 per cent a year) is much higher than in the rest of Europe. Hence the relative cost of the two methods of achieving equilibrium differ significantly.

REFERENCES

Barrell, R. (1990), |Has the EMS changed wage and price behaviour in Europe?', National Institute Economic Review, November, No. 134.