The world economy.

Gurney, Andrew ; Barrell, Ray

In the course of the last 2 years economic performance in the major 7 economies has become less synchronised. in 1988 GNP grew by more than 3.5 per cent in all seven economies, with growth rates either at or close to cyclical highs. However for 1991 we expect negative GNP growth for Canada and the United Kingdom, negligible growth in the United States, growth of around 1.5 per cent in France and Italy, and of over 3 per cent in Germany and Japan. Table 1 shows that GNP growth in the major 7 economies is expected to slow to 1.2 per cent in 1991. Chart 1 highlights the different responses among the major 4 economies.

These different responses can be in part related to the policies pursued within each economy. Box A indicates that both the United Kingdom and Canada experienced severe monetary tightening between 1988 and 1989, which contributed to their respective recessions. Germany also experienced a tight monetary policy, but the effects of this were countered by the large fiscal stimulus arising from increased transfers to the east following German Monetary union in 1990. France, Italy and Japan have experienced less severe monetary tightening, accompanied by a mildly expansionary fiscal stance in France, and by a contractionary fiscal stance in Italy and Japan. Japanese fiscal policy was however expansionary in 1990. It is more difficult to relate the recent performance of the US economy to policy developments. Monetary policy has been progressively relaxed since the second half of 1989, and the fiscal stance has changed little. The US recession therefore appears to have been either a result of private sector behaviour or a delayed response to the tightening of policy in 1988. Chart 1 shows that the decline in US GNP growth has occurred progressively since the second half of 1988, although recession did not occur until the second half of 1990. The slowdown has been associated with slower growth of consumers' expenditure, and declines in the level of housing investment, as consumer confidence has been eroded by failing real estate values and fears of a credit crunch.

Table 1 summarises our forecast. GNP growth in the major 7 economies is expected to decline from 2.4 per cent in 1990 to 1.2 per cent in 1991, before rising to 2.9 per cent in 1992. All major 7 economies are expected to grow more slowly in 1991 than in 1990, although for Japan and Germany this means GNP growth of 4 per cent and 3.1 per cent respectively, whilst the United States will have zero growth, and Canada and the United Kingdom negative growth. Japan and Germany are expected to again grow more slowly in 1992, while growth-rates in the other major economies are expected to increase.

Our forecasts represent an unwinding of recent cyclical developments, and their associated policy responses. Charts 2 and 3 show recent developments in interest rates in the major 4 economies. The most striking feature in chart 2 is the sharp decline in United States interest rates, especially since last November, as the monetary authorities have attempted to stimulate a recovery from recession. As yet it is unclear whether this attempt has been successful, and although the preliminary estimate of second quarter GNP showed a small increase, it is not clear that this was led by lower interest rates. Leading indicator and survey evidence suggest that a weak recovery may be underway. Our forecast suggests that the recovery will strengthen in the second half of the year in response to the monetary easing that has already occurred, leaving GNP for the year as a whole unchanged from 1990.

Chart 1 indicates that GNP growth in Japan in the year to 1991Q1 was actually higher than in the two previous quarters. Nevertheless there are signs that domestic demand growth has slowed in response to tighter monetary policy, and this prompted the monetary authorities to reduce interest rates at the beginning of July. The Tokyo stockmarket remains extremely volatile, and there is a risk that financial scandals will affect consumer and investor confidence. The yen appreciated 14 per cent between the second quarters of 1990 and 1991, and this will reduce the contribution of net exports to GNP growth in the latter half of 1991, and in 1992.

There are signs that the German economy may also be growing more slowly. Monetary policy remains very tight, and fiscal policy was tightened through a tax package introduced in July. The tax increases affected consumer price inflation, which rose to 4.5 per cent in July, leading to speculation that monetary policy may be tightened further. However survey evidence and July unemployment figures suggest that the economy is already beginning to slow, and year on year growth rates will fall sharply in the third quarter, when comparisons will no longer straddle German monetary union in July 1990. GNP growth for the year is therefore likely to be around 3 per cent, with a further decline expected in 1992 in response to fiscal tightening and continued tight monetary policy.

France and Italy have benefited from higher demand in Germany, but have lost competitiveness against non-ERM countries as their currencies appreciated with the D-Mark in 1990 (see chart 8 and table 3). Interest-rate differentials with Germany have narrowed in the past two years so that the monetary squeeze has been less severe than in Germany. However without the direct benefit of the German demand stimulus, growth rates have declined. Some recovery is expected in 1992, in part in response to lower interest rates, in part to recovery in the world, and also to a more competitive exchange rate following depreciation this year.

Chart 4 shows recent developments in inflation rates. Inflation in the major 7 countries is expected to fall from 4-4 per cent in 1990 to 4.3 per cent in 1991. However inflation is only expected to fall in the United States, Italy and the United Kingdom. Higher inflation in Japan and Germany arises out of strong pressure of demand, and in Canada from the introduction of the Goods and Services Tax at the start of the year.

Chart 5 shows capacity utilisation rates in the major 4 economies. Capacity utilisation remains above 1985-1987 levels in all four economies except the United States, although it declined in all 4 economies in the first quarter of 1991. it is now much clearer that all these countries have passed their cyclical peaks. Unemployment remains low in Germany and Japan, but has risen in both France and the United States, reducing inflationary pressure emanating from labour markets.

The decline in major 7 GNP growth has led to a decline in world trade growth, table 1. World trade in manufactures is expected to grow by 3.3 per cent in 1991, the lowest growth rate since 1986, and total world trade by 2.6 per cent, the lowest since 1983. OECD industrial production is expected to grow by 0.4 per cent, the lowest growth rate since 1982. OECD industrial production and world trade growth will both recover next year in line with recovery in G7 GNP.

Chart 6 shows recent developments in commodity prices. The price of metals and minerals has fallen by 20 per cent since the third quarter of last year, with the prices of zinc, lead and aluminum falling by over 25 per cent. Metals prices have been more stable since the first quarter of 1991, with a rise in the price of silver of almost 20 per cent, but falls of over 1 0 per cent in the price of both aluminum and zinc. Chart 7 shows that while real metals prices have fallen by 30 per cent since their peak in 1989, they remain 15 per cent higher than in 1986. Metals prices are expected to remain weak in 1991 as mining and production capacity have expanded following the large increase in prices between 1987 and 1989, while slower growth in OECD industrial production has weakened demand.

The price of wheat rose 20 per cent between September 1990 and April 1991, but has since returned to last September's levels. This development accounts for a rise in developed country food prices in the first half of the year. Less developed country food prices have fallen by 10 per cent in the last year with cocoa prices 28 per cent lower than in 1990OQ2, coffee prices 15 per cent lower, and sugar prices 20 per cent lower. Sugar production has been reduced following price falls of 30 per cent in 1990, and this is expected to lead to higher sugar prices in the second half of the year.

Our exchange-rate forecasts are contained in table 3. We continue to assume that exchange rates will in future change in line with interest-rate differentials so that the uncovered arbitrage condition continues to hold. We do not presume that we have more information than the market, and hence we are not currently predicting a major realignment of exchange rates. Although there is some evidence (for instance in Barrell and in't Veld in this Review) that the dollar is overvalued, we do not expect this to be rapidly corrected as US interest rates are below those elsewhere. However, rising US interest rates and higher US inflation will cause the US to depreciate in real terms in the mid-1990s. The yen also appears to us to be misaligned, but its recent undervaluation is expected to be removed only very slowly. We are predicting at least one re-alignment of EMS parities before the forecast move to Monetary Union in 1997. The D-Mark-franc rate, however, changes very little in our forecast, and the realignment is designed to accommodate the effects of the current italian overvaluation and the forecast level of excess inflation in that country. However, we are predicting that convergence will proceed sufficiently for union to be supportable.

United States

US GNP is estimated to have risen by 0.1 per cent in the second quarter of 1991. Whilst the US economy ceased to be in recession, the level of GNP remained 1 per cent lower than in the third quarter of 1990. Domestic demand grew by .7 per cent in the second quarter of 1991, as a result of increases of .9 per cent in consumer spending, .5 per cent in Government expenditure and a lower level of de-stocking. However investment expenditure fell 0.2 per cent, and a deterioration in net exports meant that GNP rose by 0.1 per cent.

The modest recovery in GNP is consistent with the evidence provided by cyclical indicators. The Department of Commerce's leading indicators index has been increasing since February, while the co-incident index stopped falling in April and registered small increases in May and June. The National Association of Purchasing Managers index has also been increasing since February, and registered 51.8 in July. However the Conference Board index of consumer confidence, which rose from 54 in January to 81 in March, declined to 77.7 in July.

The US recession between the third quarter of 1990 and the first quarter of 1991 was caused by a fall of 2.4 per cent in domestic demand. Consumer spending fell by 1.2 percent, investment expenditure by 6.3 per cent and there was a reduction in inventories equivalent to 1 .2 per cent of domestic demand. The weakness of domestic demand can be attributed to the tight monetary stance maintained in 1989 and 1990. The US Federal funds interest rate rose from 6.6 per cent in March 1988 to 9.9 per cent in March 1989, and remained above 8 per cent until November 1990. Since then there has been a substantial easing of monetary policy, with the Federal funds rate falling to 5.5 per cent in early August. Lower interest rates should stimulate the economy into stronger growth in the second half of the year.

The reason that the US monetary authorities maintained a tight monetary stance through most of 1990 was their concern about inflation. The consumers' expenditure deflator increase of 5 per cent in 1990 was the highest inflation rate since 1982. Underlying inflationary pressures were exacerbated by the increase in oil prices in the second half of the year.

The subsequent reversal of the oil price increase has contributed to a reduction of inflation in the first half of this year. By the second quarter of the year CED inflation had fallen to 4.5 per cent. Further declines in inflation are likely in the second half of the year, as producer price inflation has moderated from 3.6 per cent in 1990 to 1.7 per cent in April and May. Inflationary pressures should also ease as a result of the rise in unemployment, which reached 8.7 million in June 1991, an increase of 2.2 million compared to a year previously. The rate of capacity utilisation in manufacturing fell to 77.1 per cent in April, down from 84.6 per cent in the first half of 1989.

Our forecast is presented in table 5. We expect that domestic demand will grow by 1.6 per cent in the second half of the year, with consumers' expenditure growing by 1.4 per cent and a slower rate of destocking. This recovery in the second half of the year will not compensate for the fall in the first half of the year, so that domestic demand for the year as a whole is expected to decline by .6 per cent. However net exports should improve relative to their level in 1990, as weak domestic demand suppresses the growth of imports whilst relatively buoyant demand in the rest of the world maintains the growth of exports. This improvement in external demand will leave GNP little changed for the year as a whole, with growth in the second half of the year nullifying the effects of the earlier recession.

Inflation is expected to continue to fall in the second half of the year, leading to an inflation rate of 4.3 per cent for the year as a whole. As recovery gets underway it will prove more difficult to reduce inflation further. Our forecast assumes that the US monetary authorities will raise interest rates to 6.0 per cent towards the end of the year in order to slow the speed of recovery and keep inflation under control. This assumption is consistent with money market rates at the start of August.

In 1992 GNP is expected to grow by 2.8 per cent, with domestic demand growth of 3.2 per cent. Consumers' expenditure is expected to grow by 2 per cent, with domestic demand growth boosted by investment expenditure growth of 7 per cent, and by a return to positive stockbuilding. Net exports are expected to deteriorate as higher domestic demand leads to a resurgence of imports. Renewed growth in demand will eliminate deflationary pressures, so that the rate of inflation is likely to remain at around 4 per cent. Against this background it is likely that there will be further upward pressure on US interest rates, and we envisage a rise to 7 per cent by the end of the year.

Table 6 shows our forecast for US trade and the balance of payments. The US current account balance was in surplus in the first quarter of the year for the first time since 1982. This surplus was due to transfers from abroad of around $20 billion to help pay for the war with Iraq. (These unrequitted transfers affect the current account but not the level of measured Gross National Product.) Nevertheless there has also been an underlying improvement in the current account balance as imports have declined as a consequence of the US recession, while exports have continued to grow. In January-May the visible trade deficit was $26 billion compared to $42 billion in January-May 1990. For the year as a whole we expect a visible trade deficit of $63 billion and a current account deficit of $20 billion. Renewed growth of the US economy in the second half of this year and in 1992 will result in renewed growth in imports, which will cause the visible trade deficit to deteriorate slightly in 1992. The current account deficit will no longer benefit from Gulf War transfers and consequently will deteriorate more markedly to $65 billion or 1 1 per cent of GNP. This remains a significant improvement from the deficit of 3.6 per cent of GNP recorded in 1987. The correction has come in part from a depreciation of the US dollar over the last four years, and also because the economy is no longer operating above capacity. This reduction in demand has been associated with a reduction in the cyclically adjusted public sector deficit. Past details and our forecast for the US public sector are set out in table 7.

Japan

Japanese GNP grew by 5-7 per cent in 1990, with an apparent slowdown in the second half of the year when GNP was 5.2 per cent higher than a year previously. It appeared likely that GNP would continue to grow more slowly in the first half of 1991. This expectation was confounded by the publication of GNP estimates for the first quarter, which showed GNP 5.9 per cent higher than a year earlier, and 2.7 per cent higher than in 1990Q4. The growth-rate between 1990Q4 and 1991Q1 is equivalent to an annualised growth-rate of 11.2 per cent. This estimate almost certainly overstates the current strength of the Japanese economy, but nonetheless indicates that earlier forecasts of a significant slowdown were somewhat premature.

In 1990 GNP growth of 5.7 per cent was composed of domestic demand growth of 5.8 per cent and a deterioration in net exports equivalent to 0.1 per cent of GNP. The 1991Q1 figures showed domestic demand growth of 4.5 per cent compared to a year earlier, and an improvement in net exports equivalent to 1.4 per cent of GNP. Although both elements of demand grew by more than was expected, the improvement in net exports is especially striking. It appears that the Japanese traded sector is exploiting the gain in competitiveness derived from the depreciation of the yen. In the first quarter of 1991 the yen was 1 1 per cent lower than in the first quarter of 1989, in spite of an appreciation of 1 1 per cent since the second quarter of 1990.

Japanese monetary policy has been tight for much of the last year. The discount rate was raised from 5.25 per cent to 6 per cent on 30th August 1990, and was held at that level until 1st July 1991, when it was reduced to 5.5 per cent. The policy has aimed to reduce the growth of domestic demand, which increased by over 20 per cent between 1987 and 1990, contributing to increased capacity utilisation, and inflationary pressures emanating from a tight labour market. Unemployment has remained at 2.1 per cent of the labour force since the first quarter of 1990, and the ratio of unfilled vacancies to job seekers stood at 1.44 in May 1991, close to its 16 year high of 1.47 reached in June 1990. The annual Spring wage negotiations produced wage increases averaging 5.7 per cent in 1991, which while slightly down from 1990 will contribute to domestically generated inflation.

Consumer price inflation peaked at 4.5 per cent in January 1991, and averaged 3.6 per cent in the second quarter, as against 4.2 per cent in the first quarter. Domestically generated inflation remains high, but imported inflation has declined as a result of the fall in the oil price in early 1991 and the 14 per cent appreciation of the yen between the second quarters of 1990 and 1991. The first quarter GNP figures also indicate that domestic demand is growing more slowly. Consumers' expenditure was only 2.2 per cent higher than a year previously, and although investment expenditure was 10.4 per cent higher than a year previously, it was growing more slowly than in 1990, when it averaged 12.7 per cent.

The decision to reduce the discount rate by .5 per cent on 1st July can be seen as an indication that the Japanese monetary authorities felt that the economy had slowed sufficiently to avoid the possibility of overheating, and that consequently some monetary easing was justified. As well as slower growth in consumers' expenditure, housing investment has declined substantially, with a fall of 19 per cent in housing starts in March-May compared to a year previously. Survey evidence indicates that business investment is expected to grow more slowly this year. A Bank of Japan survey in May showed planned increases of 7 per cent, well below outturns for the last three years, and also the lowest survey figure for several years.

Our forecast presented in table 8, shows GNP growth of 4.0 per cent for 1991. We expect that the economy will continue to grow more slowly in the second half of the year as domestic demand growth is restrained by high interest rates and the growth in external demand is restrained by the appreciation of the yen over the last year. Domestic demand is forecast to grow by 3.3 per cent, with growth of consumers' expenditure slowing to 2.2 per cent, growth in investment slowing to 6.8 per cent and a decline in the level of stockbuilding. Net exports are forecast to add 0.7 percentage points to GNP growth, as a consequence of slower growth in Japan combined with a lagged response to the depreciation of the yen since 1989.

Inflation should continue to decline in the second half of the year, resulting in an average of 3.3 per cent for the year as a whole. If inflation does continue to fall in this way it should prove possible for the Bank of Japan to reduce interest rates again, although this is unlikely to occur until the fourth quarter. Unemployment is expected to edge up a little, but the increase will be small, and labour markets are likely to remain tight.

We expect that in 1992 the growth rate of domestic demand will rise to 4.0 per cent, but that the growth rate of GNP will decline to 3.6 per cent, as net exports deteriorate. The recovery in domestic demand growth will be caused mainly by higher growth in consumers' expenditure. This will result from higher growth in real personal disposable income as the rate of inflation subsides, and from a lower level of interest rates than in 1991. The deterioration in net exports will arise from a higher growth of imports due to higher growth of domestic demand, and from an erosion of competitiveness caused by the appreciation of the yen through both 1991 and 1992.

Chart 9 shows the evolution of the Japanese current account balance over the 1980s, in relation to the level of the Japanese real exchange rate, and table 9 shows our forecast for Japanese trade and the balance of payments. The chart shows that the Japanese real exchange rate was relatively stable in the early 1980s, at a level which enabled Japan to generate increasing current account surpluses. Between 1985 and 1988 the real exchange rate appreciated by 44 per cent, leading to a decline in the current balance surplus from 4.2 per cent in 1986 to 1 .2 per cent in 1990. However the real exchange rate has depreciated since 1988, making further erosion of the surplus less likely. In 1991 we expect that the current account surplus will increase to 2.2 per cent of GNP as the traded sector benefits from the real depreciation in 1990, but that thereafter the surplus will gradually erode, as the real exchange rate is expected to appreciate again in 1991 and 1992.

Germany

Although it is over a year since German economic and monetary union, and almost a year since political unification, the German economy is still absorbing the short-term consequences of these two events. The economy of the western Lander has had to absorb increased demand for the products, at a time when it was already close to capacity output, while much of the economy of the eastern Lander has been unable to compete with western goods, with the result that the east has experienced a severe recession. The partial integration of the two German economies is reflected in the gathering of economic data. Our analysis and forecast focusses on developments within the western economy, although data for trade and the current balance is now only available for the whole of Germany.

West German GNP grew by 5.2 per cent in the year to the first quarter of 1991, compared with growth of 5.4 per cent in the second half of 1990. Domestic demand grew by 5 per cent in the first quarter of 1991 compared with 5.4 per cent in the second half of 1990. it could therefore be argued that demand growth is abating, although the effect is small. Comparisons with a year ago do not give a true picture of underlying demand growth, since they compare demand before and after monetary union."

There are, however, some signs that the west German economy has now absorbed the initial demand shock arising from German monetary union. The rate of capacity utilisation in manufacturing fell in both 1990Q4 and 1991Q1, and is now at its lowest level since 1989Q1. Seasonally adjusted unemployment, which fell from 7.3 per cent of the labour force in June 1990 to 6.2 per cent in January 1991, remained little changed at 6.3 per cent in June 1991. New orders to manufacturing were lower in February-April 1991 than in any 3-month period since May-July 1990, before monetary union. Survey indices of the business climate and the future tendency of production also indicate that demand is growing more slowly.

German monetary union has led to a higher level of inflation as the west German economy has had to meet increased demand at close to capacity output. Consumer price inflation rose from 2.5 per cent in the first half of 1990, prior to monetary union, to 3.1 per cent in the second quarter of 1991, and 4.5 per cent in July 1991. The July increase included the effects of increases in indirect taxes, including taxes on petrol, which were introduced in order to fund some of the costs of unification. Inflationary pressures have also become apparent in the labour market. The annual increase in hourly wage rates rose from 3.2 per cent in the first half of 1990 to 6.9 per cent in June 1991. These increases have occurred despite the increase in German labour supply following unification.

The integration of the German labour market is occurring slowly. East German GNP is estimated to have fallen by around 30 per cent following unification, with the result that unemployment has risen strongly. In July 1991 east German unemployment reached 11 million, or 12 per cent of the workforce. Unemployment would be substantially higher if the authorities had not extended their part-time working support scheme. However, the labour market is adjusting, albeit slowly. Around .5 million east Germans are estimated to be working in western Germany. The reconstruction of east Germany has to date been slow. However it has received a welcome impetus from the decision to relocate the Bundestag in Berlin, which will transfer considerable spending power into the heart of the eastern Lander.

German monetary policy remains tight. Three-month interest rates averaged 8.4 per cent in 1990, up from 4.3 per cent in 1988 and they rose further to 9.1 per cent in the first half of 1991, following increases in the Lombard and discount rates on 1st February. There has been a widespread expectation that the Bundesbank will tighten monetary policy further. Our forecast includes an increase in the short-term rates to 9.5 per cent in the fourth quarter of 1991, in line with money market expectations. However given that there are signs that economic growth may be slowing, the Bundesbank may decide to maintain interest rates at current levels.

Table 10 presents our forecast of west German GNP. National accounts statistics for the eastern Lander now exist but the disruptions to the eastern economy means that the quality of this data is unknown, and for the time being it remains more useful to focus on developments within the western Lander, which will continue to influence the setting of German macroeconomic policy. GNP in the western Lander is expected to grow by 3.1 per cent in 1991. This represents a decline from 4.7 per cent in 1990, both because the initial stimulus arising out of monetary union has subsided as the east has experienced recession, and because monetary policy has continued to bear down on the growth of German demand. Nevertheless growth for the year will remain high, with consumers' expenditure growth of 3.3 per cent, higher than in 1988 or 1989, and investment growth of 9 per cent, the second highest rate recorded in the last twenty years. High growth in domestic demand, combined with weak external demand will produce a contribution to GNP growth from net exports of 0.6 per cent of GNP, compared to 0.5 per cent in 1990 and 11 per cent in 1989.

Inflation is forecast to increase to 3.8 per cent for the year as a whole. Inflation in the second half of the year is expected to be higher than in the first half, as wage and indirect tax increases lead to higher prices. Monetary policy will remain tight, and the Bundesbank will continue to press for a reduction in the fiscal deficit. These policy measures will exert downward pressure on both demand and inflation. Domestic demand growth is expected to slow to 2.9 per cent in 1992, and GNP growth to 2.5 per cent. Inflation may stay above 4 per cent for the year as a whole. The Government has announced that VAT will be increased in January 1993, which will cause inflation to increase temporarily in that year.

Table 11 shows our forecast of German trade and balance of payments. German imports have grown rapidly since monetary union, while export growth has slowed in response to weaker demand abroad. in the first half of 1991 the German trade surplus totalled DM5 billion, against DM65 billion in the first half of 1990. The current account has also been affected by payments associated with the Gulf War of DM10.4 billion. These contributed to a current account deficit of DM20 billion in the first half of the year compared to a surplus of DM50 billion in the first half of last year.

The current account is expected to return to surplus in the second half of the year, as the special factor of Gulf War payments no longer applies, and as German growth slows and external demand recovers. For the year as a whole the current account may register a deficit of $1 billion. The more favourable current account developments will continue in 1992, leading to a current surplus of $7 billion or .5 per cent of West German GNP.

Table 12 shows our forecast of the German public sector deficit, which emphasises the continuing disequilibrium in the German economy following monetary union. The public sector deficit for 1991 is expected to be DM156 billion, or 6 per cent of west German GNP. However this figure excludes debts incurred by the railways, post and the Treuhand agency, engaged in privatising east German industry. The Finance Ministry estimate that transfers to east Germany will amount to DM93 billion this year, DM109 billion in 1992, DM105 billion in 1993 and DM113 billion in 1994. The public sector deficit will remain at around DM150 billion in 1992, but is expected to decline to DM108 billion in 1993 when recovery should be underway in the eastern Lander, and when revenues will be enhanced by the projected increase in VAT in January 1993.

France

French GDP grew by 0.25 per cent between the fourth quarter of 1990 and the first quarter of 1991, after a fall of 0.1 per cent between the third and fourth quarters of 1990. The French economy has therefore avoided recession, according to the widely used technical definition of a recession as two consecutive quarters of falling output. Nevertheless GDP growth remains weak, with GDP only 1.2 per cent higher than in the first quarter of 1990.

The weakness in GDP growth can be largely attributed to weak domestic demand growth. Consumers' expenditure in the first quarter of 1991 was 1.9 per cent higher than a year previously, but private sector investment was only .6 per cent higher than a year earlier. Net exports reduced GDP growth by .3 per cent, although this represented an improvement on 1990, reflecting slower import growth in response to slower domestic demand growth.

The slower growth of demand has been reflected in failing capacity utilisation and rising unemployment. Capacity utilisation fell to 82.9 per cent in the first quarter of 1991, having reached a peak of 86.3 per cent in the second quarter of 1990. Unemployment rose to 2.7 million in May 1991, an increase of 200,000 since May 1990. These developments should exert downward pressure on labour costs. Hourly wage-rates in industry rose by 5 per cent in the year to March 1991, compared to 4.5 per cent in 1990. Low pressure of demand has also been reflected in continuing low inflation. Consumer price inflation registered 3.3 per cent in June 1991, and was lower than in Germany for the first time since 1973.

Survey evidence points to stronger growth in the French economy in the second half of the year. In May those expecting manufacturing production to increase outnumbered those expecting it to fall for the first time since July 1990, and whilst there remained a negative balance in response to a question on the prospects for the industrial sector, the degree of pessimism has declined since February. The leading indicators index has also been rising since February. The more optimistic outlook has been partly conditioned by a relaxation in monetary policy, with 3-month interest rates failing from 10.3 per cent in January 1991 to 9.2 per cent in May.

Our forecast, presented in table 13, is for GDP growth of 1.5 per cent in 1991. Domestic demand is expected to grow by 1.3 per cent, with a fall in investment expenditure and consumers' expenditure growth of 2.1 per cent. Net exports should add 0.2 per cent to GNP growth. Import growth will be restrained by weak domestic demand growth, while exports will benefit from strong growth in Germany, although other export markets are either in recession or growing more slowly than in 1989 and 1990.

GNP growth is expected to pick up in the second half of this year and into 1992. Domestic demand growth should respond to the reduction of interest rates in the second quarter of this year, and external demand will benefit both from a recovery in demand in French export markets and from the recent depreciation of the franc against both the US dollar and the yen. The stance of monetary policy is unlikely to change significantly, although a small increase in French interest rates may occur if Germany interest rates are increased. In our forecast we assume that French rates rise by 0.2 percentage points in the fourth quarter, but this is unlikely to restrain the recovery in economic activity.

The reduction in capacity utilisation and increase in unemployment that have arisen from slower growth in the second half of 1990 and first half of 1991 should enable inflation to remain at around 3 per cent in both 1991 and 1992. French inflation will consequently remain below German inflation, and this should provided scope for further narrowing of the interest-rate differential between the two countries.

The French Government expects that tax revenues for 1991 will be around FF40 billion below their budget projection, and that consequently the budget deficit will exceed the target of FF80 billion. The target of FF80 billion has been retained for 1992, and since the Government does not wish to increase the tax burden, this is expected to entail lower growth in public spending. Spending on education and research remains a priority, but defence spending is likely to be cut.

Our forecast for French trade and the balance of payments is shown in table 14. The trade deficit for 1991 Q2 was FF7.6 billion, down from FF1 3.7 billion in 1991 Q1. Import growth has slowed in response to slower growth in demand. Export markets are expected to grow by 4.8 per cent in 1991, with high growth in Germany compensating for lower growth in other markets. In the latter half of the year the trade balance will improve as a result of improved competitiveness following the franc's depreciation against the US dollar and the yen. The visible balance should therefore improve compared to 1990, and this will also contribute to an improvement in the current balance, which is expected to register a deficit of 7.8 billion in 1991 and 4.8 billion in 1992.

Italy

Italian GDP was unchanged between the third and fourth quarters of 1990. This meant that growth over the preceding year slowed from 1.8 per cent in the third quarter to 10 per cent in the fourth quarter. This deterioration was mainly due to a deterioration in net exports, since domestic demand growth over the preceding year remained at 2.8 per cent in the fourth quarter, compared to 2.9 per cent in the third quarter. The growth-rates of domestic demand and GNP were both lower in 1990 than in 1989, as the economy continued to respond to the restrictive monetary and fiscal stance adopted in 1989.

Recent indicators suggest that the economy has continued to grow slowly in the first half of 1991, but may begin to grow more rapidly in the second half of the year. The leading indicator index increased from 112.1 in January 1991 to 113.7 in May. Industrial production increased by .3 per cent between the fourth quarter of 1990 and the first quarter of 1991 although it declined in April for the fourth consecutive month. Business survey evidence shows that manufacturers have become more optimistic in recent months, although it remains the case that more respondents expect economic prospects to deteriorate than to improve. Employment in industry increased in both the fourth quarter of 1990 (by .7 per cent) and the first quarter of 1991 (11 per cent), although unemployment also increased in the first two quarters of 1991.

Our forecast is presented in table 15. GNP growth is expected to slow to 1.4 per cent in 1991. This will be the lowest annual growth-rate since 1983. Lower growth in 1991 is due to lower growth of domestic demand, since the change in net exports as a proportion of GDP is expected to improve from 0.9 per cent in 1990 to -0.4 per cent in 1991. Italian interest rates rose from 114 per cent in the third quarter of 1990 to 13.1 per cent in the first quarter of 1991, and although they were reduced to 12.3 per cent in the second quarter, it is expected that the higher level of interest rates will restrict domestic demand growth in 1991. Lower domestic growth will also restrict the growth of imports, and net exports will also benefit from strong growth in German demand and from the depreciation of the lira since the middle of the 1990. In July 1991 the effective rate for the lira was 3.2 per cent lower than in August 1990. These developments will also lead to a reduction in the italian current account deficit to $3 billion, or 0.3 per cent of GDP.

Inflation is expected to fall to 5.8 per cent in 1991 from 6.2 per cent in 1990. In the second half of last year inflation rose to 6.6 per cent, but had fallen to 6.4 per cent in February 1991. Wholesale price inflation registered 0.1 per cent in April 1991, reflecting the weakening of demand pressures. This is also evident in the decline of capacity utilisation from 80.8 per cent in the second quarter of 1990 to 77.2 per cent in the first quarter of 1991, and in the rise in unemployment since the fourth quarter of 1990. Continuing progress in reducing inflation will however require a reduction in wage inflation especially in the non-traded goods sector and in the public sector. Hourly wage rates in industry increased by 7.8 per cent in the year to February 1991.

Our forecast shows that GNP growth will increase to 2.9 per cent in 1992. Domestic demand growth is expected to increase to 2.8 per cent, as consumers' expenditure responds to an increase in real earnings caused by wage inflation declining less rapidly than price inflation. Net exports are also expected to improve as a result of recovery in external demand and continuing depreciation of the lira. The current account balance is expected to deteriorate slightly as a proportion of GDP. Consumer price inflation may fall below 5 per cent for the first year since 1969. However, inflation is still expected to be higher than in other ERM member countries, and this will make it difficult to maintain the value of the lira within the ERM. In our forecast the lira is devalued within the ERM in the mid-1990s.

Canada

Canadian GDP declined for the fourth consecutive quarter in the first quarter of 1991, to record a cumulative decrease of 2.8 per cent over the preceding year. Domestic demand also fell by 2.8 per cent in the year to the first quarter, with net exports unchanged as a proportion of GNP. Consumers' expenditure and investment continued to decline in the first quarter, but there was some rebuilding of inventories, for the first occasion since the fourth quarter of 1989.

The severity of the Canadian recession can be attributed to increases in interest rates from 8.5 per cent in the first quarter of 1988 to 13.6 per cent in the second quarter of 1990, and to the weakening of demand in the United States. These developments were exacerbated by the effects of the free trade agreement between Canada and the United States, which has resulted in some relocation of production from Canada to the States. However, as interest rates have fallen to 9.7 per cent, and prospects for the United States economy have improved, there are grounds for expecting some recovery in the Canadian economy in the second half of the year. The leading indicators index rose in April, although it remained below its February level. The reduction in interest rates appears to have already stimulated activity in the housing market, with seasonally adjusted housing starts registering 186,000 in July 1991, an increase of 85,000 on the average for January-March. Survey evidence shows increased optimism in manufacturing industry. In the second quarter of 1 991 more respondents expected that the production would increase in the future than that it would fall, whereas in the three previous quarters the balance of respondents had expected that production would fall.

Our forecast, shown in table 16, indicates that the Canadian economy will move out of recession in either the second or third quarter of the year. However recovery in the second half of 1991 is likely to be weak, so that GDP for the year as a whole will be 0.7 per cent below its level in 1990. Domestic demand is expected to be .5 per cent lower in 1991. Increases in stockbuilding, housing investment and consumers' expenditure are expected to occur by the third quarter, but the recovery in business investment may not occur until the fourth quarter, as capacity utilisation has fallen considerably in the course of the recession. in the first quarter of 1991 capacity utilisation registered 73 per cent, compared to over 85 per cent in 1987 and 1988. Despite the decline of domestic demand net exports are expected to deteriorate as a proportion of GNP in 1991. This is because recession in the United States, appreciation of the Canadian dollar against the US dollar, and the higher rate of Canadian inflation will reduce Canadian exports. The current account deficit is however expected to decline to US$16.2 billion, or 2.7 per cent of GDP.

The introduction of the Goods and Services Tax (GST) at the start of 1991 has caused a step change in the price level, and consequently raised the rate of inflation in 1991. The GST was introduced in order to control the federal budget deficit. However the Canadian recession has reduced Government revenue, and causing the budget deficit in the fiscal year commencing in April 1991 to increase to C$30.5 billion. The February 1991 Budget reduced programme spending, increased some excise levies and raised unemployment-insurance premiums in an attempt to maintain the deficit at C$30.5 billion for the fiscal year 1991/92.

By 1992 we expect the recovery in GDP to be well under way, with housing investment and consumers' expenditure growing strongly, although business investment is expected to recover more slowly. The recovery in the Canadian economy will be supported by the recovery in the United States. Inflation is expected to fall to 3.6 per cent, as the recession will have reduced price pressures within the economy, and because the inflation figures will no longer include price increases caused by the introduction of GST.

Developments in Eastern Europe are, at least in the short term, shrouded in uncertainty. Even in EastGermany the development of a new set of market mechanisms has been slow, and the scale of capital inflows has been rather disappointing. However as systems of property rights and of contract change then we would expect large scale capital inflows to take place, and these are likely to be matched by current account deficits. In long run equilibrium real wages in East Europe are likely to rise toward western levels, and their subsequent evolution over time will be affected by the rate of population growth and by the other factors considered by neoclassical growth theories. These factors include the scale and nature of technical progress. However, in the short to medium term East Europe has both a large pool of cheap and reasonably educated labour and a shortage of capital.

Net capital outflows from the major western economies are likely to help augment the level of domestic saving in the East. These capital outflows will be responding to the availability of higher rates of return in the East, and this will have an effect on the rate of return in the west. Investors will have a tendency to equalise the rate of return on assets in all locations, and if returns rise outside the major economies then resources will be bid away from them. This will push up the domestic rate of return, and saving will be increased. We need to be able to judge the scale of the potential change in the real interest rate in order to analyse the effect of capital flows to the East on western economies.

Real interest rates depend upon the desire to save, upon the absorption of resources by the government, and on the availability of productive investment opportunities at both home and abroad. if returns are high overseas then capital will flow out, the current account will be in surplus and the real interest rate will be high. Chart Al plots the relationship between real interest rates, current balances and government deficits in the G7.

The average real interest rate over this period was 2.8 per cent, the average G7 current balance deficit was 0.2 per cent and the average government deficit was 2.6 per cent. We would expect that a rise in investment opportunities outside the G7 would cause real interest rates to rise, and we have assumed that a rise in the G7 current balance surplus of 1 per cent of GNP would raise real interest rates by 1/2 a per cent. We have used this relationship to analyse the effects of a permanent 50 per cent increase in imports into eastern Europe financed largely by capital inflows.

We have assumed that exchange rates do not vary, and that nominal interest rates rise by the same amount everywhere. Real interest rates rise on average by less than a half a per cent, but the effects vary between countries because the effects on inflation and output differ. Output rises slightly in most countries, but higher interest rates reduce it in Japan. Table Al summarises the effects after four years.

Output changes much less than does investment in each of the major economies. The rise in real short-term interest rates affects long interest rates, and this has particularly strong effects in Japan, Germany and the UK. Current balances improve everywhere, but especially in Germany. Capital flows to the East would be substantial. Table A2 gives the effect of the shock on world trade and on East European exports and current balances after four years.

The commodity price equations on GEM have been re-estimated. We tested down from a general specification of the form:

 where WDPCOM = logarithm of commodity price
 WDPXG = logarithm of world export prices
 M7GNP = logarithm of major 7 GNP
 T = time trend

This specification gives a long-run solution for the real commodity price. We would expect that technical change in the production processes in agriculture and mineral extraction should put downward pressure on real commodity prices, and that this pressure would not be offset by increases in demand. As both technology and demand evolve at relatively constant rates in the long run we would expect our equations to contain trends. The time trend captures any trend in the real commodity price. The term in AM7 GNP provides a short-term response to changes in the growth-rate of major 7 economies.

Table 1 specifies the four commodity price equations. All equations were initially estimated over 1970Q1-1990Q3, and tested for econometric specification and structural stability. The equation for metals and minerals proved satisfactory on both counts, but the three other equations failed the structural stability tests, and have therefore been estimated over shorter time periods. The equations shown in Table 1 all reject tests for serial correlation, mis-specified functional form and heteroskedasticity at 5 per cent significance levels. The equation for agricultural non-food does not reject the test for nonnormality of its residuals. All the equations appear to be structurally stable.

The effects of these new equations on the simulation properties of the model are shown in charts 1 and 2. In the case of the US appreciation the real prices of food and agricultural non-food initially rise, as the dollar price of world exports falls more rapidly than the $ price of these commodities. However by year 4 of the simulation all 4 real commodity prices have returned to their base values. Our previous equations which were discussed in the August 1988 Review-did display some dollar inhomogeneities and these have been removed. Chart 2 shows that the oil price increase has a small effect on real commodity prices, with an initial fall of less than 1 per cent that is largely reversed by year 4 of the simulation. Our old equations embodied permanent effects from oil prices onto all commodity prices. These reflected the experience of the early 1970s, and the relationship between oil prices and commodity prices now a ears much less strong.

We have recently re-estimated the unemployment equations on GEM for the G7 economies. Our new equations have been derived from a common functional form. The equations were obtained by testing down from the following specification:

 where
 U = Unemployment rate
 y = logarithm of GDP
 (GNP for US, Japan, Germany)

This specification has the following long-run solution:

g = steady-state rate of growth ( per cent per quarter)

They can be described as simple Okun's Law relationships that are designed to pick up the cycle in unemployment that follows from cycles in the rate of growth of the economy. The lags are very long, and unemployment cycles can lag well behind those in output. However, these equations suggest that if growth stabilises and does not cycle the unemployment will also stabilise.

The aim was to obtain a well-specified, stable equation that could explain 1970Q1-1989Q4. This could not be achieved in every case, and consequently some equations have been estimated over shorter time periods.

Table 1 lists the final equations for each country. The equations for Japan, France and the United Kingdom have been estimated from 1970Q1, the equation for Germany from 1974Q1, the equations for Canada and the United States from 1976Q1 and the equation for Italy from 1978Q1. In all cases apart from Italy the coefficients appear well-determined. Although the coefficients in the italian equation are not significant, the equation satisfies tests for white noise errors and structural stability, and also has desirable simulation properties.

Apart from Canada all equations satisfy tests for white noise errors at the 5 per cent significance level. All the equations pass tests for the restrictions imposed from the general model. Apart from the United Kingdom all the restrictions imposed were setting coefficients on dynamic terms to zero. In the United Kingdom the coefficient on lagged unemployment level was poorly determined, and it proved acceptable to impose a value of 0-01, which was more negative than its unrestricted estimate. None of the equations revealed significant structural instability as measured by the predictive failure and Chow tests, although the equations for Canada and the United States reject the null hypothesis of constant variance at 5 per cent significance levels.

In four of the equations it proved possible to restrict the coefficients on lagged changes in unemployment to zero. The exceptions were Germany, Canada and the United Kingdom, where the coefficients on the first lag proved significant with values of .55, .49 and .89 respectively. The coefficient of .89 in the United Kingdom is especially high and may prove problematic if it is incorporated in the model.

The new equations contain a long-run solution for the unemployment rate in terms of the rate of growth. Table 2a shows the average growth-rate and average unemployment rate over 1970-89 for each country, together with the long-run solution of the unemployment equations for the average growth-rate. The long-run solutions are reasonably close to the historical averages for the United States, Japan, UK and Canada, but over-predict for Germany, France and Italy. Table 2b shows the same comparison for the period 1980-89. The long-run solutions for Germany, France and Italy still over-predict, but by less than in Table 2a, while the solution for the United Kingdom under predicts the period average. However mean lags in these equations are very long, implying considerable hysteresis in the unemployment rate, and hence a slow convergence on the implied long-run solution.

In both forecasts and simulations the steady-state growth rate is determined by other equations in our model. The model has a tendency to settle down to a stable rate of growth, and this will be associated with a stable level of unemployment. The interaction of demand equations and price equations ensure that if demand is growing too rapidly then inflation rises which crowds out excess demand growth. Similarly if demand is growing slowly, inflation falls, and this generates extra demand through the effects of lower prices in raising real wealth. Table 3 shows the rate of growth and the level of unemployment generated from extending our forecast until the year 2005. The table is remarkable both for the diversity of levels of unemployment in various countries and from the tendency of the level of unemployment to stabilise. The former feature flows largely for institutional differences, the latter from the fact that our forecast, which extends to 2005, does not contain cycles.

BOX A. MONETARY AND FISCAL POLICY DEVELOPMENTS IN THE MAJOR 7 ECONOMIES Table Al provides indicators of monetary and fiscal policy stance for the major 7 economies for the last three years. We have chosen to assess monetary policy stance in terms of the ex-post real interest rates, and fiscal policy stance by the ex-post general government financial balance. Both measures show actual outcomes of policy decisions rather than expected outcomes, and hence unanticipated developments in the economy will have resulted in the measured policy stance being tighter or looser than the intended policy stance. These measures do not therefore reflect just the discretionary element of monetary and fiscal policy, but indicate how policies combined with contemporary developments in their impact on the economy.

Changes in real interest rates suggest that the monetary stance was loosened in all major 7 economies in the first half of 1988, following the stock marketcrash in October 1987. In the second half of 1988 and the first half of 1989 real interest rates increased in all economies except Japan, and further increases followed in the second half of 1989 in all countries except the United States, where there was a substantial decrease. Between the first half of 1988 and the second half of 1989 the largest real interest-rate increases occurred in the United Kingdom, Germany and Canada. Real interest rates remained high in the four European countries in 1990, but fell significantly in Canada, and rose in both Japan and the United States. In the first half of 1991 there appear to have been significant increases in real interest rates in italy and Japan, and significant decreases in the United States and United Kingdom.

The cyclically adjusted financial balances show that fiscal policy was contractionary in 1988 in the United Kingdom, Japan and Canada and expansionary in Germany, France and italy. In 1989 fiscal policy was contractionary in Germany, Italy and Japan, expansionary in Canada and broadly neutral elsewhere. In 1990 there was an especially large expansion in Germany caused by additional expenditure following monetary union, expansion in the United Kingdom and France, and a broadly neutral stance elsewhere.