Macroeconomic convergence in Europe.

Anderton, R. ; Barrell, R. ; Veld, Jan Willem 等

Economic convergence within the EMS member states is a precondition for further moves to economic and monetary integration. The Delors Report has set out a path toward EMU in three stages, where transition to the next stage depends on some degree of convergence being achieved in the previous stage. In October 1990, the European Council in Rome agreed that 'in order to move on to the second phase (of economic and monetary union), further satisfactory and lasting progress toward real and monetary convergence will have to be achieved' (Press communique, European Council, Rome, 28 October 1990). The draft treaty text for the Maastricht summit states that before the start of the second phase in January 1994, the council shall assess the progress made with regard to economic and monetary convergence, and in particular with regard to price stability and balanced public finances.

Before transition to the third stage, and not later than December 1996, progress made in economic convergence must be reassessed. On October 28th the Dutch released a draft Treaty text for the Maastricht Summit (to be held on 8th and 9th of December) which stated four criteria for transition to the third stage relating to inflation performance, budget positions, exchange-rate stability and interest-rate convergence. The draft treaty text was accompanied by separate protocols laying down the criteria for economic convergence. Financial Times, October 29th 1991). First, each prospective member state must have 'a price performance that is sustainable and a rate of inflation observed over a period of one year before the examination that does not exceed that of the, at most, three best performing member states in terms of price stability by more than 1.5 percentage points. inflation shall be measured by means of the consumer price index (CPI) on a comparable basis'. The protocol states that the government budgetary criterion shall not be met if there is a council decision that an 'excessive' deficit exists for the member state concerned. Another protocol stipulates that the deficit should not exceed 3 per cent of GDP and the debt to GDP ratio should not exceed 60 per cent. This criterion refers to the general government, that is, central, regional and local government and social security funds. The third criterion relates to a country's record in the ERM. Any prospective candidate for the third stage must have respected the normal, that is, narrow, fluctuation margins in the ERM without severe tensions for at least the last two years before its examination', ruling out devaluations. The fourth criterion states that nominal long-term interest rates must not have exceeded that of the, at most, three best performing member states in terms of price stability by more than 2 percentage points in the last year before examination.

The analysis of convergence

In this note we will discuss the convergence achieved so far in Europe and assess the prospects of further convergence in the coming years.(1) In general terms convergence can be defined as the narrowing of international differences in the development of certain economic variables. As a requirement for a system of stable exchange rates, a distinction must be made between nominal convergence, which is the convergence of the development of costs and prices and their underlying determinants, real convergence of working conditions and living standards and the convergence of economic institutions or structures. Real convergence is one of the fundamental objectives of a fully intergrated Europe, but it is a long-term process and it is not a necessary condition for a successful transition to an economic and monetary union. The note by Begg and Mayes in this Review analyses the issues around real convergence. This note is mainly concerned with nominal convergence but the convergence of economic structures is also briefly discussed.

In a monetary union inflation rates must be similar, and hence it is prudent to insist that countries demonstrate that they are willing and able to converge in this respect before the union is finalised. This demonstration involves getting inflation rates into line, and doing so without strain. The strain would show up in indices of competitiveness, in unemployment and in external balance. Lack of strain implies some conditions, on underlying economic variables, such as fiscal and external balances. Convergence of price performance can only be maintained when such underlying factors do not put pressure on prices to diverge again.

This does not of course mean that fiscal and current account balance are required, rather that they must be consistent with internal and external equilibrium.

If a region is far away from its non-accelerating inflation rate of unemployment, NAIRU, or its fundamental equilibrium exchange rate, FEER, then this cannot be a sustainable situation and some sort of adjustment toward equilibrium will take place. A prolonged overvaluation of the real exchange rate associated with a current account imbalance is not sustainable. Such a situation will lead to lower exports, higher imports and the gradual decumulation of wealth. These will all lead to lower demand and this will put downward pressure on prices. Of course, current account deficits will be easier to finance in a fully integrated Europe with free capital flows, but this will be at the cost of higher interest payments. Such a situation does not rule out an economic and monetary union, but it has implications for the costs of adjustment. An economy cannot continually diverge from its NAIRU and FEER because the automatic stabilising mechanisms in the economy will eventually produce adjustment. The further a country is away from equilibrium, and the longer this lasts, the more painful will be the adjustment. (2)

The transition from high to low inflation also involves reducing nominal interest rates and (in some cases) reducing fiscal deficits. It is arguable that this transition should be made before joining EMU. Fiscal balance is not required for all countries, nor a full convergence of public debt positions. But it is generally accepted that a situation of prolonged fiscal deficits and a rising debt to GDP ratio is not sustainable. The increasing burden of servicing the public debt reduces fiscal flexibility. If countries are not in a monetary union this burden may put upward pressure on real interest rates, and may ultimately lead countries to resort to monetary financing or inflation as methods to reduce the real debt stock. Once inside a union individual countries face fewer financing problems, and hence they may feel some relaxation of the pressure to observe fiscal discipline. A minimum political requirement for transition to EMU may well therefore be that all countries are on a sustainable debt path. This rules out countries that have not managed to bring the medium term trend of public debt under control. For this trend to be stabilising, the primary surplus, the overall public sector balance excluding interest payments, must exceed the product of the desired, or initial, debt to GDP ratio and the difference between the long-term nominal interest rate and nominal growth rate (3). We will show below that not all Community countries have achieved this target.

In the following we will review the progress that has been made so far in the European Community in achieving the necessary convergence of the most important economic variables. We consider not only the countries that are members of the EC at present, but include Austria in our comparison, as Austria has de facto followed the D-Mark over the last 10 years and has arguably achieved closer convergence to Germany than many ERM countries. We will first discuss the exchange-rate developments and the realignments in the ERM. Next we review consumer price indices, as well as different cost measures and measures of competitiveness and real exchange rates. We will also look at the performance of monetary variables such as short and long-term interest rates. We will then consider measures of external and internal imbalances.

Exchange-rate developments

Since its inception in 1979, the ERM has seen 12 realignments. The first years were characterised by frequent tensions within the system and some large realignments. These were partly caused by large differences in countries' external positions but mainly by continuing inflation differentials, which were a reflection of divergent economic performances. These divergences provoked expectations of exchange-rate realignments and speculative capital flows that put irresistable pressure on the system to realign. Table 1 shows all the changes in parities that have taken place since 1979. Only after 1983, when several countries' domestic economic policies changed, did a stable exchange rate in the EMS became the objective for monetary policy. Only then were serious efforts made to achieve the nominal convergence needed for exchange-rate stability. it is often said that the ERM became hard at that time. During the first 10 years, the Italian lira participated in the ERM in wider fluctuation margins of 6 per cent around bilateral central rates and only since 1990 has adopted normal margins of 2-25 per cent. Spain and the United Kingdom joined the ERM in 1989 and 1990 respectively and have opted also for the wider fluctuation margins of 6 per cent. Over the last decade, the italian lira has devalued most, by more than 40 per cent, followed by the French franc and the irish pound. The Dutch guilder has continually shadowed the D-Mark closely and has only devalued by 2 per cent in 1979 and 1983 against the D-Mark. The Austrian schilling has depreciated only 10 per cent against the D-Mark over the period 1979 to 1991, and has been virtually pegged to the D-Mark since 1981.

Prices and costs

The general trend of the last decade suggests an emerging nominal convergence among ERM members. In particular, consumer price inflation differentials have narrowed dramatically over the 1980s. Table 2 and Charts 1 to 2 show that the Netherlands and Belgium, and more recently France, ireland and Denmark, have achieved the greatest inflation convergence towards German rates in absolute terms. Inflation in the Netherlands has actually been lower than that in Germany since 1987 and Austria has also achieved close convergence to German rates. France, Denmark and Ireland have managed to bring their inflation rates down to close to the German rate after having all experienced very high inflation during the first half of the 1980s. ireland in particular has seen a dramatic decline in its inflation rate from a peak of 19 per cent in 1981 to below 4 per cent in the second half of the decade.

Chart 3 shows the inflation rates of the ERM countries that had not achieved convergence to German rates by the end of last year. italy has reduced its inflation differential vis-a-vis Germany from 15 percentage points in 1980 to 3 points at present. Spain has achieved a similar inflation differential. The United Kingdom has, after a diverging inflationary trend in recent years, achieved closer convergence since the beginning of this year. it should be noted that although this general disinflationary experience is also common among many non-ERM countries, their experience has not been so strongly associated with shrinking inflation differentials. However, there is no doubt that the recent inflationary upsurge in the newly unified Germany has exaggerated the impression of inflation convergence. Much of the achieved convergence can be attributed to the increase in the rate of inflation in Germany from around zero to 4 per cent. Furthermore, our latest forecast given in Chapter 2 of the Review suggests that a high inflation rate in the newly unified Germany will persist for the next few years.

Price inflation convergence in several ERM member countries has been associated with the steady implementation of austerity packages with the process beginning around 1982. Since the election of a new coalition government in the Netherlands in 1982, both the real minimum wage and the ratio of unemployment benefits to average earnings have been reduced, followed in 1984 by cuts in nominal public sector wages(4). In 1982 Belgium introduced a corrective policy package of increased taxes, reduced public expenditure, suspension or elimination of wage indexation and the freezing of some prices(5). Also in 1982, Denmark implemented measures consisting of 'tight fiscal policy, wage guidelines, suspension of wage indexation ..... and a fixed exchange-rate policy' (Anderson and Risager (1988)). According to Dornbusch (1989), in 1982 ireland adopted a policy of higher tax rates and tighter money, froze special pay increases in the public sector and 'hardened-up' on the exchange rate. Also in 1982, France introduced a temporary freeze on prices and wages and announced a reduction in budget deficit plans(6). Italy began the process of dismantling wage indexation (the Scala Mobile) in 1983 (7).

As table 3 shows, the above measures helped to reduce the rate of increase in unit labour costs, relative to Germany, for many ERM countries. However, Italy has been less successful in this respect. Non-convergence of Italian unit labour costs has been associated with falling export profit margins. This seems to support the hypothesis that, due to greater exposure to foreign competition, prices in the traded-goods sector are under more pressure to converge than the non-traded sector. Further evidence of this is contained in the 1990/1991 OECD Country Survey on Italy (page 88) which demonstrates that the price of Italian services are rising rapidly compared to other sectors and Italy seems to be an outlier' in this respect compared to other ERM members(8) . Chart 4 gives a comparison of French and Italian manufacturing export profit margins. The slower growth of French unit labour costs seems to have allowed France to converge in terms of manufacturing export prices without severely squeezing profit margins.

Although inflation differentials have narrowed, they have not disappeared. As a result price levels for many ERM members are continually diverging. Given that no substantial currency realignments have occurred since 1987 this means that real exchange rates have appreciated for several ERM countries. In particular, in CPI terms, the Italian real exchange rate has increased by over 15 per cent since 1987 and, in its brief period of EMS membership, Spain has experienced a 10 per cent real appreciation. Ireland, Denmark and France have also experienced less severe real appreciations. There seems little change in fundamentals to justify these real appreciations. On the contrary, in the August Review we argued that France and Italy have experienced a depreciation of their FEERs over the 1980s(9). The real appreciation will make future adjustments more difficult. Charts 5, 6 and 7 plot relative normalised unit labour costs which are commonly used as a measure of the real exchange rate. Once again it is possible to divide these countries into three groups with Denmark, Germany and the Netherlands experiencing real appreciations since 1985. These reflect the nature of their economies rather than the emergence of structural problems. The second group consists of Austria, Belgium, France and ireland where real exchange rates have been constant in recent years. The third group contains the UK, italy and Spain. All have previously used the nominal exchange rate to overcome the effects of inflation differentials. Since they have eshewed the use of this instrument they have appreciated in real terms, reflecting emerging competitiveness problems.

Although not members of the ERM, Greece and Portugal belong to the EC and may participate in a future monetary union if sufficient economic convergence is achieved. However, tables 2 and 3 reveal that substantial adjustment is required within these countries if acceptable growth rates for prices and unit labour costs are to be attained.

Interest rates

Short term interest-rate differentials have narrowed considerably among ERM members. The Netherlands have followed the most credible policy since the inception of the ERM with the Dutch Guilder virtually moving in line with the D-Mark. As a consequence, the Dutch/German interest-rate differential is now almost zero and has been for some time. The same can be said of interest rates in Austria. Chart 8 plots interest rates in these countries along with that in Germany. Throughout the rest of the ERM member countries, interest rates that were almost double that of Germany in the pre-ERM period are now within 1 or 2 percentage points of the German rate. These remaining ERM countries can be divided into two distinct groups, those where interest-rate convergence has recently taken place, and those where it has not. This latter group consists of Italy, Spain and the UK. Even though the differential has been cut to around 2 percentage points this has been at least in part the result of rising rates in Germany rather than failing rates in the UK, Spain or Italy. German short rates are 6 points higher than in the middle of 1988 whilst rates in the UK are only 2 points higher. Chart 9 plots interest-rate differentials for these three countries against Germany.

The third group of ERM countries form a central lane in contrast to the fast and slow groups described above. As Chart 10 shows, interest rates in France and Belgium, and to a lesser extent in Ireland, have been moving together for some time, and in the last two years the differential against Germany has narrowed from around 1.5 per cent to approximately zero. Some of this narrowing has been associated with policy announcements. A good example is the effects of the Belgian announcement in March that they would peg the BFranc to the D-Mark. The interest-rate differential against the D-Mark disappeared almost immediately. In recent months Dutch, Belgian, Danish and Austrian short rates have all been within 1/3 of a per cent of German rates, whilst French and Irish rates have recently moved to within 1/2 percentage point of German rates. it is even possible that French rates may fall below those in Germany in 1992. Short rates in the UK and Italy remain about 1 1/2 to 2 points above those in Germany, whilst Spanish rates (buoyed up by capital inflow controls) have maintained a differential of over 3 percentage points. This period of interest-rate convergence has been combined with the virtual elimination of substantial capital controls, and France and Italy have virtually eradicated onshore/offshore interest-rate differentials.

The draft Maastricht treaty makes great play of the role of convergence of long-term interest rates. Longer term interest differentials have not shown such clear evidence of convergence as short-term rates especially for italy, Denmark, Spain and the UK. Table 4 gives long-term interest rates for the EC countries, and Chart 11 plots long rates on a quarterly basis for the major four over the same period. Long rates embed expected future short rates, and hence if short rates are expected to converge (as is necessary in a Monetary Union ) then long rates must already show signs of doing the same. The evidence from recent changes in long rates suggests that short rates are expected to converge in Germany, France, Belgium, Netherlands, Austria and even Ireland. However, UK, Italian and Spanish long rates show fewer signs of an expected convergence in short rates. This may not only be due to expectations of future higher inflation but, for countries such as italy, the situation may be exacerbated by large government debt/GDP ratios. Nevertheless, if long rates do not converge it is a clear sign that markets are not expecting a union to be formed.

Long rates strongly suggest that Portugal and Greece are not expected to participate in a monetary union in the near future.

Government debt

In recent years, government primary balances have switched from deficit to surplus in France, Ireland, Denmark, Belgium and the UK. In contrast, primary deficits have been consistently recorded over the past decade for Italy , Greece and the Netherlands with Germany joining this group only recently. The sustainability of these positions depends upon whether the net government debt to GDP ratio is stable (although consideration should also be given to the absolute size of the net debt/GDP ratio). Chart 12 shows that, although very high, Ireland's net debt ratio is decreasing as are the smaller debt ratios of Denmark and the UK. Chart 13 reveals that the Netherlands, Belgium and italy have substantial debt ratios, although Belgium seems to have reversed the deterioration whereas Italy and the Netherlands seem to be on a less stable path. As can be seen from Chart 14, sustainable net government debt positions seem to have been achieved by Germany, France, Spain and Austria. Consequently, taking into account both the absolute size and trajectory of debt ratios, convergence conditions require corrective fiscal action in Italy, Belgium, Ireland Greece, Portugal and, perhaps to a lesser degree, the Netherlands(10). However, the link between 'unsustainable' debt positions and real interest rates is somewhat tenuous-Italy and France have had very similar real interest rates over the past five years even though their fiscal positions have diverged substantially.(11)

External balance

From the viewpoint of external balance, Charts 15 to 17 show that Belgium, Denmark, Germany, France, ireland, Italy, Austria and the Netherlands seem to be in sustainable positions. However, this external equilibrium has been a fairly recent phenomena for some countries and provides some tentative evidence of convergence; Denmark and Belgium registered current account deficits of over 5 per cent of GDP in the 1980s and Ireland began the decade with a 14 per cent deficit. Italy may be in a less favourable position than the chart suggests, as she has a small but deteriorating overseas deficit which may worsen due to the high Italian real exchange rate and narrowing of export profit margins. Germany has recently moved into current balance deficit and it should be remembered that the unusual temporary position of high demand for imports by Germany (because of re-unification) has optimistically overstated the sustainability of the trading position of some ERM members. Chart 17 shows that Spain and the UK may have structural trade problems which may only be solved by real devaluation or relatively slower growth (12). However, capital flows are now more mobile and substantial deficits relative to GDP can be sustained for long periods.

Obviously, both fiscal and current balance deficits will be more easily sustained once monetary union occurs. Therefore, rapid convergence of these factors may not be so necessary as other macro-variables.

Unemployment

Both Italy and, in particular, Ireland currently have unemployment rates well above 10 per cent (and also reached higher levels in the process towards inflation convergence). But France, Belgium and Denmark also have high unemployment rates around 9 per cent. The newcomers to the ERM, Spain and the UK, already have substantial unemployment rates of around 16 per cent and 9 per cent respectively (see table 5). The average unemployment rate within the ERM for 1990 of around 10.5 per cent compares unfavourably with the USA and Japan at 5.5 per cent and 2 per cent respectively. There is of course no reason to expect that all countries would have the same level of unemployment when they are at the NAIRU.

High unemployment rates seem to be partly the result of achieving nominal convergence with Germany via the implementation of deflationary packages.(13) The output/unemployment cost of these deflationary policies partly depends upon the credibility of the particular government's commitment to the exchange-rate parities of the ERM. By adopting credible policies consistent with the ERM exchange-rate bands, a government can borrow' the anti-inflation 'reputation' of Germany and reduce inflationary expectations. Studies such as Weber op cit argue that reputation and credibility were not attained for the majority of the ERM countries until the later 1980s (around 1987) which may partly explain the high rates of unemployment prevalent in the last decade within the ERM(14). Artis and Nachane (1987) claim that price expectations were shifted downward in the ERM period. They find that price expectations for ERM countries in the 1980s, relative to the 1970s, were more influenced by German inflation. Barrell (1990) shows that wage and price behaviour in some ERM member countries has undergone structural change in the last decade. However, although this change in structure has reduced the output costs of deflationary policies, it is possible that the fall in inflation has cost 700,000 and a million job losses in France and Italy respectively. (15)

The completion of the internal market program in 1992 also has important implications for the labour market. Integration of the goods market in Europe should result in greater product price elasticity (as monopoly suppliers diminish) and hence a more price elastic demand for labour. Either wage setters will become more responsive to the new conditions in the labour market or unemployment will rise further. It should be noted that the whole process of European integration, since the formation of the EEC, has been associated with rising unemployment in Europe relative to outside the EEC. One would expect the reduction of internal tariffs, greater cooperation and integration to actually decrease unemployment.

Conclusions

It is evident that substantial progress has been made towards nominal economic convergence among the ERM member countries. However, countries such as the UK, Spain and Italy are finding it more difficult to control price inflation and maintain competitiveness. Furthermore, countries such as France, Belgium, Denmark and Ireland have achieved inflation convergence vis-a-vis Germany only at the cost of substantially higher unemployment. Indeed, the ERM could be described as a low inflation-high unemployment regime. Conversely, the other members of the EC, Greece and Portugal, have not attained nominal convergence but have experienced relatively lower unemployment. It is worth noting that the best macroeconomic performance over the period analysed was achieved by a non-ERM country, namely Austria.

Two major factors have had a favourable influence upon nominal convergence. First, most of the last decade has been associated with virtually worldwide macroeconomic stability. Convergence may have been more difficult to achieve if the ERM countries faced, for example, another substantial adverse oil price shock. Second, the inflationary impetus in Germany caused by re-unification has given a flattering impression of convergence. However, compared to the past, German anti-inflation discipline is expected to be somewhat weaker for some time, therefore convergence may be further enhanced. But this does expose the fact that the ERM reputation of providing stable, and low, price inflation does depend on the ability of one nation to maintain control over price inflation.

The convergence of economic structures is an important factor influencing the sustainability of a monetary union once it has been formed. Inflation rates can be made to converge, and eventually it should even be possible for all economies in a union to operate at full employment with low inflation as long as no external shocks hit the system. Diversities in economic structures may well engender differing responses to external shocks, and differential experience of common events may make it politically impossible to hold a union together. Some differences in structure, such as those stemming from natural resources or from culture may be immutable.

The UK, Norway and the Netherlands are energy producers and hence their response to an oil price shock will differ from that of the rest of the Community. Trading patterns, and in particular the Uk's links with the US, inevitably depend on culture. The effect on the Community of developments in the US will depend on trade patterns, and shocks to the US will feed through differentially. The effects of such shocks are inevitably asymmetric, and their implications are analysed in Barrell (1990a).

In Barrell (1990b) we argue that other differences in structure such as those in labour markets may be more malleable. Labour market structures depend in part on the institutions that have been developed to deal with inflation and bargaining. Some countries, such as Italy and the UK, appear to react rapidly to inflationary shocks, whilst others such as Germany respond only slowly. These differences will amplify the effects of common shocks. Even if a shock to the union is in the long run symmetric in its effects, differences in the dynamic patterns of response may make effects asymmetric in the short run. As Barrell, Gurney and In't Veld (1991) argue these asymmetries may make a union difficult to sustain.

A successful union would be more likely if a shock had only small effects on the location of the new equilibrium for the exchange rate and unemployment in the economies of the union. This may require some convergence of structures and institutions. The mere existence of the union may help produce such convergence of structures. Labour market institutions will evolve in relation to the more stable environment. This process can be speeded up in various ways. The single market programme should increase competition within Europe, and this will produce pressures on wage bargainers to act in the same way throughout the Community. it will also increase the sensitivity of trade flows to competitiveness, and as Barrell and In't Veld (1991) show this will reduce the effects of shocks on the location of the equilibrium exchange rate. However, institutional adaption without political assistance can be very slow.

As Europe moves closer towards a full economic and monetary union the whole issue of convergence becomes more complicated. If the date for union is known, this opens up the possibility of devaluations immediately before exchange rates cannot be changed. This action would be beneficial for the devaluing countries in terms of short-term price competitiveness and reductions in the real value of government debt, but it might harm the anti-inflation reputation of the counties concerned. For this reason, some authors (for example, Froot and Rogoff (1990)) argue that convergence has peaked as expectations of the- above scenario will cause prices and interest rates to diverge now. Conversely, the prospect of irrevocably fixed exchange rates may provide extra policy credibility and enhance the process of convergence.

REFERENCES

AMEX Bank Review (1991),18 September 1991, vol.18, no.7.

Alogoskoufis, G., Papedemos, L. and Portes, R, External constraints on macroeconomic policy: the European experience.

Andersen, T., and Risayer, 0. (1988) Stabilisation policies, credibility and interest rate determination in a small open economy,' European Economic Review, Papers and Proceedings, 32 pp. 669-673.

Artis, M.J. and Nachane, D., (1990), Wages and prices in Europe, a test of the German leadership hypothesis', Weltwirtschaftliches Archiv 126 pp. 59-77.

Barrell, R.J. (1990a) 'European currency union and the EMS', National Institute Economic Review no. 132 May.

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

Barrell R., A. Gurney and J.W. in't Veld (1992), The real exchange rate, fiscal policy and the role of wealth: an analysis of equilibrium in a monetary union', Journal of Forecasting, forthcoming.

Barrell R., and J.W. in't Veld (1991), FEERS and the path to EMU', National Institute Economic Review, no. 137.

Barro, R.J. and Xavier Sala-i-Martin (1990), World real Interest Rates,' National Bureau of Economic Research Macro-economics Annual 1990 pp. 15-74.

Chan-Lee, J.H., Coe, D.T., and Prywes, M.(1987) Macroeconomic changes and Macroeconomic wage Disinflation in the 1980s', OECD Economic Studies no. 8, Spring pp. 121-157.

Commission of the European Communities (1990), 'One market, one money: an evaluation of the potential benefits and costs of forming an economic and monetary union', European Economy, no. 44 October.

Commission of the European Communities, (1989), 'Economic convergence in the Community: a greater effort is needed', European Economy, no. 41.

Commission of the European Communities, (1990), Annual economic report 1990-91', European Economy, no. 46, 1990.

Dornbusch, R. (1989), Credibility, Debt and Unemployment; Ireland's failed stabilisation,' Economic Policy, no. 8 (April)

Dornbusch R. (1991),'Problem of European monetary integration', in A. Giovannini and C. Mayer (eds): European Financial Integration, Cambridge University Press.

Froot K.A. and Rogoff K. (1991), The EMS, the EMU, and the transition to a common currency', NBER Working Paper no. 3684, 1991.

Kremers, J. (1990), Gaining policy credibility for a disinflation: ireland's experience in the EMS', IMF Staff Papers, vol. 37, no.1.

Layard, R., Nickell, S. and Jackman, R. (1991), Unemployment: Macroeconomic Performance and the Labour Market, Oxford University Press.

Mehta, F. and Sneesens, R. (1990), Belgian unemployment: the story of a small open economy caught in a worldwide recession', Europe's Unemployment Problem edited by Bean, L.R. and Dreze, J.H. MIT Press.

Ungerer H., Evans, O., Mayer, T., Young, P. (1986), The European Monetary System: recent developments', Occasional Paper 48, international Monetary Fund, Washington D.C.

Ungerer H., Hauvonen, J.J., Lopez-Claros, A., Mayer, T.(1990), The European Monetary System: developments and perspectives,' Occasional Paper 73, International Monetary Fund, Washington D.C., 1990.

Weber, A. (1991), Reputation and credibility in the European Monetary System', Economic Policy, no. 12 (April).

NOTES

(1) For other studies of convergence in Europe see Ungerer et a/(1986,1990), European Economy (1989,1990), Froot & Rogoff (1991) and AMEX Bank Review (1991).

(2) In Barrell, Gurney and in't Veld (1992) we demonstrate that the process of real exchange-rate adjustment may be very protracted and the adjustment costs are unequally shared, which could cause severe strains to be put on the union.

(3) The requirement for a debt/GDP ratio to stabilise at its present level is p = (i-g).d where p is the primary balance, i the nominal interest rate, g the nominal growth rate and d the desired or present debt/GDP ratio. This simply states that the non-interest surplus must be large enough to off set the increase in the debt/GDP ratio due to interest payments on debt. This condition can be modified to allow for taxation of interest payments on debt, valuation changes on existing debt and money financed borrowing.

(4) Chan-Lee et al (1987) point out that the Netherlands 'reduced nominal minimum wages for workers under 23 by 10 per cent in 1983'.

(5) See Mehta and Sneesens (1990) for further details.

(6) Weber (1991) makes it clear that the French austerity programme really began in earnest after the March 1983 realignment of the French Franc.

(7) Barrell (1990) gives details of the Scala Mobile and shows that in 1983 the degree of wage indexation was reduced from 1 to 0.85. However, he states that the most significant reforms of the italian indexation mechanism occurred in 1985.

(8) The OECD report goes on to state that 'the level of wages for both private and public sector services (in italy) has continued to exceed that in the exposed sector, suggesting that both employers and employees share the economic rent caused by the lack of competitive pressure'.

(9) See Barrell and in't Veld (1991).

(10) For calculations of the required primary surplus for debt sustainability for the ERM countries see European Economy: One Market, One Money.

(11) Barro (1990) estimates reduced-form models for expected real interest rates for ten OECD countries and generally finds fiscal variables to be an unimportant determinant of real interest rates.

(12) Alogoskoufis et al (1990) provide a thorough examination of external constraints for European economies.

(13) in contrast, the failure of Greece and Portugal to achieve nominal convergence has been associated with relatively lower unemployment rates.

(14) Opinions differ here as to the role of credibility. For example, Dornbusch (1989) argues that irish inflation convergence was attained via old fashioned deflationary policies with no extra credibility effect, whereas Kremers (1990) argues that extra ERM credibility may have decreased the unemployment cost of reducing inflation in Ireland.

(15) The necessity for persistently high unemployment rates within the ERM in order to reduce inflation is a controversial issue. For example, in 1988 the long-term unemployed accounted for more than 60 per cent of the unemployed in Belgium, Ireland, italy and Spain (see page 51 of Layard, Nickell and Jackman (1991)). As the long-term unemployed are generally thought to contribute very little to the disinflationary process, the inflation cost of re-employing them may be negligible.