ARENA Working Papers
WP 97/10



Can EMU Benefit From the Norwegian Experience?
Some Hypotheses

Ton Notermans
ARENA and Department of Political Science, University of Troms�


1. Europe's Failure to Tackle its Persistent Mass-Unemployment is the Main Threat to EMU. [1]

Popular support for further European integration will critically depend on whether the European electorates can be convinced that this is in their economic interests. Despite the hopes of the early European idealists, this pragmatic approach has characterised the process of integration very much from the beginning. [2] Survey's persistently indicate that the European peoples have not (yet) developed a strong sense of European identity. As Elisabeth Bakke (1995, p. 2) has quite correctly noted: ?It seems that the Union so far has failed to capture the imagination of ordinary people.? In the absence of a strong European identity support for integration will hence have to rely on a more utilitarian type of calculus. [3]

The fear of a return to the European wars so common before 1945 might support such a calculus. Indeed for the European Union of Federalists, which experienced its heyday in the years immediately following 1945, the rapid creation of a federal Europe was the only effective safeguard against a relapse into war. And although federalism rapidly lost much of its influence since the early fifties, at present the pacification of Europe still is frequently understood by many as the raison d'�tre of European integration. As e.g. former German Chancellor Helmut Schmidt has argued, the failure of EMU would threaten the whole project of European integration. Without EMU the common market would degenerate into a free trade zone, and this free trade zone would be likely to give way to a resurgence of protectionism and conflict in Europe. ?Wenn es deshalb nicht zur Einbindung Deutschlands und Frankreichs k�me, dann st�nde Europa bereits fr�h im 21. Jahrhundert wieder dort, wo es im ganzen 19. und in der ersten H�lfte des 20. Jahrhunderts gestanden hattte.? (Schmidt 1996). [4]

However, the fear of war would at present hardly seem able to lend sufficient legitimacy to projects like EMU. Not only has the extension of the EU beyond its original six members and the replacement of older politicians by a newer generation made such concerns less pressing, most important, to argue that without EMU, Germany and France might sooner or later go to war again would be to deny the quite obvious success of the reconciliation which has taken place between these two nations, and between West-European nations in general. The possibility that Germans and Frenchmen could again be brought to entertain the idea of going to war with each other seems quite remote indeed, even if their separate currencies would continue to exist after January 1, 1999. In addition the radically changed geopolitical situation since 1945 serves to make a return to warfare even more unlikely. The threat of war within Western Europe, in short, is not foremost on the mind of its people.

What does preoccupy most citizens though is the continuous economic crisis in Europe. Europe is now entering its third decade with stagnation and mass-unemployment. Low growth and durable mass-unemployment, in turn, severely disrupt public finances and welfare state arrangements. With such a large percentage of the European workforce doomed to inactivity, the pensions of future generations are no longer secure. Affordable health care for everybody, moreover, is a principle which is eroded more and more as European states desperately seek ways to reduce their persistent budget deficits. Employment protection is progressively weakened throughout Western Europe in an effort to cut labour costs and thereby allow European firms to fare better in competition with competitors from low-cost countries.

Yet how prosperity is to be restored is an issue which at present is more controversial than probably at any moment during the history of European integration. The political viability of integration projects generally has been promoted by a shared economic orthodoxy between European policy-makers, be it the Keyensian-interventionist orthodoxy of the early decades or the neo-liberal-deregulationist view which manifested its dominance with the conclusion of the SEA. Such shared understandings provided the basis for agreement on the general type of economic governance to which integration projects should conform, while simultaneously providing the basis on whicht the conflicts of national interests over the specific modalities of integration could be fought. [5]

In 1945 it seemed clear to most European policy-makers, given the two recent catastrophes of the Great Depression and the War, that establishing and maintaining economic prosperity and political stability would require a larger amount of politico-economic intervention than had hitherto been practised. As a result of this conviction Europe, despite the many cultural and political differences developed a particular model of political economy which distinguishes it from the USA [6] and Japan in terms of a more extensive welfare state, more emphasis on macroeconomic stabilisation policies, and relatively well developed access of organised interest groups to the process of national decision-making. [7] Reflecting this dominant political consensus in Western Europe from the end of the second world war until the mid seventies European integration has tended to display an interventionist-social democratic rather than a deregulatory bias. Two of the three main integration projects which were undertaken in Europe, namely the ECSC and the CAP were extremely interventionist as they aimed to politically determine quantities produced and prices, in coal and steel and agriculture respectively. Although some parts of the ECSC treaty bore the imprints of US anti-cartel laws and the French planner's convictions that economic interest groups should be excluded as much as possible from decision-making such provisions were ineffectual. In fact, the ECSC could be considered a continuation of the inter-war cartels in coal and steel, but this time with government backing and more efficient international co-ordination, thereby making it more effective. [8] The CAP, in contrast, was explicitly conceived as an arrangement which would suppress market forces in agriculture in favour of the political determination of prices and quantities in very close co-operation with agricultural interest groups. [9] And while the Customs Union agreement did aim at liberalisation, [10] it constituted an attempt to remove trade barriers, which might have potentially debilitating effects for the highly trade dependent West-European economies, without threatening the interventionist forms of economic governance common in the region. [11]

Defence of the European model of political economy hence might seem a promising way to maintain and extend support for European integration. And indeed, especially Social Democrats in Europe presently tend to argue that European Integration and particularly EMU is the only available road to do so. In this view the ongoing internationalisation of the economy, and in particular the internationalisation of financial markets has undermined the successful policies of the `Golden Age' of the fifties and sixties. European integration hence is to be interpreted as an effort to recoup the state's autonomy in economic policy-making by creating a political counterweight to the already internationalised business community. As e.g. the editors of the Swedish LO's journal argued: ?Nu �r kapitalet europeiskt och internationellt. Nu saknas en fackling motkraft mot koncernerna, som sedan l�nge struntar i nationella gr�nser. Nu byggs p� andra sidan upp en altmer verklig r�str�tt p� europaniv� och embryon till en Europeisk v�lf�rdspoltikk.? (?Stegt in i EG? in Tiden 9/10 1990, p.503).

Such a Social-Democratic interpretation, however, has increasingly failed to convince voters and politicians since the late seventies. For many the extensive welfare state arrangements, corporatist patterns of policy-making and the generally high cost level which emerged in Europe since 1945 are not things to defend but the root cause of Europe's economic problems. It was the popularity of the neo-liberal version of supply-side policies which motivated the enthusiasm for the relance of Europe in the form of the SEA. By removing the many-non-tariff barriers behind which European firms had been able to hide, not only a level playing field for competition would be achieved, but a healthy pressure for industrial innovation would be exerted thereby promoting European competitiveness with respect to the rest of the world.

Yet the enthusiasm which informed the SEA has virtually vanished by now. Politically the combination of restrictive macroeconomic policies and deregulatory microeconomic policies, which came to characterise Western Europe since the mid eighties, derived much of its viability from the fact that the effects of unemployment and recession seemed to be confined to a relatively small section of the population. As long as most section of the middle class - the crucial constituency of the European welfare state - were not threatened by mass-unemployment, the deregulation of labour markets and the reduction of support to the unemployed, on which the first wave of neo-liberal policies focused, could hardly be seen as a threat. Rather such reforms could be seen a strengthening the financial viability of those welfare state programmes which benefited the employed middle class. Moreover, since it seemed to be confined to a relatively narrow stratum of blue collar workers, unemployment did not exert a strong downward pressure on middle class wages.

Yet, restrictive monetary policies, successive rounds of spending cuts and market openings (like the SEA) have failed to rekindle growth and stem the spread of unemployment. In the nineties it has become evident that mass unemployment and stagnation in Europe is threatening the core foundations of the European welfare state. And as the living standards of broad sections of the middle-class seem increasingly threatened, the political support for completion of the single market programme has visibly waned.

The core reason why EMU has been so contentious, therefore, seems to be not that it is a programme which leads to a significant reduction of national policy autonomy, but that it has come to appear to many as a programme which intensifies the pressure for the down-sizing of the welfare state. As the date at which the council of ministers will decide on who will participate in EMU moves closer, many governments have to intensify their efforts to cut spending so as to bring their budget deficits within the Maastricht limit of 3% of GDP. The wave of public spending cuts which is presently sweeping across Europe to many rather seems to indicate that EMU further undermines the European welfare state. Moreover, many find it hard to understand why the institutionalisation of the German model of monetary policy at EU level should contribute to growth and employment. The public strikes which were provoked in response to President Chirac's budget proposals, the ongoing protests against Chancellor Kohl's spending cuts, and the successes of extremist parties like the Austrian FP� and the French Front National on the basis of an anti-EMU, anti-austerity platform may be indications of things to come in the next two years.

In the longer run it may prove more dangerous for the future of Europe that the disappointment with the neo-liberal programme of the SEA leaves the mainstream political parties with no viable policy programme, thereby opening up the political space for extremist movements advocating programmes of market closure which start to be reminiscent of some of the proposal vented during the Great Depression. As they themselves have accepted that their traditional recipes do not work in a globalised economy, more interventionist minded social democrats find it hard to capitalise on the disappointment with neo-liberal policies. It would seem hard to understand why, unless Europe resorts to radical protectionism, vastly increasing the competencies of Brussels should solve the problem. After all, also a federal Europe will be confronted with globalised markets for goods and services. [12] The emphasis which is placed on the argument that no policy alternatives exist in a thoroughly internationalised economy hence might easily backfire as it promotes a relapse into more protectionism. Europeans generally have realised that the openness of their economies means that they cannot profit from protectionism in the longer run. Yet if the immediate problems become so pressing the concern for immediate relief might come to outweigh the threat of longer-term setbacks, just as had been the case during the Great Depression of the thirties. It is worth noting that the Front National has replaced the Parti Socialiste as the biggest political party amongst French blue collar workers on the basis of slogan's like: ?Globalisation threatens your job. The Front National fights globalisation.? [13]

2. The Swedish Employment Initiative is not Sufficient.

The main initiative for a concerted attack on unemployment at the EU level presently comes from the Swedish government. At the ongoing Intergovernmental Conference Sweden has presented a proposal for a separate employment chapter to be included in the Maastricht treaty. Despite initial German and British opposition the Swedish proposal is likely to be unanimously accepted. And also the Commission itself and in particular Padraic Flynn are strongly in favour of the Swedish initiative.

Yet the Swedish initiative is unlikely to contribute to a significant reduction in European unemployment mainly because it does not envisage a change in present macroeconomic strategies, and hence is conceived as an addition to and not as an alteration of the Treaty on European Union (TEU). The proposal for a separate employment chapter mainly aims to strengthen European efforts in labour market policies. Swedish social democratic governments, as is well known, have pioneered active labour market policies. Such policies are generally recognised to have contributed favourably to Swedish economic performance [14], and since the sixties they have been upheld by organisations like the OECD as a model for other European countries to emulate.

Yet, active labour market policies were never designed to tackle a situation of mass-unemployment. Swedish labour market policies historically rather were designed as a tool to help combat inflation in a situation of tight labour markets. The dilemma which confronted the Swedish governments almost continuously between 1945 and 1991 was that, on the one hand, the spectacular employment performance led to continuous inflationary pressures, while, on the other hand, the political commitment to full employment ruled out the traditional way of dealing with inflation, namely a restrictive macroeconomic policy which created unemployment. The active labour market policy aimed at reducing the skill mismatch in the labour market thereby preventing potentially inflationary bottlenecks from arising. In other words, active labour market policies were an instrument to keep frictional unemployment to a minimum.

Since the mid seventies virtually all West European countries have switched to a macroeconomic austerity strategy which has increased unemployment in order to contain inflation. And the Swedish social-democratic government has taken this step in the early nineties. Frictional unemployment is not the main problem in Europe today and accordingly an active labour market policy which assures that the unemployed will rapidly an efficiently acquire the skills in demand in the labour market, though doubtlessly a positive contribution, cannot solve the unemployment problem. [15] If the demand for labour is insufficient, programmes which are aimed at the supply of labour will not reduce unemployment. Instead the attempt to solve unemployment by means of active labour market policies may, as has been the case in Sweden reduce the quality of these programmes. [16]

Despite the historical experience of Swedish labour market policies, the replacement of Keynesian demand management by supply-side policies as the preferred cure against unemployment has led some to argue that such policies can create jobs by improving competitiveness. In this view the faltering competitiveness of European industry is not so much due to excessive costs but to its weakness in high-quality, high-technology products. By upgrading the skills of the labour force active labour market policies may help upgrading the quality of the end product.

The German case is commonly presented as the paradigm example of the importance of high quality production. In contrast to what neo-liberal supply side views would expect, the presence of strong trade-unions, extensive labour market regulations and generally high costs did not seem to form an insurmountable hurdle to competitiveness as the German economy has frequently attained an export volume which exceeds that of Japan and the USA. The reasons for this unexpected success is said to be the successful specialisation of German exporters in high-quality products. [17] This version of supply side policies hence held out the hope that high costs and strong labour were not incompatible at all with economic prosperity. Yet exactly the German model illustrates the fundamental flaws of wanting to rely on such a supply-side strategy.

Although it would seem eminently logical from a microeconomic (i.e. firm-level) perspective that improved competitiveness means more jobs it is not clear at all on the macroeconomic level whether a supply-side strategy of this kind is a sufficient of even a necessary condition for full-employment. Given sufficient aggregate demand a failure to keep up with technological developments may rather translate in lagging real incomes and a shift to a less advanced export package than in job-losses per se. In the growth environment of the fifties and sixties even a country with as rigid and conflict-ridden an industrial relations system as the United Kingdom could safeguard full-employment just as Germany. Rather the failure of the British industry to sufficiently modernise its products and production strategies was translated into a rapidly increasing real income gap between Germany and the UK.

If aggregate demand is lacking, however, success in international competition would not automatically seem to spell growth and full-employment. Despite its huge export volume and its almost continuous current account surpluses Germany has experienced high unemployment ever since the Bundesbank decided to stamp out inflation by a dear money policy in the mid seventies. Rather than a high quality production strategy solving the unemployment problem, continuous unemployment seems to undermine the high quality / high costs strategy and replace it with the traditional supply-side policies of costs cutting, deregulation and down-sizing of welfare arrangements. Under the pressure of continuous unemployment, and the resultant fiscal problems, both the employers association and the IG Metall are loosing control of its members. Under persistent conditions of mass-unemployment employers and unions at the firm level have an ever stronger incentive to break away from centrally negotiated collective agreements in order to safeguard employment. As a result IG Metall openly is considering abandoning centralised bargaining (Fl�chentariffvertr�ge) simply because such collective agreements may not be enforceable anymore in the future. [18] At the same time, the government, in order to bring its budget within the Maastricht limits and safeguard export competitiveness sees no alternative but to reduce welfare protection. The framework law of October 1, for example, which allows for the reduction of sick pay from 100% to 80% has already caused serious labour conflicts in the German car industry.

Despite the failure of supply-side strategies to solve Europe's unemployment problem, this is not to say that European economies did not, and do not, suffer from microeconomic problems. The claim that the post-war decades of prosperity gave rise to an over-extension of public regulation and to labour market rigidities which had a negative impact on industrial flexibility is certainly not without grounds. The British trade union's attitude towards industrial modernisation and wage bargaining in the seventies still stands out as a frightening example. Labour market reforms and deregulation (in some areas) certainly have had a positive effect on the competitiveness and flexibility of European industry. And as the example of the British Conservative governments since 1979 shows, such reforms have been popular with the electorate. Yet to significantly reduce unemployment a change in macroeconomic policies will be inevitable.

European mass-unemployment is not the outcome of a dramatic loss of competitiveness of its industry with respect to the rest of the world. Rather mass-unemployment seems to be the result of a policy regime which relies on restrictive macroeconomic policies in order to contain inflation. To expect a significant reduction of unemployment from improved industrial competitiveness under the conditions of generally restrictive macroeconomic policy regimes would rather seem to be an example of the fallacy of composition which holds that those conditions which induce individual firms to employ more labour will also contribute to a reduction of overall unemployment. For European employment to increase in the aggregate, aggregate demand will have to increase. Juhana Vartiainen's (1995, p. 20) observation about Sweden and Finland certainly seems to have relevance for the rest of Europe: ?My suspicion is that many economic problems in Sweden and Finland that are basically due to the failure of macroeconomic management in the 1980s have been falsely attributed to the assumed rigidities of labour market institutions.? [19]

3. The Restrictive Monetary Regime is the Main Obstacle in the Way of Improving Europe's Employment Performance

To reduce unemployment by increasing aggregate demand, unfortunately is a strategy which in Europe has become associated with a Keynesian policy of deficit spending. According to the interpretation of European economic history which became the textbook orthodoxy in the fifties, Europe managed to overcome the persistent unemployment problem which plagued it during the interwar period by abandoning the liberal norm of balanced budgets in favour of Keynesian counter-cyclical spending. Yet this textbook orthodoxy cannot be considered an adequate account of policy-making. As is well known by now, Keynesian deficit spending was largely irrelevant in overcoming the Great Depression. [20] Nor did Western-Europe safeguard full-employment during the fifties and sixties by counter-cyclical fiscal spending which successfully managed to iron out the instability of private sector activity. Even though, Keynes argued that private sector investment was inevitably prone to large swings because it was driven primarily by necessarily uncertain expectations about the future, [21] during the first three decades after 1945 the rate of investment turned out to be not only exceptionally high, but also rather stable. Rather than having to counteract the fickleness of `animal spirits' short term fiscal management was confronted with the task of dampening down demand in order to prevent inflationary overheating. [22]

In fact, Keynesian deficit-spending policies were only practised in earnest during the seventies when a host of countries implemented expansionary fiscal packages in an attempt to counter the rise in unemployment which followed in the wake of the first oil price shock. Those policies are now generally considered to have been unsuccessful. Keynesian pump-priming did not re-ignite private investment. But as the growth rates of the fifties and sixties did not return, the combination of deficit-spending and a welfare state system which was constructed on the background of vigorous economic growth led to unacceptable budget deficits. At present, therefore, there is not much room in Western Europe for fiscal expansion, even if the Maastricht criteria were to be relaxed. If only to prevent debt-servicing from becoming an ever expanding share of public expenditures, fiscal moderation seems inevitable. Europe's best bet to improve its growth performance hence would seem to be a round of aggressive interest rate cuts. [23]

The proposition that expansionary monetary policies can promote employment and growth is a highly contentious one. The traditional theoretical divide concerning the effects of monetary policies on growth and employment is a direct reflection of differences in judgement about the strength of price rigidities in modern day economies. On the one side of the divide are those who hold that no long-term effects of restrictive money can be discerned, and that accordingly durable mass unemployment must have microeconomic causes. In brief the argument runs as follows: In the short term restrictive monetary policies will provoke a recession as they lead to a lower price level (rate of inflation) at given nominal wages implying higher real wages and a lower profit share. Through the market process, unemployment, however, will set in train downward wage adjustments leaving a lower price level (rate of inflation) as the only long term effects of monetary restriction. On the other side of the divide are those who believe that price and wage rigidities are an inevitable element of any modern economy. Monetary austerity will accordingly have durable negative effects.

Empirically both arguments seem somewhat suspect because in many European countries the profit share in the nineties has returned to its level of the sixties without replicating the vigorous private investment of that decade. Theoretically these arguments are problematic because of their neglect of the role of time and money in growth and investment activity. Investment in productive assets is a long-term commitment and accordingly current profit rates and current interest rates cannot function as a sufficient criterion. Instead expectations about the future, i.e. the expected profit rates and expected real interest rates must be seen to exert a crucial influence on private investment. [24] Accordingly also the long term orientation of economic policies, rather than its day to day variations, i.e. the policy regime, becomes of crucial importance. [25] A policy regime which has to resort to monetary restriction in order to combat inflation, however, can only serve to depress expectations about future activity. If central banks must be expected to tighten policy whenever growth accelerates so as to reduce unemployment and increase the probability of inflationary pressures in the labour market, the vigorous growth expectations of the first post-war decades will be impossible to recreate. Hence it is not too surprising that despite high profits in many sectors, firms frequently prefer financial to productive investment. [26] At the same time it becomes understandable why those instances of fiscal reflation in the seventies and eighties failed to rekindle private sector activity and hence faltered on the problem of excessive deficits.

In other words, macroeconomic policies, or better, macroeconomic policy regimes, do have long term effects which are not primarily due to the presence of nominal rigidities, but to the influence they exert on the state of long term expectations. Accordingly a solution to Europe's growth and employment problem cannot be expected from more expansionary short-term demand management policies, e.g. through a relaxation of the Maastricht fiscal criteria, but will instead require a clear, and clearly perceptible shift to a policy regime which no longer is willing to sacrifice growth and employment in order to exclude any possibility, however remote, of a resurgence of the wage-price spirals of the seventies.

In this perspective the fundamental change which has taken place in European economic policies since the seventies is not the internationalisation of financial markets or the emergence of low cost competitors but the disintegration of a model of political economy which did not need tight money but could rely on a set of negotiated incomes and (occasionally) price policies in order to keep inflation in check. And it is the disintegration of the different national models of negotiated wage setting which has been the main driving force behind the trend towards monetary convergence in Europe, rather than a functionalist logic of arriving at political union via the road of economic policies. The idea of monetary union in Europe is not as novel as it may seem. The classical gold standard (1880s - 1914) and the interwar gold standard had many traits in common with EMU. Also the gold standard system required that nation-states abdicate most of their monetary sovereignty in favour of maintaining irrevocably fixed exchange rates under conditions of free international capital movements. In the area of fiscal policies the gold standard system was based on the norm of balanced budgets and hence was even stricter than the 3% norm contained in the Treaty on European Union. The gold standard system is generally seen to have collapsed because of the twin, and interconnected, developments of democratisation and the increasingly dense organisation of civil society. As a result of general suffrage it became politically increasingly difficult to orient the conduct of macroeconomic policies towards external balance rather than towards domestic concerns. The emergence of labour market organisations which made wage determination increasingly a matter of bargaining rather than market pressures undermined the domestic price flexibility on which the Gold Standard relied. The demise of the Gold Standard, in other words, could be seen as the result of the democratic nation-state's need for greater autonomy in the field of economic policies. The two post-war attempts to create monetary union in Europe, i.e. the ECA plans of the late forties and the Werner Plan of the early seventies, floundered on this unwillingness of the nation state to abdicate from economic policy autonomy. The fundamental turn which allowed agreement on the TEU to be reached occurred only when the lack of an external constraint on macroeconomic policies came to be considered problematic by many policy-makers given the increasing inability to contain inflationary pressures in the labour market and demands of additional fiscal spending from civil society. Only at the point where domestic policy autonomy seemed to lead to runaway inflation, exploding budget deficits, and rapidly deteriorating exchange rates could governments like the French and Italian be brought to accept the need for a German style model of monetary policy as enshrined in the Treaty on European union.

4. The Main Obstacle to a Change in Monetary Policies is the Traumatic Experience of the Seventies and Eighties

Despite the emphasis governments currently place on microeconomic solutions to the unemployment problem, the recognition that expansionary monetary policies might be required is not entirely absent. Partly in response to the successes of Le Pen, a substantial minority of the French governing coalition presently advocates cheap money policies instead of EMU. In Germany, the SPD opposition has recently toyed with the idea of internationally co-ordinated reflation. Gerhard Schr�der, a likely Chancellor candidate of the same party, advocates postponing EMU. In Sweden a report, commissioned by the government, on the question whether the country should join EMU came to the conclusion that this would not be advisable in the short run since this, amongst other things, would eliminate the ability to reduce the currently very high level of unemployment by means of more flexible monetary policies (SOU 1996:158). Yet while the lasting unemployment crisis does seem to undermine the taboo of macroeconomic stimulation governments at the same time seem convinced that a strategy of dear money is essential in order to keep inflation low.

That those instances, like the wage price (exchange rate) spirals, frequently accompanied by the spread of pure financial speculation, become a trauma which will inform policy-making for years to come is a well known phenomenon. [27] After all, the instability to contain inflation does mark a fundamental failure of a policy regime which is founded on the discourse of market failures and politically guaranteed full-employment. As a result a fundamental break with old habits seems necessary, and this is best achieved through the institutionalisation of a new set of policies. The centre piece in the regime changes of the last two decades has been a reinterpretation of monetary policies which has accorded central banks the freedom to concentrate on low inflation / fixed exchange rates while disregarding any other concerns. Institutionalising a new regime, however also implies that policies which may have been appropriate in a certain constellation may continue to be pursued long after that original constellation has disappeared. Giving greater independence to central banks implies a choice for monetary policies which will prefer the certain over the uncertain and maintain a high base level of unemployment. Keeping labour markets slack is the best way to prevent a resurgence of inflation. Even if the chances for a renewed wage-price spiral would seem very remote, a significant relaxation of monetary policies does imply that labour market partners have the possibility to conclude inflationary wage contracts. But a central bank which is judged purely by its inflation performance has no reason to take such a risk, however remote it may be.

Yet institutions can be changed. The main cause of regime inertia would hence seem to be politicians' fears of a resurgence of old problems. The breakdown of the post-1945 regime generally implied the advent of a new generation of politicians with a new policy agenda. The sunk political costs included in such a new agenda and the risk of loosing credibility promotes a tendency to call for higher doses of the same medicine if the long term effects of the original strategy prove to be rather different than expected. Yet apart from such purely political reasons, much of the tenacity with which European politicians currently hold on to the policies initiated in the seventies and early eighties seems to be based on the traumas of those years. In particular the following three concerns seem to dominate.

(1) A resurgence of inflationary wage bargaining. A change in monetary regime to the extent that the central bank again assumes part of the responsibility for growth and employment may reduce wage earners' fears that wage increases will result in unemployment and hence may remove incentives to moderate wage bargaining.

(2) A renewed escalation of fiscal spending. Although Keynesian demand management did not play a role of importance up to the early seventies, the pervasiveness of the Keynesian discourse did create well-neigh irresistible political pressures to solve the unemployment problem by expanding the public sector. But also after the decline of Keynesianism in the early eighties many governments find it difficult to pursue fiscal austerity mainly because the rapid growth of public sectors which had taken place since 1945 implied the emergence of a substantial constituency which is dependent on the state for employment and / or income. The combination of monetary disinflation policies, which in themselves increased budget deficits and the political inability to substantially reduce spending led to a constellation where the public sector seemed to be crowding out private employment. Arrangements like the EMS and the EMU for many governments have played the useful role of strengthening their hand with respect to civil society as exchange rate crisis have provided an opportunity to push through fiscal restraint packages which otherwise would have floundered on political resistance. Relaxing those constraints and accepting again that expansionary macroeconomic policies are necessary to stimulate growth and that exchange rate policies should be subordinated to domestic concerns hence may again strengthen political pressures for public sector expansion.

(3) Jeopardising the ongoing process of microeconomic reforms. Whereas in countries like Germany and Sweden prolonged prosperity did not seem to undermine but rather strengthen the willingness of wage earners to accept rapid technological change, in countries like Britain breaking trade union resistance to the adoption of (labour saving) new technologies seemed to require a serious unemployment crisis. Reducing labour market pressures hence may reduce the pace of innovation again.

Accordingly a change in economic policy regimes in Europe would require that those issues be addressed, whereby the first two clearly seem to form the decisive obstacle.

5. The Low Norwegian Unemployment Rate Holds No Lessons for Other European Countries

As persistent mass-unemployment increasingly seems to result in growing political resistance against EMU, policy makers, especially at the EU level have come to pay more attention to the problem. The problem of unemployment played a crucial role at e.g. the council Summits in Edinburgh in December 1992 and Copenhagen in June 1993. More recently the Commission has embraced the idea of a pact between the social partners in order to tackle the unemployment problem and hence increase the support for EMU. At the round table on Employment of 28-29 April 1996 representatives of the Commission, European trade unions, and employers associations, for example, agreed on a so-called European Confidence Pact for Employment which stresses a co-operative approach to Europe's economic problems. In the macroeconomic field the pact however advocates no changes to the present strategy as it sees a policy of combating inflation and budget deficits as the best way to promote investment, growth and employment.

The Commission's interest in social pacts may well have been influenced by the findings of a report written by Stefan Collignon (1994) and commissioned by the European Parliament. In contrast to the Commission, Collignon points out explicitly that more expansionary monetary policies will be needed to solve Europe's unemployment problem. A social pact, or productivity oriented wage policies [28] are considered a prerequisite for such a change in monetary policies. [29] ?It is therefore necessary to focus on labour markets in order to achieve convergence and low inflation growth in the Community. The trap of tight money, low growth and high unemployment could be overcome, if the pressure to combat inflation is taken off central banks by social partners agreeing on a `co-operative game' where all parties win and nobody loses.? (Collignon 1994, p. xxii) [30]

The interest in social pacts which will allow the government to pursue macroeconomic policies for employment without having to fear escalating inflation, inevitably directs the attention towards Norway.

Since 1945 Norway has developed a strong tradition of co-ordinated tripartite bargaining, a tradition stronger perhaps than in other countries. [31] Yet during the eighties the model was confronted with much the same problems that bedevilled incomes policies in other European countries, and eventually broke down. Like in many other countries the dual pressures of structural changes in the composition of the labour force which tended to undermine the traditional predominance of blue collar organisations, and the inter-union pressures resulting from a system of centralised bargaining which bereaved local union members of much of their autonomy, proved too much to handle. By the mid eighties wage drift had attained such proportions that the central agreements reached by the leaders of the peak organisations became rather meaningless. [32]

Table 1: Yearly Wage Increases of Adult Males in Manufacturing

Confronted with escalating nominal wages and inflation, also the Norwegian authorities pulled the macroeconomic breaks, thereby ending the subordination of monetary policy which in fact had guided overall policy strategies since the end of the Gold Standard in 1931. [33] And also here macroeconomic disinflation caused a major recession and rapidly rising unemployment. Yet in contrast to most other countries the change in monetary policy strategy did not establish a durable new regime in which the Central Bank unilaterally handled inflation by means of dear money, while the labour market partners were left to, necessarily unsuccessfully, deal with the resulting unemployment on their own. The shock of unemployment set off a frantic search for a remodelled system of negotiated wage bargaining which again might be able to resume its former macroeconomic r�le. Based on the recommendations of a government appointed commission, headed by former finance minister Per Kleppe (Labour), [34] the social partners and the state in 1993 concluded the so-called solidarity pact for the duration of five years, which again gave negotiated wage bargaining a crucial r�le.

Although there are presently explicit social pacts in operation in Ireland, Finland and the Netherlands (Accord of Wassenaar, 1982), only the Norwegian one might be called an unmitigated success. The Netherlands, Finland and Ireland have significantly higher unemployment rates which makes it difficult to determine to what extent it is the social pact which contributes to low inflationary pressure. In short, it seems that only Norway has been able to again combine something close to full-employment with low inflation.

In line with other strongly social-democratic countries like Sweden and Austria, Norway performed much better in terms of unemployment than other West-European countries during the seventies and most of the eighties. The advent of macroeconomic disinflation in 1987, however caused unemployment to rapidly increase. Open unemployment reached its peak in 1993 with a rate of 6% which meant a tripling compared to 1986 (See figure 1). [35] After 1993 unemployment decreased again. According to the latest figures from the Norwegian statistical office (SSB 1997A) unemployment had come down to 4.2% in the last quarter of 1996.

Figure 1: Unemployment in Norway and the EU

The Norwegian performance also looks impressive when examining labour force participation rates. Unemployment rates often do not give an entirely accurate reflection of labour market pressures. A low overall participation rate generally means that substantial reserves are available which can be mobilised in times of rising employment, implying that the labour market actually is more slack than unemployment figures would seem to suggest. As figure 2 shows, Norwegian labour market participation rates are amongst the highest in Europe. Only Sweden has historically had a higher participation rate than Norway, but since the onset of the Swedish crisis in the early 1990s participation rates in both countries are virtually the same. In the Netherlands, in contrast, which at present display an almost equally favourable performance in terms of inflation and unemployment as Norway does (See figure 4) the labour force participation rate is almost 20% lower.

Norwegian authorities, trade unions and employers generally attribute this favourable economic performance to the aforementioned solidarity pact. As finance minister Sigbj�rn Johnsen, for example, argued recently at a meeting of EFTA and EU finance ministers, the close co-operation between trade unions and employers is the foundation of the Norwegian economic success. [36] The main element of the solidarity alternative was the agreement to keep wage growth at a rate 2% lower than in competing countries, thereby achieving a total improvement in competitiveness of 10% over the duration of the pact. [37] The main aim of the pact hence was to increase private sector employment by improving the competitive position of Norwegian industry. [38] To support such a strategy the report also called for supply-side measures (structural policies) in order to reduce the equilibrium rate of unemployment and an expansion of active labour market policies.

Figure 2: Labour Force Participation Rates

In contrast to the tripartite concerted wage bargaining of the period 1974-78, initiated by finance minister Per Kleppe, which aimed at reducing nominal wages, the Solidarity Alternative focuses on real wages reflecting a shift from a Keynesian interpretation of economic policies according to which employment is determined by aggregate demand to a neo-classical position according to which employment is primarily determined by the price of labour. To be sure, the report does not believe that a combination of improved competitiveness and supply side policies will be sufficient to solve the unemployment problem, at least not in the short run. It does see a r�le for demand side measures but it warns for a vicious circle of growing budget deficits. [39] Fiscal expansion hence should primarily take place using the room which will be created by the higher growth and lower unemployment resulting from improved competitiveness and structural measures. Moreover, fiscal policies should create more employment by reducing transfers and increasing public consumption, within the same budget framework. In the longer run the report calls for an improvement in public finances. Moreover, the report's mandate was to elucidate policy alternatives assuming a fixed exchange rate policy. As a result fiscal polices are subordinated to monetary policies, i.e. fiscal expansion can only be practised to the extent that it does not threaten the exchange rate.

This shift in interpretation from Keyensian to more neoclassically oriented policies seems to have been mainly inspired by the conclusion reached by the Labour government already in the late seventies, that a strategy whereby the state tries to buy nominal wage moderation by means of tax concessions and social security expansion undermined the willingness of unions and employers to come to a bilateral agreement at all as the state could be expected to pay the difference between the unions demands and the employers offers. Technically the Solidarity Alternative hence differs from the Kleppe packages of the seventies in that it is a form of centralised agreement without direct state involvement.

The other major difference between the present Solidarity Alternative and previous tripartite corporatist wage and price moderation arrangements concerns the orientation of monetary policies. Up to late 1986 the central bank basically pursued accommodating monetary policies. From the mid seventies to 1986 this implied that the currency was repeatedly devalued in order to compensate for higher nominal wage growth and inflation rates in Norway as compared to its main competitors. Since 1986 maintaining a fixed exchange rate however, has become the sole policy goal of the central bank, and although the Norwegian Krone was forced to abandoned its unilateral link to the ECU in the exchange rate crises of 1992, any substantial devaluation has been avoided. In late 1986 with high inflation, budget and current account deficits, and downward pressures on the currency, the decision to pursue a fixed exchange rate policy was a decision to pursue restrictive monetary policies. Indeed the change in monetary policies can be seen as a cornerstone in a strategy which was aimed at increasing unemployment in order to get wage bargaining and inflation under control again.

As the Unions by the early nineties were ready to enter into a solidarity alternative whose main conclusion was that a improvement in cost competitiveness was required one might have expected the exclusive orientation of monetary policies on the exchange rate too be softened. If, as the government argued, real wage cuts were necessary to improve competitiveness and employment, and since the unions were willing to accept this, then a devaluation could have been employed without the fear of provoking compensating wage increases. Moreover, it would seem that such a devaluation would have achieved the stated goal of improving export competitiveness much more rapidly than the reliance on lower domestic wage agreement. Yet even with the backing of an explicit wage moderation pact the government was unwilling to consider a more active use of exchange rate policies again.

Table 2: Competitive Positions: Relative Unit Labour Costs in the Manufacturing Sector

Judged by its own criteria the Solidarity pact must be considered a failure as the improvement in competitive position, measured in terms of relative unit labour costs did not come about. As table 2 shows, the competitive position of Norwegian manufacturing did improve during the early nineties but has deteriorated again since 1993. And also the 1996 wage bargaining round, which took place under ever tighter labour market conditions, provided for wage increases in excess of the framework set by the Solidarity Pact.

The major reason why the set goal was not reached was not that the Solidarity Pact failed to moderate the rise of unit labour costs measured in the national currency (See Table 3) but that the floating of exchange rates in the wake of the 1992 international crisis affected some of Norway's main trading partners in such a way that incomes policies were overburdened with the task of improving the competitive position. As table 2 shows, Sweden's position improved dramatically after the Krona was floated in 1992 although the breakdown of centralised wage bargaining in 1991 has proven to be durable there. Similarly the combination of a depreciating currency, very high unemployment and a solidarity pact has radically improved Finland's competitive position. Germany and the Netherlands, in contrast experienced a deterioration reflecting the appreciation of their currencies. In Norway, the Krone appreciated substantially in 1994, thereby undoing much of the improvement in competitiveness which should have resulted from lower wage increases. Pressures for appreciation reappeared again in 1996, yet this time the bank tried to resist by means of lowering interest rates.

In addition there are indications that the Norwegian Solidarity pact did not lead to more moderate wages growth above and beyond what the labour market situation would lead one to expect. In a recent study Evjen & Nymoen (1997) found that the low wage growth in 1993 and 1994 can be very well accounted for by the relatively high unemployment and does not seem to have been affected significantly by the Solidarity Pact.

Since the envisaged improvement in competitive position did not come about, what caused the favourable Norwegian unemployment rate? The low unemployment in Norway, as such holds no lessons for other European countries, as it seems to be mainly a result of oil-financed public sector expansion. Over the last 9 years overall employment in the private sector has fallen by 10.6% from 1.597 million in 1987 to 1.496 million in 1996. Over the same period employment in the public sector has risen by 23.5% from 536,000 in 1987 to 661,800 in 1996 (see figure 2) [40]. In addition open unemployment has been reduced by a significant increase in labour market programmes and education enrolment. In other Western European countries the recession which followed the advent of macroeconomic disinflation generally also meant that a halt was called to public sector expansion, while in some countries public sector employment even declined. Assuming a unchanged level of public employment from 1987, 219,800 instead of 94,000 Norwegians would have been unemployed in the last quarter of 1996 which implies an unemployment rate of 9.8%.

Yet it may be argued that the Solidarity pact has had a beneficial effect on Norwegian unemployment during the last few years. The decline in private sector employment, which started in 1988, reached bottom in the year the Solidarity pact was concluded (1993) with 1.409 millions, and has since recovered by 6.17% to 1.496 millions in 1996. Yet since the improvement in competitiveness did not come about a more likely explanation for the recent private sector employment growth is increased demand due to fiscal expansion. And indeed, most of the private sector employment growth during the last years has occurred in the sheltered sector. [41]

Figure 3: Public Sector Employment as a Percentage of Total Employment, Norway 1962-1996

Without its oil wealth Norwegian unemployment rates hence might look more like those in Sweden and Finland. [42] Indeed, during the eighties economic policies in all three countries developed in remarkably similar ways. All three experienced a vigorous private sector boom with very low unemployment rates and fairly high inflation rates. All three experienced a explosion of the financial system after deregulation which in turn led to a switch to more restrictive monetary policies. The restrictive policies caused a serious downturn in all cases. Yet soon afterward the Norwegian developments started to deviate. Whereas the economic collapse in Sweden and Finland led to rapidly growing budget deficits and hence forced governments to implement drastic fiscal cutbacks, in Norway fiscal policies became highly expansionary again in 1989.

Confronted with strong union complaints over the high unemployment, and declining electoral popularity, the ruling Labour Party apparently decided to reduce unemployment by means of public sector jobs. Politically such a strategy also had the advantage that given the declining share of blue collar labour and the declining popularity of the Labour Party amongst blue collar workers, rapid expansion of public sector employment would serve to enlarge the reservoir of potential labour voters. Yet while many other Social Democratic parties have been tempted to become public sector rather than blue collar parties the economic reason why the Norwegian Labour Party could successfully pursue such a strategy is that the incomes from oil production meant it did not have to face any serious fiscal problems despite the crisis in the private sector. Indeed with the exception of the years 1987 and 1988 the public budget would have been in substantial deficit every single year during the 1980s and 1990s. If one looks e.g. at the years 1989-1991 - years which were critical in stopping the rapid increase in unemployment - the budget deficit corrected for petroleum activity would have been 4.5, 6.2 an 8.6 % of GDP respectively. [43]

In sum, as far as employment creation is concerned the Norwegian Solidarity alternative holds no lessons for other West-European countries. Given the fiscal situation in the rest of Europe, expansion of public sector employment is out of the question. Nor would continuous expansion of public sector employment seem desirable. [44] Moreover, even if the Norwegian Solidarity Pact had succeeded in dramatically reducing unemployment by means of improving cost competitiveness it would hardly be an example for emulation, as such a strategy clearly has a beggar your neighbour character. For the solution of the unemployment problem Western Europe countries will have to rely on a strategy which makes the growth in private sector employment in all countries the driving force behind an overall reduction in unemployment rates.

6. But the Low Inflation in Norway Might Hold a Lesson for Other Countries.

But although the low unemployment is mainly the result of a strategy which may be neither feasible nor desirable for other countries, the fact remains that Norway has successfully combined low unemployment with low inflation (See Figure 4).

An equally favourable picture emerges from Table 3. Table 3 shows the average annual unemployment rate and the average annual percentage increase in unit labour costs during the period 1990 - 1995 for all EU countries - except Luxembourg - plus Norway. With an average annual increase of unit labour costs of 0.9% Norway ranks second after Ireland with -0.1%. Yet Ireland's unemployment rate is more than twice as high. Denmark has the same increase in unit labour costs but also the Danish unemployment rate is twice as high. Adding the figures for unemployment and unit labour costs and ranking them, it can be seen that Norway clearly outperforms all the EU countries. [45]

In short, in terms of controlling inflationary pressures by means of negotiated wage setting the Norwegian Solidarity Alternative appears as the, albeit modified, rebirth of successful corporatist incomes determination. [46] How then could this arrangement be made to work, despite that many analyses have diagnosed the inevitable demise of such policies? And could it serve as a possible model for other European countries, inside or outside the EMU, along the lines of a co-operative policy game as proposed by e.g. Stefan Collignon?

Figure 4: Average Unemployment and Inflation, 1992-1995

Table 3: Unemployment and Unit-Labour Costs in the Business Sector, 1990-1995

7. The Need for Legal Constraints on Bargaining Autonomy

The most obvious reason for the relative success of the Solidarity Alternative would seem to be the centralisation of Norwegian trade unions and employers organisations. Since the 1970s the argument that the institutional structure of organised interests [47] exerts a crucial influence on wage bargaining has become common. The argument in favour of labour market centralisation usually rests on some kind of free-rider argument according to which small sections of the labour force have an incentive to engage in strategies which, although damaging to the economy overall, may improve their own position. Encompassing organisations, i.e. organisations which organise the overwhelming part of the labour market obviously cannot benefit from engaging in a strategy which is damaging to the economy overall. As a result centralisation of wage bargaining institutions is expected to correlate with wage moderation.

Different views as to what the exact nature of this relationship is exist. For some the relation is linear, i.e. more centralisation correlates positively with wage moderation. Others conclude that a U-curve exist between wage bargaining behaviour and institutional centralisation. Very small unions will be dominated by market forces and very large unions will have an incentive to be moderate. Problematic instead are those institutions which are big enough to override market pressures yet too small to have an incentive to pursue macroeconomically responsible strategies. Others have pointed out that, although high centralisation may be beneficial, also intermediate centralisation can achieve the same effects by means of informal, i.e. non hierarchical co-ordination. [48] But whatever the exact relation, it would seem that centralisation / co-ordination should be preferred to decentralised and uncoordinated bargaining systems. To expect the solution of the unemployment and inflation problem from a complete decentralisation or even individualisation of wage bargaining would not seem very promising. In economic upswing completely uncoordinated bargaining is very prone to inflationary wage leap-frogging. In times of high unemployment, in contrast, a generalised downward flexibility of wages is more likely to cause problems of deflation and aggravate the recession than to solve the unemployment problem. [49]

With about 53% the Norwegian unionisation rate [50] is lower than in Sweden, Finland, Denmark and Belgium. Yet, Norwegian labour relations can be considered amongst the most centralised in comparative European perspective. On the union side wage bargaining has traditionally been dominated by the blue collar LO. Although organising workers from all sectors, the metal sectors and large firms tend to be over-represented in LO, as indeed is also typical of e.g. the German DGB and the Swedish LO. During the last decades LO has lost in relative power, both due to a decline of traditional blue collar sectors and to the emergence of two separate union federations; the AF which mainly organises service sector employees with high education, and the YS which has its centre of gravity amongst low income public sector workers. Yet, also partly due to its excellent relations to the ruling Labour Party, LO remains the dominant union. At present LO has roughly 790,000 members and AF and YS have 238,00 and 210,00 members respectively. [51] The employers' side is dominated by the national federation NHO.

The rational choice argument about the relation between institutional structure and wage bargaining, however, is not entirely convincing because of the problem, common to this type of analysis, that the institutional structure itself apparently cannot be understood as the outcome of rational choice. Trade unions are voluntary organisations whose members can join and leave as they please. If small sections of the labour force have an incentive to pursue myopic strategies, then they also do so if they are a part of a encompassing organisation. Indeed smaller sections should have a quite strong incentive to defect, either by actually leaving the organisation or by local bargaining resulting in wage drift, after a moderate encompassing agreement has been struck by the central organisation. But if everyone does so, centralised bargaining breaks down. In other words, if centralisation indeed leads to moderation for the reasons assumed in rational choice analysis then it would seem difficult to explain how centralised organisations can continue to exist. [52]

As the existence of trade unions would seem to indicate, wage bargaining can not simply be analysed as being determined by the institutional features of the labour market but has to be understood in terms of ideology, social norms, solidarity and trust. [53] Centralisation as such may very well lead to radicalisation unless unions feel they operate within a political and economic environment in which moderation will be rewarded. If such an environment exist, then centralisation of wage bargaining can generally serve to increase wage moderation to the extent that it forces (aggressive) individual unions to justify their behaviour in terms of such norms. In the Norwegian context maintaining centralised bargaining has been facilitated by two pronounced ideological traditions, which are less widespread in other European countries. First Norwegian society displays a pronounced paternalistic tendency in the sense of deference to (public) authorities. Secondly, and perhaps somewhat in contrast to Max Weber's views, Norwegian Lutheranism has traditionally considered accumulating wealth and consumption as morally somewhat suspect activities. As a result it may be somewhat more difficult for local unions or union members to press for more autonomy in order to bargaining for higher wages, than it is in other countries. Moreover, centralisation may have been easier to maintain in Norway as compared to Sweden because, even though wage dispersion is small, wage bargaining has put less emphasis on across the board wage equalisation and hence has avoided many potential sources of severe inter-union conflicts. [54] Yet the rapid growth of public sector employment seems to have created more pressures for equalisation in recent years.

Centralisation is only one part of the story of Norwegian wage bargaining. A crucial difference with respect to virtually all other countries is that the Norwegian government maintains the right to force the parties in a labour market conflict to accept compulsory arbitration. [55] Even though Norwegian labour law does not provide for direct state interference in wage bargaining the government, contingent upon approval by the parliament, can force unions and employers to accept compulsory arbitration by the tripartite National Wage Board. According to Fr�land (1997, p. 13) compulsory arbitration has been used about 90 times since 1953. Moreover, the state has used additional legal means to interfere directly in wage bargaining. During 1987-80, e.g. a general wage freeze was proclaimed and in 1988-89 the state forbade non LO unions to accept a better deal than the one reached between NHO and LO. [56] Compulsory arbitration has the single big advantage that it can prevent individual unions /employers associations from setting off a wage-wage spiral. In addition it has also helped maintain centralisation, both by making it more difficult for individual union or unions outside LO to strike significantly different agreements, and by relieving internal tensions in the trade unions. Compulsory arbitration, in other words can at times provide a useful scapegoat function for the union leadership as it can transfer the task of getting militant member unions to accept a moderate central agreement to the state. [57]

Instituting some form of compulsory arbitration hence might be a valuable measure in other European countries, if only to give governments an additional instrument they can rely on to contain nominal wage escalation in tight labour markets so less emphasis can be placed on unemployment creating monetary policies. Admittedly many European unions, like e.g. the Swedish, British and German ones, have traditionally been opposed to government interference in bargaining. However, given the continuing unemployment crisis in Europe it is not so clear any more whether this resistance could not be overcome relatively easily, especially if the possibility of such interference can be coupled with a credible promise of a change in the macroeconomic policy regime. [58] Yet, the Norwegian system has several drawbacks which would make it seem unfit for export as it is. Constraining the right to free bargaining makes it in principle possible for the government to completely undermine the right to strike by simply calling for compulsory arbitration the moment a strike is announced. [59]And given the rapid growth of the public sector throughout Western Europe since 1945, compulsory arbitration means giving the by far largest employer direct access to the state's coercive apparatus in order to enforce wage moderation.

The potential abuses inherent is such a system have not become an important issue in Norway mainly because compulsory arbitration takes place on a background of long-standing relations of trust between the main governing party and the dominant trade union. In contrast to many other West-European unions LO is one-sidedly oriented towards the labour party, even though an ever declining share of its members actually votes for it. [60] This means that non-Labour governments have generally found it very hard to employ compulsory arbitration, especially against LO unions, because of the possible hostile reactions. The close relationship between the Labour party and LO has also implied that, in order not to provoke conflicts, the instrument of compulsory arbitration is not used if a majority of LO unions is against co-ordinated bargaining. [61] Instead such measures have, in the main been confined to breakaway LO unions or non LO unions. On the one hand this arrangement has prevented abuse by the government. On the other hand it has at times implied government intervention in favour of LO and to the detriment of other unions.

Most other West European nations, however, lack the constellation of a hegemonic Labour Party with very close relations to the dominant trade union. In order to make the instrument of compulsory arbitration independent of the political hue of the government, and to safeguard against government abuse or preferential treatment of a particular union, the decision to use compulsory arbitration, in other European countries, should not be left to the government itself. Activation of such an instrument may be best left to a qualified majority of a board composed of union and employer representatives and chaired by state representatives. Representation on the board should be proportional to the number of employees in the specific sectors employers and unions represent. Moreover the same board should appoint mediators. Mediation should generally be guided by the principle that wage increases should be limited to productivity increases although deviations from this rule may be justified in exceptional circumstances. Where the state is needed nevertheless is to make mediation outcomes legally binding on both parties to the conflict. Moreover, the government should have the right to ask the board to consider using compulsory arbitration.

8. The Maastricht Fiscal Criteria Should be Maintained

Although it is important to defend bargaining autonomy in the sense that in general the state should not unilaterally be able to force conflict parties to come to an agreement, the concept of bargaining autonomy in the public sector has always had a somewhat different status. Although it is technically possible to try and isolate state negotiators from direct influence from the government, the state per definition is always involved in public sector bargaining. While it may be possible to maintain that private sector bargaining takes place in the context of a distributional conflict between private interests which means that the state should be reluctant to take sides, public sector employment is paid for out of tax money and hence the general interest is necessarily involved.

At the same time public sector bargaining has been the source of many major conflicts in West European wage setting over the last two decades. [62] One might consider therefore giving the government the right to unilaterally imposed compulsory arbitration in the public sector. To improve the legitimacy of such measures, however, one might also leave the decision to a board consisting of representatives from public and private employers and unions. In any case public sector bargaining should be subjected to the rule that its wages cannot rise faster than comparable wages in the private sector, both because productivity increases in many part of the public sector are notoriously difficult to measure and to prevent public sector employment from becoming more attractive than private sector employment.

As the date on which will be decided who will participate in EMU draws nearer and as most governments fail to meet the criteria the call to allow for larger budget deficits than envisaged in the TEU becomes louder. Yet, a policy regime which couples restrictive monetary policies with a more relaxed fiscal stance, in the longer run can only aggravate budgetary problems. Instead, to assist governments in containing public sector wage escalation, and to prevent an ever increasing level of public debt, maintaining the Maastricht fiscal criteria, or accepting some form of stability pact as proposed by German finance minister Theo Waigel and partly agreed upon during the Dublin summit of December 1996, though in a less rigid form, might be a useful device. [63]

9. EMU Needs to Recognise the Joint Responsibility for Employment and Price Stability

The present labour market situation in Europe with disintegrating unions and falling unit labour costs in several countries seems to indicate that monetary policies rather than labour market partners should take the first step. Given the state many European unions are in at present, it would seem somewhat illusory to call for increased centralisation and co-ordination, possibly even at the European level, in order to allow monetary policies to take a more relaxed stance. And since the unemployment problem, as argued above, results primarily from the long term consequences of the present regime a turnaround in monetary policy would have to imply a rejection of the philosophy of long term money neutrality underpinning present monetary policies. Initiating EMU in its present form might hence give exactly the wrong signal.

Admittedly it is true that the Maastricht Treaty, as the Bundesbank points out [64], does not provide watertight guarantees that the ECB will pursue the same policies as the Bundesbank. Like in the case of the Bundesbank the ECB statutes do not mention a specific inflation target but instead the bank is committed to maintaining price stability. Accordingly there is room for discretion on the board of the ECB. And even though neither European governments nor the Commission will have the right to issue directives to the ECB, independence does not necessarily have to imply inflation aversion. [65] Rather the trade-off the ECB will make between price stability and stimulating employment would depend on the specific outlook of the majority of its members. But even if they cannot be recalled by their governments or the Commission, they are not immune from being influenced by the political and economic developments around them. Indeed the inflation aversion of the Bundesbank may be said to rest primarily on the broad support such policies enjoy amongst the German population rather than on its political independence. ECB board members from other countries may hence have different view on how to handle the unemployment-inflation trade-off than Bundesbank members. Accordingly also, a widespread opposition against the ECB's policies throughout Europe may influence its members.

Nevertheless it seems likely that the ECB will pursue policies which very much resemble those of the Bundesbank, not only because of the reputation of its president designate Wim Duisenberg for ultra-orthodox policies, but also because of the political isolation and the lack of political accountability. Unlike other European central banks, whose statutes can be changed by a simple parliamentary majority, a change in ECB statutes will require a renegotiation of the treaty and hence unanimity. The possibility of a change in statues however is an important factor in preventing that the central bank drifts too far away from the general views in society. [66] In addition because the ECB will not face a single economic policy authority as others central banks do it will operate not only in a climate of larger political independence but also isolation.

It would seem therefore that, in order to communicate the necessary signals, the EMU treaty, it there is to be an EMU at all, would need to be changed so as to state explicitly that restrictive monetary policies very well can have long term damaging effects on economic growth and that therefore the bank necessarily also must accept part of the responsibility for it. Moreover, the council of ministers should be given the explicit right to override the bank's decisions.

10. References

Bakke, Elisabeth (1995), ?Towards a European Identity?? ARENA Working Paper, No 19, April 1995.

B�ckstr�m, Anders (1996), ?En Politik att Sk�mmas F�r.? Dagens Nyheter, 3 October 1996, p. A4.

Calmfors, Lars & John Driffill (1988), ?Centralization of Wage Bargaining.? Economic Policy:14-61.

Coddington, Alan (1982), ?Deficient Foresight: A Troublesome Theme in Keynesian Economics.? American Economic Review, Vol. 72, No. 3:480-487.

Collignon, Stefan, et al (1994), Europe's Monetary Future. London: Pinter Publishers.

Conolly, Bernard (1995), The Rotten Heart of Europe. London: Faber and Faber.

Crouch, Colin (1993), Industrial Relations and European State Traditions. Oxford: Clarendon Press.

de Haan, Jakob & Sylvester Eijffinger (1994), �Het Democratische Tekort van de Europese Centrale Bank.� Economisch-Statistische Berichten (ESB), 7 September 1994.

Dore, Ronald, Robert Boyer & Zo� Mars, eds. (1994), The Return to Incomes Policy. London: Pinter.

D�lvik, Jon Erik, Mona Br�ten, Frode Longva & Arild H. Steen (1997), ?Norwegian Labour Market Institutions and Regulations.? In: Jon Erik D�lvik & Arild Steen, eds., Making Solidarity Work? Oslo: Universitetsforlaget.

D�lvik, Jon-Erik (1996), ?International Change and Transformation of the Norwegian Labour Market Model?. ARENA Working Paper, No. 14, August 1996.

Elster, Jon (1989), The Cement of Society. Cambridge: Cambridge University Press.

Evjen, Snorre & Ragnar Nymoen (1997), ?Har solidaritetsalternativet bidratt til lav l�nnsvekst i industrien?? Norges Banks Arbeidsnotat, 1997/2.

Freeman, Richard B. (1997), ?Are Norway's Solidaristic and Welfare State Policies Viable in the Modern Global Economy?? In: Jon Erik D�lvik & Arild Steen, eds., Making Solidarity Work? Oslo: Universitetsforlaget.

Fr�land, Hans Otto (1997), ?Utvidet inntektspolitisk samarbeid som svar p� ytre utfordringer.? ARENA Working Paper, No. 7,1997.

Gillingham, John (1991), Coal, Steel , and the Rebirth of Europe, 1945-1955. Cambridge: Cambridge University Press.

Gillingham, John (1995), ?The European Coal and Steel Community: an object lesson?? In: Barry Eichengreen, ed., Europe's Post-War Recovery. Cambridge: Cambridge University Press.

Hamilton, Carl & Dag Rolander (1993), Att leda Sverige in i krisen. Stockholm: Norstedts.

Heldring, J.L (1997), ?De Europese motor is in panne.? NRC Handelsblad, NRC Webpagina's, 28 February 1997.

Kielmansegg, Peter Graf (1996), ?Integration und Demokratie.? In: Markus Jachtenfuchs & Beate Kohler-Koch (1996), Europ�ische Integration. Opladen: Leske & Budrich.

Majone, Giandomenico (1996), ?Redistributive und sozialregulative Politik.? In: Markus Jachtenfuchs & Beate Kohler-Koch (1996), Europ�ische Integration. Opladen: Leske & Budrich.

Milward, Alan S. (1992), The European Rescue of the Nation State. London: Routledge.

Moene, Karl Ove & Michael Wallerstein (1995), ?Economic Integration and Unemployment in the EFTA Countries.? In: Barry Eichengreen, Jeffry Frieden & J�rgen von Hagen, eds., Monetary and Fiscal Policy in an Integrated Europe. Berlin: Springer.

NOU 1992:26, En nasjonal strategi for �kt sysselsettning in 1990-�rene. Oslo: Statens Forvaltningstjeneste.

Sassoon, Donald (1996), One Hundred Years of Socialism. London: I.B. Tauris.

Schmidt, Helmut (1996). ?Der zweite Anlauf, die letzte Chance.? Die Zeit

Sk�nland, Hermod (1992), ?Oppgavefordelingen mellom Norges Bank og Finansdepartementet - hvordan den var of hvordan den b�r v�re.? In: Kredittliberalisering og Bankkrise. Det Nye Pengesamfunnet, Raport nr. 35.

Sk�nland, Hermod (1996), ?Hvor stramt arbeidsmarked t�ler sysselsettingen?? Aftenposten, L�rdag 12. Oktober 1996.

Soskice, David (1990), ?Wage Determination: The Changing R�le of Institutions in Advanced Industrialized Countries.? Oxford Review of Economic Policy, No. 6: 36-61.

SOU (1996:158), Sverige och EMU. Stockholm: Finansdepartementet.

SSB (1992A). L�nnsstatistik 1991. Oslo:SSB.

SSB (1992B), De Offentlige Sektorers Finanser 1986-1991. Oslo: SSB.

SSB (1995A), Historisk Statistikk 1994. Oslo: SSB.

SSB (1995B), Statistisk �rsbok 1995. Oslo: SSB.

SSB (1995C), Arbeidsmarkedsstatistikk 1994. Hefte I, Hovedtall. Oslo: SSB.

SSB (1997A), ?Sterk sysselsettningsvekst gjennom hele 1996.? Ukens Statistikk, Nr. 7, 1997.

SSB (1997B), Kvartalsvis Nasjonalregnskap. Kv. 4, 1996. Olso: SSB.

Stocking, George W. & Myron W. Watkins (1946), Cartels in Action. New York: The Twentieth Century Fund.

Streeck, Wolfgang (1991), ?On the Institutional conditions of Diversified Quality Production.? In: Egon Matzner & Wolfgang Streeck, eds., Beyond Keynesianism: Socio-economics of Production and Full Employment.? Aldershot: Edward Elgar.

Temin, Peter (1989), Lessons from the Great Depression, Cambridge: The MIT Press.

Tietmeyer, Hans (1994), �Europ�ische W�hrungsunion und politische Union - das Modell mehrerer Geschwindigkeiten.� Europa Archiv, Vol. 49, Nr. 16, 25 August 1994:457-460.

Tobin, James (1980), Asset Accumulation and Economic Activity. Chicago: The University of Chicago Press.

Tobin, James (1987), ?The Case for Incomes Policies.? In: Peter M. Jackson, ed., Policies for Prosperity. Brighton: Wheatsheaf Books.

Vartiainen, Juhana ?Can Nordic Social Corporatism Survive? Challenges to the Labour Market.? Labour Institute for Economic Research Discussion Papers , 125. Helsinki.


[1] I would like to thank Jon-Erik D�lvik for helpful comments.

[2] See e.g. Milward 1992.

[3] Put differently, the present problem of insufficient popular support for further integration may not be primarily due to the lack of a `grand vision' but to the inability of the EU to deliver the economic goods. See also Heldring 1997, p. 1

[4] Bernhard Connoly (1995, p. xvi) has suggested the opposite, namely that a single currency will make a European war more likely. ?My central thesis is that the ERM and EMU are not only inefficient but also undemocratic: a danger not only to our wealth but to our freedoms and, ultimately, our peace.?

[5] For example, during the late fifties there was common agreement that the problems in agriculture had to be solved by means of extensive market regulation. What exact form the CAP was to take, however, was a highly contentious issue where each nation tried to promote proposals which would benefit its own agricultural sector most.

[6] On the comparison between the EU and the USA see Majone 1996, p. 229.

[7] Admittedly this model of political economy is not entirely unique to Europe. Versions of it can be found in Canada, Australia and New Zealand, and within Europe there are strong differences as e.g. between Sweden and Portugal. See also Sassoon 1996, Ch. 24 and Crouch 1993

[8] The first International Steel Cartel (ISC) (1926-1931) in which producers from Germany, France, the Saar, Luxembourg and Belgium participated was rather unsuccessful. The second ISC (1933-1939) turned out to be much more stable. Partly because general economic conditions were more favourable, but also because governments in the mean time had taken a generally positive attitude towards such arrangements. The British industry, for example, could not take part in the first ISC simply because it lacked the domestic centralisation required to enforce agreements. After 1931 the Conservative government more or less forced the industry to reorganise and centralise allowing it to join the second ISC. On the ISC see Stocking & Watkins 1964. On international cartels during the interwar period also see Gillingham 1991, chapter 1. On the ECSC see Gillingham 1991, 1995.

[9] On the CAP see Milward 1992, chapter 5.

[10] With respect to trade between Europe and the rest of the world the customs union had a protectionist effect.

[11] Provisions like e.g. freedom of capital movements were contained in the original Rome treaties but remained a dead letter until the SEA.

[12] See also Scharpf 1996, p. 133.

[13] In response the French Parti Socialiste probably is going to embrace an explicitly anti-EMU platform.

[14] Since the early nineties, Swedish active labour market policies, however, have been criticised by some Swedish economists because they reduce open unemployment, hence reduce the downwards pressure on wage adjustment and therefore retard the process of disinflation.

[15] About this point there seems to be general agreement in Sweden now. See SOU 1996:158, p. 367.

[16]See Anders B�ckstr�m 1996: ?Arbetsmarknadspolitiken kan inte skapa arbeten. Den kan inte ers�tta en halv miljon jobb som f�rsvinner. F�ljden av dessa f�rs�k har blivit en kraftigt �kad volym av �tg�rder men p� bekostnad av kvaliteten.?

[17] See e.g. Soskice 1990 and Streeck 1991.

[18] And even within large firms IG Metall presently seems to have problems preventing the workforce at individual plants from striking separate agreements. Within Daimler Benz, e.g. some individual plants (i.e. the Gaggenau plant, the axle plant at Kassel and the truck plant at W�rth) have agreed to wage cuts against the advice of the IG Metall.

[19] It may be useful to point out that the view that Europe's unemployment problem is to be attributed to microeconomic (structural) rigidities was embraced politically only after the radical change in macroeconomic strategies which has occurred in Western Europe during the last two decades. Initially the new (monetarist) orthodoxy held out the hope that the increase in unemployment provoked by restrictive monetary policies would be temporary only. As unemployment persisted, however, rather than reassessing the long-term effects of macroeconomic restriction, microeconomic deficiencies were seen to be at the root of the problem.

[20] See Temin 1989.

[21] For a critique of the Keynesian treatment of the effects of uncertainty see Coddington 1982.

[22] This assignment of short-term fiscal management was formalised in the so-called Rehn-Meidner model developed by Swedish trade union economists. It had, however, significant relevance for other West European countries too.

[23] This assumes that the exchange rates of European currencies with respect to the Dollar and the Yen can, and should be treated with benign neglect.

[24] See also Tobin 1980.

[25] On the concept of policy regimes see Temin 1989.

[26] See also Moene & Wallerstein 1995 who present a model of the effects of regimes on firm's hiring policies under fixed capital stocks. Whether firms will fire or hoard labour in a recession depends on whether it is perceived to be long-term or short-term. In other words, a shift to a restrictive macro-regime, also in this model, should induce firms to switch from what Moene & Wallerstein call a fixed employment policy (hoarding of labour) to a variable employment policy (hire and fire).

[27] See Temin 1989 for examples from the interwar period. For a more recent example from Sweden see Hamiton & Rolander 1993.

[28] These are policies which limit wage increases to productivity increases thereby exerting no inflationary pressures.

[29] See also Tobin (1987, p. 376), ?Extra points of unemployment - carrying with them, as Okun's law tells us, multiple extra points of national output lost - are very costly insurance against accelerating inflation. The macroeconomic role of incomes policies is to provide much less costly insurance. Any incomes policy imposes microeconomic costs well known to economists, the allocational distortions our profession loves to tell the world about. That exercise is not sufficient. It is hard to imagine, and hard to argue empirically, that those costs could come close to those of insuring against inflation by permanent excess unemployment.?

[30] Unless one sees EMU as a purely political project, it may be considered somewhat peculiar to call for co-ordinated wage bargaining in order to increase acceptance of EMU since historically the breakdown of negotiated strategies of wage determination in the seventies and eighties formed the background against which policy convergence and agreement on EMU only became possible. If therefore the `co-operative game' between social partners could be brought about, i.e. if governments would not have to fear that a national monetary policy would inevitably lead to renewed inflation, then it might simultaneously become less clear why EMU is so urgent.

[31] See Fr�land 1997.

[32] Between 1980 and 1987, e.g., 70% of the wage increases in industry derived from wage drift. Source: Fr�land 1997, p. 19.

[33] As Hermod Sk�nland (1992, p. 37), who was governor of the central bank at that time put it: ?Dermed var det skapt en ny situasjon for Norges Bank. For f�rste gang siden 1930-�rene hadde vi f�tt en operativ retningslinje for v�r virksomhet ...?

[34] See NOU 1992:26.

[35] Norwegian unemployment rates tend to be biased downwards as they do not count persons in labour market programmes. Including these the unemployment rate in 1995, e.g. would have been 6.7% instead of 4.9%. Source: Freeman 1997, p. 20.

[36] See: ?Solgte Norge til EU.? Dagens N�ringsliv, October 15, 1996.

[37] For an overview see also D�lvik et al 1997.

[38] See NOU 1992:26, p. 23-4.

[39] See NOU 1992:26, p. 42-49.

[40] Public sector employment refers to administrative activities (forvaltningsvirksomhet), i.e. state-owned companies, like e.g. Statoil are not included.

[41] See SSB 1997B.

[42] See also Sk�nland 1996: ?Med et arbeidsmarked som virker etter noenlunde det samme m�nster som i andre europeiske land, er det ingen grunn til at ledigheten skulle v�re s�rlig lavere hos oss dersom vi ikke hadde hatt oljeintektene � gripe til.?

[43] Source: SSB 1992B, p. 32.

[44]One might argue that in some cases the Norwegian public sector has become hydrocephalic in a way which is sometimes encountered in less developed countries. The city of Oslo, e.g. with slightly more than 400,00 inhabitants has roughly 50,000 municipal sector employees.

[45] Admittedly there has been a long term deterioration of the ability of Norwegian wage setting to combine low unemployment with low nominal wage increases. Whereas unemployment for almost every single year between 1945 and 1986 was significantly below the 5% mark, inflation rates until the mid seventies were comparable to those in most other West European countries. The experience of wage bargaining of the last few years, however, would suggest that the labour market partners require a unemployment rate of no less than 5% in order to contain wage increases. During 1996 open unemployment in Norway fell below 5%. As a result of the tight labour markets wage increases exceeded the limits agreed upon by the unions, employers and the government. Accordingly the Central Bank, for good reasons it would seem, warns that at the present level of unemployment one may rather expect a further escalation of wages in the next bargaining round. But instead of containing public consumption and employment in order to prevent overheating the government prefers higher taxes on consumption to cool down private spending and finance increased public spending.

[46] See D�lvik 1996, p. 48.

[47] This in contrast to e.g. traditional Keynesian or neo-classical interpretations of the labour market where objective economic factors, like e.g. the level of aggregate demand are considered to determine wage bargaining behaviour.

[48] See Soskice 1990.

[49] The positive correlation between decentralisation and below average unemployment rates which has been noted e.g. for the cases of the USA, Japan and Switzerland most likely measures other effects like more expansionary monetary policies (USA, Japan) or a policy of repatriating unemployed foreign nationals (Switzerland).

[50] Defined as union membership as a percentage of the labour force. Source of figures: Crouch 1993, p.276.

[51] Source: D�lvik 1996, p. 40.

[52] Alternatively one could assume that economic actors behave according to some kind of `bounded rationality' which leads them to act rationally within a given institutional structure but to accept unquestioningly whatever institutional structure they happen to find themselves in. This, however, would seem to be theoretically inconsistent.

[53] See e.g. Dore et al 1994, Elster 1989 chapter 4 and Crouch 1993.

[54] The relatively widespread occurrence of profit-sharing in Norwegian industry also helps counteract excessive equalisation of wages.

[55] This in fact means that Norwegian regulations are in violation of the ILO convention.

[56] See Fr�land 1997, p. 13-14.

[57] See also D�lvik et al 1997, p. 90.

[58] It should also be noted, that at several instances in the past German, Swedish and British unions have voluntarily co-operated with the state on incomes policies.

[59] Strikes which are continued during and after arbitration are illegal.

[60] Although the German DGB, e.g., is generally sympathetic towards the SPD it cannot allow itself as partisan a political position as LO, because a substantial part of its members vote for the CDU and because the CDU itself has traditionally had an organised working class wing (CDA).

[61] As was the case in the 1996 bargaining round, for example.

[62] The Bundesbank's shift to a monetarist strategy in 1974, e.g., followed after the Brandt government had suffered a humiliating defeat in a conflict with the main public sector union.

[63] Privatisation of those activities which the state does not necessarily perform better, i.e. mail, telephone, transportation, might also strengthen the hands of the government in public sector bargaining as strikes in those sectors are generally very disruptive and hence make it difficult for the government not to make concessions.

[64] See e.g. Tietmeyer 1994.

[65] An extreme example of this is the German Reichsbank, who, under pressure from the allies was made independent during the hyperinflation of the early 1902s. Yet the Bank refused to act against inflation, because, like the majority of German politicians it did not want to assist in fulfilling what it considered to be grossly unfair conditions imposed on the country by the allies. The allies, in response, retorted by making the Bank dependent on them.

[66] See also de Haan & Eijffinger 1994.

List of Abbreviations

AF Akademikernes Fellesorganisasjon
CAP Common Agricultural Policy
CDA Christlich Demokratische Arbeitnehmerschaft
CDU Christlich Demokratische Union
DGB Deutscher Gewerkschaftsbund
ECA Economic Cooperation Administration
ECB European Central Bank
ECSC European Coal and Steel Community
EMU Economic and Monetary Union
EU European Union
ILO International Labour Office
ISC International Steel Cartel
LO Landsorganisasjon i Norge
NHO N�ringslivets Hovedorganisasjon
SEA Single European Act
SPD Sozialdemokratische Partei Deutschlands
TEU Treaty on European Union

[Date of publication in the ARENA Working Paper series: 15.4.1997]