The "Cologne Process": a Neglected Aspect of European Employment Policy

Arne Heise*

The inclusion of an „Employment Chapter“ in the 1997 Amsterdam Treaty closed a fundamental loophole in its Maastricht counterpart, signed six years earlier. For, even after the path-breaking agreement on a single currency, critics of European integration had still been able to complain – with some justification – that the convergence criteria set for participation in European Monetary Union (EMU) placed too much weight on the new currency’s price stability, and too little on the employment situation in the EU. Indeed, it could be said that the Maastricht Treaty still bore the hallmarks of the economic - and political - priorities of the 1980s, the „monetarist decade“. As growing importance was attached to combating the rising trend in unemployment, a pressing need became apparent; the European Central Bank’s remit to ensure price stability in the future „Euroland“ had to be balanced by assigning the Union explicit responsibilities for employment. Thus was a new field of EU policy opened up.

The Three Pillars of Employment Policy in the European Union

(Both the Luxembourg and Cardiff processes stress the importance of functioning markets and are based on microeconomic considerations of allocation theory and supply-side policy; they are concerned with improving employability rather than quantitative employment. By contrast, the Cologne Process is founded on macroeconomic, demand theory notions.)

Without doubt this development was due in part to the constant political pressure applied by those member states which have traditionally had more active labor market and employment policies; Sweden and Austria. But it can also be attributed to the EU’s gradual „social-democratization“, a process whose crucial moment was the arrival in power of Lionel Jospin in France and Gerhard Schröder in Germany. Yet the Union’s assumption of employment policy responsibilities in the Amsterdam Treaty was merely the beginning of a conceptual process which shaped the three pillars of current EU employment policy. These are:

• the „Luxembourg Process“, initiated at the 1997 Luxembourg Summit, which established that member states would coordinate their labor market policies. In annually approved Employment Policy Guidelines (EPGs), based on work by the Commission and the newly created Employment and Labour Market Committee, the European Council identifies priorities for individual members’ labor market policies, on the implementation of which national governments are required to report in National Action Plans (NAPs)

• the „Cardiff Process“, named after the Cardiff Summit of 1998, which embodies the hope that liberalization of product and financial markets can stimulate the structural changes and dynamism required to create knowledge-based economies in the various European countries, while simultaneously deepening European integration. In annual „national reports“ and „Cardiff reports“, the Commission and national governments exchange information on progress, priorities and omissions.

• the „Cologne Process“, the most recent of the three. Agreed at the 1999 Cologne Summit, it recognizes that a favorable macroeconomic situation is the prerequisite for lasting improvements in growth and employment, and that coordination of budgetary, monetary and incomes policy – so-called „EU macro-dialogue“ –is therefore desirable. For the first time the ECB’s activities have been set in an employment policy context, and the European social partners tied in to the process.

Both the Luxembourg and Cardiff processes stress the importance of functioning markets, and are thus unmistakably based on microeconomic considerations of allocation theory and supply-side policy; they are explicitly concerned with improving „employability“ rather than purely quantitative „employment“. By contrast, the Cologne Process is founded on macroeconomic, demand theory notions, so that it can also be seen as part of a „Euro-Keynesian“ strategy. It is thus a clearly distinctive strategy – indeed, it has been heavily criticized for that very reason by advocates of the structural reform approach of the Luxembourg and Cardiff Processes - and worthy of examination as such. But focusing on the Cologne Process can also be justified in that little attention has so far been paid to it. Nor has it yet found much expression in the policies of the individual member states - unlike, for example, the goal of price stability, which has „spilled over“ from European to national level as a result of the single currency and the Maastricht convergence criteria. Our intention here is therefore to redress the balance, by assessing the potential and limitations of EU macro-dialogue. This will involve setting out the theoretical basis for coordinating the behavior of macroeconomic actors, which will then provide the background for an assessment of initial experience of the Cologne Process.

First, however, we wish to examine briefly how the Cologne Process came into being, as an aid to understanding its limited importance to date. The idea that lasting improvements in the employment situation demand that policy be given a stronger macroeconomic emphasis is most clearly expressed in works produced by European social democrats; the so-called Larsson Report of 1993, and the 1999 Guterres Report. However, it was also favored by elements within the European Commission, whose views were reflected in the 1994 White Paper „Growth, Competitiveness, Employment“. As social democrats’ influence within the Union grew (by 1998 they governed, alone or in coalition, 13 of the 15 member states) opposition to a strategy of macroeconomic intervention weakened. Nonetheless, it was not until the change of government in Germany that a window of opportunity opened, enabling macro-dialogue to be placed on the EU agenda and later institutionalized at the Cologne Summit. One of the key advocates of macro-dialogue as a means of tying the ECB into a European employment strategy was Oscar Lafontaine, and with his resignation as German Finance Minister the window quickly closed again, as soon became apparent. First, a more radical initiative by the French and Italian governments (see Council document 8906/99 of 2 June 1999) was countered by a joint proposal from the Spanish (conservative) and British (Labor) administrations, which stressed workers’ employability, and the capacity to adapt of labor and product markets (see Council document 8906/99 of 2 June 1999). Then, at the Special Summit held in Lisbon in March 2000, the change in priorities was confirmed. There the Cologne process was mentioned only in the preliminaries; nowhere was coordination of macroeconomic behavior as a means of promoting market conditions favorable for expansion identified as one of the EU’s fundamental policy tasks. Moreover, unlike its Luxembourg and Cardiff counterparts, the Cologne Process is not subject to reporting requirements, and thus neither to public or indeed any control. Its limited significance can probably be ascribed to the disagreements among European social democrats over strategic direction. But that does not alter the fact that within the EU interest exists in exerting greater macroeconomic influence. In the following we wish to explain this interest, and to examine the prospects for EU macro-dialogue against the background of initial experience.

Rationally Coordinated Economic Policy

(Were it possible to compare an economy with horizontal coordination against another without it, then the latter would display higher interest rates, higher prices, higher unemployment and more public debt – in other words, a clearly inferior overall situation.)

(The Stability and Growth Pact is based on an understanding of how economic policy responsibilities are assigned that is incompatible with cooperative economic policy in the sense of horizontal coordination designed to create a positive-sum game.)

The effects – both positive and negative – of concentrating responsibility for independent sub-areas of economic policy are far from accepted, as can be seen from the sometimes strong reactions evoked, not just by the notion of macro-dialogue, but also by the corporatist „Alliance for Work“ in Germany. The alternative approach, so-called assignment, involves a clear division of responsibilities for fear that „corporatism could run amok“.[1] Before we can assess such criticisms, however, a closer examination is required of the prerequisites for, and the forms and established effects of a cooperative economic policy, with specific reference to EU macro-dialogue.

Cooperation Versus Corporatism

The economic interaction of various individuals or actors can be analyzed in a two-dimensional space. One dimension relates to the nature of their goals; the actors concerned may pursue one (or more) common aims. Alternatively, their actions and choices may be determined by contradictory, competing goals, as may be assumed in the case of consumers and producers. The second dimension relates to the nature of interaction, with a distinction being drawn between horizontal and vertical coordination of actors’ activities (see Fig 1)

Market coordination occurs when two (or more) actors interact on the same hierarchical level, in the absence of common goals. In this case, interests are reconciled by the price mechanism. For vertical coordination to occur, some actors must be clearly subordinated to others. This is true of the various parts of an organization (e.g. a firm), but can also be expected in macroeconomic policy processes where different policy fields are mutually dependent on each other. Vertical coordination without common goals is only conceivable in prisons or planned economies,[2] and requires no further comment. More interesting is the combination of common goals and horizontal coordination. This variant occurs, for example, when interdependencies can be assumed between different macroeconomic policy areas controlled by independent agencies not subordinated to those state actors with relevant policy responsibilities (normally the government). Where central banks or social partners enjoy independence, as the agencies of monetary and wage policy respectively, such horizontal cooperation becomes necessary. It should also be stressed that horizontal cooperation must be clearly distinguished from horizontal coordination of behavior in the absence of common goals, so-called „antagonistic cooperation“,[3] as represented in Germany by the  „Alliance for Work, Training and Competitiveness“, or the former, unsuccessful „concerted action“ envisaged by the „Stability and Growth Act“  and based on traditional Keynesian ideas.[4] Both these initiatives are – or were  –attempts to imitate the effects of pure market coordination by means of corporatist concession bargaining; critics take their approach to its logical conclusion, and demand the reinstatement of true market forces.

Figure 1: Types of coordination

Is Coordination Possible?

EU macro-dialogue seeks to initiate horizontal cooperation between three policy areas which in the Union are largely independent:

·        monetary policy, the responsibility of the ECB;

·        budgetary policy, the concern of the Ecofin committee and the Commission; and

·        incomes policy, which is a product of European social dialogue.

Below, the coordination procedure concerned will be described in detail and assessed, but first two other questions must be examined: the economic rationality of such coordination, and the conditions for its success.

The interdependence of the various policy areas can be justified by construction of a Phillips curve, in which those responsible for them pursue the common goal of a given combination of price stability and labor market parameters, but display differing preferences with respect to inflation and unemployment. Using a formal model[5] it can be shown that, if behavior is not coordinated, none of the political actors can reach their desired utility, expressed by a position on the Phillips curve. In that case, all actors – central bank, government and social partners – must accept a loss of utility relative to the situation where cooperation occurs. Specifically, it transpires that monetary policy is tighter than would be required merely on price stability grounds, while budgetary policy is subject to „hegemonic coordination“ by the independent central bank, the result being higher borrowing and reduced room for manoeuvre. Meanwhile incomes policy, and in particular the unions, must accept increased unemployment with no compensating distributional improvements. Were it possible to compare an economy with horizontal coordination against another without it, then the latter would display not only higher interest rates and higher prices (or inflation), but also higher unemployment and more public debt[6] - in other words, a clearly inferior overall situation.

Cooperation therefore seems desirable, in that it increases utility - and not just for the individual actors but also for the economy as a whole. It might therefore be expected to arise spontaneously. Yet empirical studies clearly indicate that coordinated behavior occurs by chance, if at all. That does not mean that actors are ignorant, ill-directed or downright malicious, however. Rather they are both rational (i.e. they pursue defined goals in a consistent matter) and selfish (i.e. they value increases in their own utility more highly than those of other actors). Under these circumstances, actors will be caught in the classic Prisoner’s Dilemma unless it is somehow assured that all the actors concerned will play their part in cooperation (in the case of macroeconomic coordination, that would involve the central bank permitting a more expansive monetary policy, the government adopting a less restrictive, but sustainable budgetary policy,[7] and the social partners agreeing to a non-inflationary incomes policy, or wage moderation). Otherwise it remains preferable, and indeed necessary, for the individual actors to accept a macroeconomically inferior outcome in a situation of general non-cooperation. For the alternative is to become a „willing victim“ who makes the concessions required by cooperation, and is exploited by other actors who do not.

Game theory considerations do not merely warn us against unjustified assumptions of „spontaneous cooperation“, however. They also point to the conditions which must be fulfilled if the Prisoner’s Dilemma is to be resolved, and cooperation occur. The most elegant option would be to bind the actors contractually, making non-cooperation sanctionable, but that is effectively ruled out in the case of macro-dialogue. A binding agreement of this type would involve dependence on a third party, to decide whether the actors had, in fact, fulfilled their contractual obligations, and is in any case barely conceivable for independent actors. The alternative is an implicit agreement, under which sanctions take the form of non-cooperation costs (the so-called „long shadow of the future“). In this case institutional structures must be devised to minimize the danger that „willing victims“ are exploited.

• Communication: The first essential is that the actors must be willing and able to communicate with each other; without communication a „cooperative game“ is unthinkable. However, „communication“ must go beyond the mere exchange of information which could in any case be gleaned from the actors’ own relevant material or press statements. Rather, communication here implies exchanges about cooperation itself, about potential gains and the costs of non-cooperation, about a cooperative strategy and about anything which might increase actors’ trust that cooperation on their part will be reciprocated by others.

• Monitoring: Communication is a necessary, but not a sufficient condition for cooperation. It must also be possible to specify and monitor actors’ individual contributions to cooperation. Only when it is clearly and generally accepted that all such contributions have been made can behavior in „the next round“, i.e. in future interactions, be determined. And that means that guidelines must be established to enable contributions to be verified.

• Sequence: To avoid the first-mover trap a sequence, i.e. a succession of cooperative actions and responses, must be established. That also addresses the problems arising from the fact that, in pursuing their policies, individual actors may have to consider other market players as well as the other policy actors involved in the „game“. Specifically, a central bank cannot ignore the financial markets, if it wishes to have any chance of achieving its goals. Indeed, the markets may demand complete central bank independence as a prerequisite for its credibility. In that case the bank cannot be a „follower“ – an actor who responds to the cooperative contributions of others – but must be a „leader“ setting the cooperation agenda.

• Game strategy: Finally, a „game strategy“ is required that minimizes the utility losses for an actor who nonetheless becomes a „willing victim“. Here game theory prescribes „tit-for-tat“ as the ideal strategy. Simple and unsophisticated, it signals willingness to cooperate while punishing non-cooperation mercilessly.

To enable lasting cooperation between macroeconomic actors, these parameters must be set in an institutional context (so-called „structural embeddedness“). This must give actors security and confidence and so underpins cooperation. In particular, communication must occur within a stable institutional framework. Monitoring of actors’ behavior and establishment of generally accepted guidelines requires a „neutral“ authority, which must be equally respected by central bank, government and social partners – only then will its decisions be effectively binding. Elsewhere I have proposed[8] a „Socio-Economic Committee“, composed of representatives of the various policy actors, and an „Expert Committee“,[9] made up of academics enjoying the relevant actors’ trust, which together could serve as the institutional framework for macro-dialogue. The Expert Committee would take on the monitoring role, working on the „papal“ principle; in other words at specified points in time it would be required to issue unanimous recommendations. That would prevent individual members’ views from being overruled by a majority, while creating pressure for agreement. The results of the monitoring process would be communicated in the Socio-Economic Committee, which would then discuss and communicate sanctions.

All in all, the prerequisites for successful coordination of macroeconomic actors’ behavior in the context of a macro-dialogue are numerous. Yet the underlying conditions are clearly favorable. Since all actors can profit from cooperation by approaching their desired position on the Phillips curve more closely, the game has a positive sum. Unlike in a zero-sum game such as „antagonistic cooperation“, there is no need for them to be constantly on guard against losing out in the process of give-and-take. That should make it relatively easy for actors to obtain the internal legitimation necessary to participate in macro-dialogue, either from their members (in the case of the social partners) or from voters (in that of the government).

EU Macro-Dialogue: Architecture and Initial Experience

(EU macro-dialogue as yet barely extends beyond information exchange.)

Among EU members Austria is particularly active in employment policy terms, and also has extensive experience with cooperative approaches to economic policy through its system of social partnership. And it was the then Austrian presidency, at the Vienna Summit of December 1998, which first picked up on work carried out by the European Commission’s economic section (DG 2) to propose the participation of macroeconomic policy actors in designing a European employment policy with good prospects of success. The outcome was a so-called „Vienna Strategy for Europe“, that laid the foundations for macroeconomic dialogue. It remained only for Germany, during the „window of opportunity“ provided by the new Schröder-Lafontaine government, to ensure formal agreement at Cologne.

There a conceptual framework for macro-dialogue was defined, by the following three conditions (see Council document 8327/99, p5): no infringement of actors’ independence; no undermining of the Growth and Stability Pact; and, maintenance of the principle of subsidiarity. Within this framework, it was accepted:

• that budgetary, monetary and incomes policy are interdependent, so that only a cooperative policy mix can be optimal in macroeconomic terms; and

• that these macroeconomic policy areas have clearly distinct responsibilities - budgetary policy for a balanced budget, incomes policy for employment and monetary policy for price stability.

The result is a delicate balance between the requirement to cooperate, on the one hand, and assignment on the other, a point to be discussed later.

First, however, it is necessary to describe the institutional framework of the Cologne Process, which consists of two different levels. The technical level is concerned with analysis of economic performance and thereby providing the basis for macro-dialogue proper. To that end a working group has been established, consisting of representatives of the Commission’s Committees on Economic Policy and on Employment and Labor Market, of the ECB and of the macroeconomic working group forming part of social dialogue (in which the social partners are represented). It meets on a six-monthly basis, once before the Commission’s economic policy guidelines are laid out, and once after presentation of the EU’s annual economic report. Macro-dialogue proper takes place at the political level, in a forum which brings together the Commission and policy actors to discuss strategies for policy coordination and confidence building. It consists of an extended Ecofin Council taking in the members of the Labour and Social Affairs Council, as well as representatives of the social partners and the ECB. This forum, too, meets twice a year, once prior to approval by Ecofin of the economic policy guidelines, and once before the conclusions from employment policy guidelines are accepted by the European Council. To keep dialogue manageable, each grouping is required to nominate two representatives only; meetings are chaired by the Council member of the country forming the Union presidency at the time.

For obvious reasons, experience of EU macro-dialogue remains strictly limited; as yet only four meetings have taken place at technical and political level. However, there are already signs that the Commission, with its bureaucratic apparatus, is taking on a lead role, at least at technical level. There, the economic forecasts produced by the General Directorate for Economic and Financial Affairs form the basis of discussion, and the effects of so-called „exogenous shocks“ (such as the oil price changes of 2000 and their effects on prices and public budgets in Euroland) are observed and analyzed. At political level, the particular attention was attracted by the recent criticism of Ireland, whose macroeconomic policy mix was regarded by the Ecofin Council as too expansive and a threat to price stability. Three points are apparent:

• the EU is willing to place greater priority on macroeconomic conditions in Euroland than on the interests of individual member states;

• unlike the Stability and Growth Pact, for example, macro-dialogue is lacking in potential sanctions; and

• the policy mix required to achieve convergence in a particular member state may differ from that demanded by cyclical conditions in Euroland’s core countries.

The Commission has reacted to this last point with extraordinary speed, by devising a strengthened mechanism to coordinate budgetary policy within Euroland on an ex ante basis (see Commission document KOM(2001) 82). This is intended to ensure that criteria established in economic policy guidelines are observed in national budgets. Yet, astonishingly, the Commission document concerned makes no mention of the Cologne process, which could clearly provide the context for such coordination. Instead, it recommends that a parallel institutional structure be set up.[10]

So far as can be judged, EU macro-dialogue as yet barely extends beyond information exchange. Certainly, nothing exists that could be described as effective coordination of the various macroeconomic actors, the Ecofin Council, the ECB and social dialogue. Indeed, even coordination within a single policy area (i.e. budgetary policy in Euroland) seems impossible to achieve. Whether this is due to lack of willingness on the actors’ part - has the „window of opportunity“ already closed? - or to the institutional arrangements for implementing the Cologne Process is the final point to be examined here.

Is There a Chance for EU Macro-Dialogue?

(Political actors who think in terms of electoral tactics currently expect a “Keynesian” strategy to arouse more opposition than support.)

The original aim of macro-dialogue was to coordinate various policy areas, and thereby to create a favorable macroeconomic climate in which the deepening of European integration (the Cardiff Process) could be eased and labor market intervention (the Luxembourg Process) made more effective. That was reminiscent of the German „concerted action“ of the late 1960s, which provided the macroeconomic context for the 1969 Employment Promotion Act. The ultimate failure of „concerted action“ was attributable in part to the lack of appropriate institutional arrangements but, above all, to the incompatibility of coordinated behavior with the policy rules to which the actors concerned were subject.[11] Consequently, the macro-dialogue originally envisaged degenerated imperceptibly into a form of „antagonistic cooperation“ in the field of incomes policy. The outcome was little more than empty phrases.

EU macro-dialogue too has three weaknesses which, unless its procedures are changed, will prevent effective coordination, and thus the creation in Euroland of a climate more favorable for growth and employment.

• The division of macro-dialogue into technical and political levels has so far impeded creation of an institution which would draw up binding, generally accepted policy rules, and enable the various policy actors’ behavior to be monitored. In order to become such an institution, the technical level would have to be markedly upgraded and its subordinate remit correspondingly extended. Only then could it form the basis of a European „Expert Committee“.

• Macro-dialogue was established as a reaction to the Stability and Growth Pact, and – it may be surmised – as a concession in the light of its terms. Yet the Pact is based on an understanding of how economic policy responsibilities are assigned that is incompatible with cooperative economic policy, in the sense of horizontal coordination designed to create a positive-sum game. Precisely in order to render coordination unnecessary, but also due to a fear that the ECB might be „exploited“ by the budgetary policy of (some) member states, budgets were made subject to severe restrictions that make no economic sense - and which derive from a world-view based on assignment rather than coordination. Thus if macro-dialogue is really to generate a strategy for which all policy actors are jointly responsible, designed to maximize Euroland’s employment potential without endangering price stability, then there is a pressing need for the Pact to be reinterpreted. Initial proposals in that direction have already been made.[12]

• Finally, it should not be forgotten that, as yet, only in the monetary policy field does a an actor genuinely capable of decisive action exist at EU level; the ECB. For budgetary and, above all, incomes policy, such actors are lacking. Macro-dialogue would therefore require multi-level coordination between the various policy areas, on the one hand, and the relevant national actors in the fields of budgetary and incomes policy on the other (see Figure 2). As a result it is in danger of falling into Scharpf’s notorious „interwoven policies trap“ (Politikverflechtungsfalle).[13] Without doubt EU macro-dialogue could be more easily initiated, and later consolidated, if it could build on national macro-dialogues at member state level. For then it could be restricted essentially to coordination within the policy fields and to providing feedback for the various national dialogues, with common monetary policy providing the necessary binding element. Unfortunately, though, the establishment of national dialogues would itself require a spill-over process from the EU level, as experienced during the discussion on monetary and budgetary policy in the run-up to EMU.