Implementing a Voluntary Wage Policy: Lessons from the Irish and Spanish Wages Policies before the Crisis

Despite the relevance given by the Post-Keynesian thought to wage and incomes policies, little attention has been paid to the institutional elements that would favour the unionsâ€™ acceptance of a voluntary moderation of wage claims. Recent wage policies have been implemented in European countries, like Ireland and Spain, which do not fulfil the requirements assumed by corporatist analysis for a successful implementation of wage policies. The success of wage policies in Ireland and Spain, in terms of economic performance and the length of current wage policies, offers a valuable insight on how wages policies can be implemented as a key piece of macroeconomic policy: It also helps our understanding of the institutional framework that favours the implementation of voluntary wages policies.

economic policy.Instruments are independent of the political conflicts and the strategies adopted by the agents, and their implementation is explained by their effectiveness to solve economic problems.
The implementation of a voluntary wage/incomes policy is a strategic decision made by the agents involved.This decision is the outcome of the valuation made by unions, employers and the government of the consequences of incomes policies on all the agents.Therefore, the analysis of incomes policies requires the study of their costs and benefits for all involved agents.Our paper focuses only on trade unions, paying attention to the conditions that must be fulfilled for unions to accept a voluntary wage policy.
The traditional view of wage policies is based on three main assumptions: i) a powerful trade union movement with close ties to a social-democratic administration; ii) a negotiated strategy of macroeconomic policy; iii) higher social expenditures as a compensation to wage moderation.In this framework, implementing an orthodox economic policy will lead to the giving up of incomes policy.However, in the late eighties and nineties, experiences of a new incomes policy emerged in economies, like Ireland and Spain, that did not have the prerequisites of the classic corporatist societies (that is, small open countries with a strong trade union movement, like Austria, Belgium, Netherlands or the Scandinavian countries, and big countries with a strong and centralized trade union movement and with social-democratic administrations, like Germany and the United Kingdom) where incomes policies were traditionally implemented (Giuseppe Fajertag andPhilippe Pochet 1997, 2000;Peter A. Hall and David Soskice 2001;Lucio Baccaro 2003) and that, moreover, implemented voluntary wages policy that were subsequently rejected by trade unions.Table 1 shows the economic performance of Ireland and Spain.In Ireland we show the data for three periods: the first one comprises the first experience with wage policies in the seventies; the second (1981)(1982)(1983)(1984)(1985)(1986) is the period without wage policies, and the last one, starting in 1987 corresponds with the current phase of social partnership.In Spain, we have four periods: in the first one there is no voluntary wage policy; the second one (1977)(1978)(1979)(1980)(1981)(1982)(1983)(1984)(1985)(1986) is the period of the voluntary wage policies; the third one is the period without voluntary wage moderation, and the final period (1997)(1998)(1999)(2000)(2001)(2002)(2003)(2004)(2005)(2006)(2007) is the new period of a voluntary wage policy starting with the signature of the first Interconfederal Agreement of Collective Bargaining.
A common feature of both countries is that the first experiences of wage policies in the seventies and eighties did not generate the expected macroeconomic outcomes.Although wage moderation (not too intense, as the growth of real wages shows) cut inflation rates; inflation remained at high levels and the labour market performance was not satisfactory.The giving-up of wage policies, however, did not perform better: inflation rates did fall, but the results in terms of economic growth and, mainly, employment and unemployment were really unsatisfactory.On the contrary, the phase of voluntary wage moderation since the late eighties in Ireland and since the late nineties in Spain has come, until the onset of the onset of the crisis in 2008, in terms of economic growth and employment creation.This long expansion phase came with a clear moderation on wage claims, a moderation that in the Spanish case led to nil growth of real wages in the last decade.
As mentioned, incomes and wages policies are a key tool in the Post Keynesian strategy of macroeconomic management.Moreover, in the framework of the European Monetary Union, the current constraints to a discretionary management of the fiscal and monetary policies, the resource to an alternative policy, like wages policy can be an useful way to help to solve the macroeconomic imbalances of most Member States (high unemployment rates, current account imbalances) without having an impact on the inflation rates.Thus, recent papers like Engelbert Stockhammer and Özlem Onaran (2012) or Nina Ponivkar, Maks Tajnikar, and Petra Došenpović (2013) defend the need of a voluntary wage policy in the EMU.In this sense, the analysis of the Spanish and Irish experiences with voluntary wages policy can help to understand the conditions that must be fulfilled for an effective implementation of these policies.
It is important to emphasize that the analysis carried out in the paper focus only in the period before the onset of the crisis in 20081 .The reason of this choice is that the paper wants to analyze the conditions that lead workers and trade unions to adopt a voluntary exercise of moderation in their wage claims during a long period of economic expansion.As a result of this exercise, workers do not fully use the bargaining power resulting from the economic growth and the improvement s in the employment and unemployment.As we will show later, the acceptance of a voluntary wage moderation during a crisis can be the result of a defensive strategy adopted with the aim to avoid the costs for workers coming from a fall in the employment and/or the implementation of measures like reforms in the unemployment protection schemes, in the modalities of hiring, the firing costs, etc.
In sum, the aim of the paper is not to analyse the impact of wage policies on the economic performance of the economies in question, Ireland and Spain: The focus of the analysis, however, is on the institutional environment in which voluntary wage moderation can be agreed between employers and trade unions.The paper is structured as follows.The next section analyzes the role to be played by incomes-wage policies in a Post Keynesian theoretical framework.Section 2 analyzes the benefits and costs for unions of voluntary wage policies and the need of compensations for the union acceptance of those policies, respectively.Section 3 studies the experiences of wages policies in Ireland and Spain before the crisis.Last section summarizes and concludes.

Wage Policies in the Post Keynesian Thought
For many Post Keynesian authors, wage-incomes policies are a key element of any strategy of macroeconomic policy that tries to reach full employment without inflationary pressures (Henry C. Wallich and Sidney Weintraub 1971;Weintraub 1972;Alfred S. Eichner 1979;Joan Robinson and Francis Cripps 1979;Laurence S. Seidman 1979, 1981;John Cornwall 1983;Giovanni Caravale 1998;Marcella Corsi and Alessandro Roncaglia 2002;Paul Davidson 2002, 2011;John Marangos 2003;Gilberto T. Lima 2004;Mark Setterfield 2006a, 2007;Lima and Setterfield 2008;Geoff Harcourt 2010;Eckhard Hein and Stockhammer 2011;Louis-Philippe Rochon and Setterfield 2012;Philip Arestis and Malcolm Sawyer 2013).In this approach, the objective of incomes policy is to avoid inflation problems arising from the conflicting-claims of employers (profits) and workers (wages).In the conflicting-claims model of inflation, inflation is the result of the struggle between workers and firms to increase their respective incomes shares (Weintraub 1958;Setterfield 2006a;Davidson 2011).Incomes policy would be required in situations of full employment or in recessions if expansionary demand-side policies must be implemented to solve mass unemployment problems without generating inflationary pressures (Hein 2002).
Since wage policies are presented as a component of a consensual strategy of economic policy, it is taken for granted that the objectives and the other instruments of macroeconomic policy are also bargained among the government and the social agents.However, with a few exceptions like Cornwall (1983) or Lima (2004), there is no explicit mention to the institutional framework in which the wage setting process takes place or to the institutional requirements that must be met to guarantee a lasting working of this policy.Some Post Keynesian authors identify wage or incomes policy with a taxbased incomes policy.In this policy, the authorities set a guideline for wage growth that if not fulfilled leads to imposing a tax on corporate profits, thus modifying the bargaining behavior of employers and workers.This kind of incomes policy is close to a compulsory wage policy with wage guidelines set by the public authority.But, though it has been the case in some cases, wage policies have been accepted voluntarily by workers and trade unions in most cases.
One weakness of the Post Keynesian analysis of incomes policies is the vague definition of this concept.For Lima and Setterfield "incomes policies are defined as formal or informal institutions that frame and mediate aggregate wage-and pricesetting behavior in such a way as to reduce conflict over incomes shares and better reconcile conflicting income claims" (Lima and Setterfield 2008, p. 438).Thus, any measure affecting the workers' capacity or willingness to claim and reach higher wage claims would be considered as a wage policy.Under the umbrella of "incomes policies" we could find measures to modify the structure of collective bargaining, to reduce labor turnover costs, or any labor market reform.In sum, any measure directed to change the income distribution (wage and profits share) would be an incomes policy.Moreover, and not least relevant, for Lima and Setterfield (2008), incomes policies can be cooperative or coercitive, depending on the distribution of power in a society.
Arestis and Sawyer see incomes policy as an "inflation policy", where "the focus of incomes policy (which may be more appropriately labeled inflation policy) should be on prices rather than on wages directly" (Arestis and Sawyer 2013).Here it is not clear whether incomes policy involves measures and guidelines regarding on prices growth, and it is neither clear the role to be played by wages in incomes policy.Moreover, by focusing only on inflation, it would not include in the definition of incomes policy, those measures focused on wages evolution whose main objective be different from inflation control, like employment or external balances.
Although it is commonly argued that Post Keynesian incomes policies (i.e.tax-based incomes policies) have not been implemented but substituted instead by monetarist incomes policies of fear (Davidson 2006;Setterfield 2006a, b), voluntary wage policies, resulting from the workers' willingness to reduce their desire to act on their bargaining power accepting a social bargain of their wage claims (Setterfield and Ted Lovejoy 2006), have been implemented in many cases.Trade unions have voluntarily accepted a moderation in wage claims, with the objective of reaching a specific target of some economic objective (employment growth, unemployment rate, inflation rate, etc.).Wage policy is, thus, embedded in a social pact, where "the acceptance of a moderate pace of increase of money wage rates on the side of the trade unions in exchange, such as for policies directed to increasing employment or fiscal policies directed to reducing the weight of taxation on dependent labor" (Corsi and Roncaglia 2002, p. 153).Wage policies become a tool of macroeconomic policy bargained at a bipartite (trade unions-employers, or trade unions-government) or tripartite (trade unions-employers-government) level.
In this paper, we adopt a more restrictive definition of incomes policies, focused on voluntary wage policies: trade unions voluntarily moderate wage claims with the objective of reaching some macroeconomic economic objectives.Thus, "(voluntary incomes) policies can be viewed as a method of coordinating collective bargaining decisions with government macroeconomic policy" (Cornwall 1983, p. 248).Voluntary or cooperative wage policies involve a self-moderation in the use of workers' bargaining power.This is a key difference with other institutional reforms that influence wage claims thereby reducing the workers' bargaining power, like, for instance, reducing firing costs, the promotion of individual wage bargaining, the cuts in unemployment compensations, etc.These reforms permanently reduce workers' bargaining power moving permanently the Philips curve to the left.However, in a voluntary wage policy workers' bargaining power remains constant, and, therefore, workers accept a less intensive use of their bargaining power.Although a voluntary wage policy reduces the workers' willingness to act on their bargaining power it does not reduce the extent of workers' bargaining power.Therefore, if wage moderation disappears, higher wage claims can move the Philips curve to the right leading to PANOECONOMICUS, 2014, 1, Special Issue, pp.107-127 potential inflation pressures.In sum, wage policies move temporarily the Philips curve to the left.
Thus, to implement an effective cooperative wage policy there must be an institutional environment that favours the setting in motion of wage moderation, and also keeping it in a permanent state.And as Robert Pollin and Andong Zhu argue: "The most basic critique of incomes policy is that, in order for the approach to have any chance of success, it is necessary that a country operates with a high level of organization among workers, and that there be some reasonable degree of common ground between workers and business.Otherwise, there will be no realistic prospects for economy-wide bargaining to yield results that will be honored widely" (Robert Pollin and Andong Zhu 2006, p. 610).This is the focus of our paper: to analyse the institutional elements that favour the acceptance by workers of a long-lasting voluntary moderation of their wage claims.This "bargaining climate", that is, "the broader set of institutions that establishes the environment for relations between labour and capital" (Cornwall and Wendy Cornwall 2001, p. 82), is the main determinant of the union's acceptance of a voluntary wages policy, in particular, or of any kind of price and incomes policy.In this sense, Harcourt defends the need of a permanent income policy, in the form of "institutions that as a start allow money incomes to be adjusted for prices and overall productivity (reflecting the terms of trade in open economies)" (Harcourt 2010, p. 245) and Arestis and Sawyer also argue that the "idea of a prices and incomes policy would be to have institutional arrangements designed to restrain the rate of increase of prices and wages to be compatible with a target rate of inflation without resort to deflationary measures" (Arestis and Sawyer 2013, p. 10).In the next two sections we analyse how this institutional environment can be generated and sustained.Next, we analyse the recent Irish and Spanish experiences with voluntary wage policies.It is commonly argued that both Ireland and Spain do not fit the features of corporatist societies.However, despite the failure of earlier experiences with incomes policies in the seventies and early eighties, since the nineties both countries have de facto implemented a voluntary moderation in wage claims that have led to excellent macroeconomic outcomes.

Benefits and Costs of Wage Policies for Trade Unions: The Need of Compensations
The implementation of voluntary wage moderation involves that agents make an exante valuation of the effects of that policy on their individual objectives.This strategic analysis is missed in most studies on incomes policies, where it is assumed that the environment, defined by a set of economic (economic performance), political (ideological adscription of the political party in the government), social (corporatist structures) or institutional (centralized structure of trade unions) elements, is the main determinant of the implementation of incomes policies.One example is the hypothesis that a centralized union movement favors the existence of a voluntary wage policy: centralization makes trade union leaders aware of the macroeconomic negative impact (higher inflation and unemployment) from excessive wage claims, thus lead-ing to a lower wage growth.Thus, from a Post Keynesian view, Cornwall (1983) argues that a centralized trade union movement facilitates workers' participation to a voluntary wage policy because unions internalise the negative impact of excessive wage demands.However, this is in contradiction with his statement that in the postwar period, governments have necessitated concessions to labour for getting the unions' support to wage policies.If this internalisation exists, concessions are not needed.These exist because, first, unions do not actually internalise those effects, and/or, second, because despite this internalisation, unions still evaluate the acceptance of a wage policy at a net cost.
These studies, therefore, do not analyze the choices that agents make among the set of strategies and tactics potentially adaptable, or the causation between the environment and the adopted/rejected strategies: an instrument is implemented because of its effectiveness to solve macroeconomic shocks.Instruments are technical, politically neutral, that is, independent of the conflicts and the strategies adopted by the agents to manage conflicts.However, in the real world the choice of an instrument is not only made according to efficiency-effectiveness criteria, but also to the political strategies of the agents involved, whose aim is to influence the distribution of power resources among the agents (Walter Korpi 1983).
The implementation of a voluntary wage/incomes policy is the outcome of the comparison between the consequences generated by an incomes policy with alternative measures (Ferreiro and Gómez 2008).Therefore, the analysis of incomes policies requires the study of the costs and benefits of its acceptance by workers, employers and government.In a voluntary wage policy, an analysis of the costs and benefits of wage moderation for workers and unions is essential to explain its acceptance or rejection.This structure of incentives arises as a result of a certain institutional framework that guarantees the existence and the maintenance of the benefits that allow workers to commit themselves to a wage moderation policy.We must keep in mind that a wage policy is not just a wage moderation but an institutionalised moderation with the explicit aim to reach some macroeconomic objective.In a neoclassical world, the flexibility of labour markets leads to internalising the consequences from excessive wage increases, thereby making unnecessary any further incentives for workers to accept right wage increases.However, in the real world, this does not apply.As we will see below, wage policies are a net cost both in the short run and in the long run for workers and only the existence of compensations guarantee the acceptance of wage moderation.But, these incentives will only be generated only with an appropriate institutional environment.
Benefits to unions of a voluntary wage policy are different from those reaped by the government and the employers.First, because these benefits are contingent and, second, because there exists a lag between the time the wage policy agreement is signed and the time the benefits are reaped by the workers (Nico A. Siegel 2005;Ferreiro and Gómez 2008).The first benefit is the macroeconomic impact of wage moderation, mainly lower unemployment and inflation.The problem is the uncertainty about the achievement of these benefits and the time they will be effective.In the short run, wage moderation increases business profits and redistributes national income in favor of capital incomes.The impact on inflation depends on prices reaction to lower wage costs, and the impact on employment depends on the extent that higher profits lead to higher investment and to the use of the higher productive capacity.
In this sense, Arestis and Sawyer (2013) argue that inflation (and we could also say that unemployment) is a multi-causal phenomenon, with its sources varying over time and economies.Therefore, the uncertainty about the causes of the macroeconomic disequilibria and the impact of the consequent economic policies is a problem for voluntary acceptance of these measures by affected agents.Thus, in the case of the voluntary wages policies, workers are accepting a voluntary moderation of wage growth (even a fall in real wages) without having the guarantee that the expected benefits of wage moderation (lower inflation and unemployment rates) be reached.
But unions can also get political benefits.Wage policies can be implemented to underpin an administration of a political party close to the union in terms of ideological affinity or institutional ties.This is the basis of the hypothesis that social democratic governments favor the implementation of incomes policies (Cornwall 1983;Korpi 1983;Gosta Esping-Andersen 1985;Nils Elvander 1990).Wage policies can also enlarge the traditional union influence arena (the firm) to the political arena, influencing the choice of the instruments and objectives of economic policy.
Trade unions can also accept wage policies for defensive reasons, that is, to avoid other instruments of economic policy with higher negative effects on their interests.For example, flexibility of the labor market, reforms in the collective bargaining system promoting decentralized-individual wage bargaining, or a deflationary economic policy.Here, unions accept wage moderation to maintain their current power resources.
Finally, although for many studies a centralized union movement is a prerequisite for wage/incomes policies, a fragmented union movement can also favor social pacts.When there are several trade unions, they can compete amongst them to increase their respective power and influence.Also, participation in social pacts can be evaluated by a union as a strategic tool to increase its individual power and reducing that of rivals.
But wage policies also involve costs to trade unions.A voluntary moderation in wage claims means a lower bargaining intensity in the wage-setting process.This changes the primary role of union, now working as a guarantor of macroeconomic stability.The new role of unions can involve a political cost if workers consider that their union stops working as an instrument to improve their economic situation and welfare, and workers can leave the particular union or affiliate other unions that reject wage policies.
Negative past experience with wage policies are also a source of cost.A past failure with the macroeconomic targets or the compensations offered to unions removes the economic benefits expected from future wage policies (Andrew Martin 1997).Therefore, the trade unions acceptance of wage-claim moderation will be related to their estimation of the degree of compromise of employers associations and the government with the fulfillment of the content of the agreement.
Another cost is the asymmetric distribution of costs and benefits: wage policies are a net cost for workers but a net benefit for the government and the employ-ers.For the latter, benefits (higher business profits, lower labor disputes, legitimation and increased effectiveness of the economic policy) are higher than costs (the nonimplementation of other tools).Consequently, the wage policy redistributes the economic and political power resources in favor of the employers and the government, thus making workers reluctant to implement a "permanent" incomes policy (Cornwall 1977;Davidson J. Alexander 1989;Ferreiro and Gómez 2008).
In sum, the costs for unions from an incomes policy can be grouped in two categories: economic and political costs.In the first category we include all the elements that have a negative effect on wage incomes, either on its absolute value or on their growth.The second category includes those elements that negatively affect the unions' power resource, reducing their capacity to influence the behavior and the decisions of the rest of the agents involved.
It is the case, of course, that unions are a tool of collective action.Collective action creates power resources and bargaining capacities above the sum of individual workers' power.Any element that damages collective action will reduce the workers' bargaining power and their influence on the decision-making of employers and government.Wage policies affect negatively collective action: uncertain benefits along with high costs make wage policy a net cost for unions.Workers can lose the incentives for collective action moving to a micro action (at firm or individual level) if they think that their interests can be better defended.Although the change can be rational from a short-term perspective, it is negative in the long run, because it reduces the resources power and the bargaining capacity of all workers.
In a bargaining process the choice of a specific tactic is determined by the subjective valuation of the bargaining power of all the agents involved in the bargaining process.The outcomes of the bargaining process can affect the future distribution of the bargaining power of the agents.Thus, if a voluntary wage policy is a net cost for workers and unions, they will only accept it if there is some compensation or if the net costs of alternative measures are even higher.These compensations can come from an economic exchange where workers demand economic compensations to offset the wage moderation; or, from a political exchange to offset the negative effect generated on the workers' power resources.This double exchange balances the distribution of the effects from wage policies.
Unions can direct these claims to employers and the government.The claims to employers have a "political" and procedural nature, claiming representation and organization privileges (Franz Traxler 1997).The objective is to increase the union's influence in the strategic decisions of the firm, offsetting the lower influence in the wage setting process (Colin Crouch 1982).Economic claims would involve higher costs for firms that would remove the benefits of wage moderation.Consequently, the exchange in the firm arena should be between wage moderation and political compensations.
Claims to government have a mixed nature.There can be an economic exchange, with unions claiming measures to increase the workers' welfare.Higher social public expenditures or tax cuts would be offered in exchange of wage moderation (Alexander 1989).However, due to the impact of these compensations on the public finances (lower taxes and/or higher expenditures), depending on the economic and fiscal situation, governments might be reluctant to provide this type of compensations.If the governments consider that these compensations are a source of disturbances, they have a strong incentive to reject wage policies.Governments would prefer alternative measures that can reduce wage growth with lower costs, like labor market reforms.There can also be a political exchange with compensations taking the form of higher unions' influence in the public, political, and economic arenas.The objective is to influence the public decision-making process, making the economic policy more favorable to workers' interests.The ultimate aim would be to identify workers' interests along with the general interests.That identification would remove the above-mentioned problem of the dual function of trade unions in modern capitalist societies and their consequent problems.
Compensations can also be directed to employers and government.Some examples are price-and profit-dividend control policies.Although employers and capital-earners are the agents directly affected, this claim is directed to the government to warrant their fulfillment.This compensation is demanded to avoid wages being the only incomes affected by the economic adjustment needed to correct the macroeconomic imbalances.This would make much easier the acceptance of wage policy (Elvander 1990).
Cornwall separated voluntary incomes policies into two types: "one general class in which the main economic groups to comply with wage and price norms or guidelines by measures that lead them to internalize the costs of their wage and price behaviour: these will be referred to as consensus policies.The second class of voluntary incomes policies involves legislative and other measures that provide incentives in the form of financial rewards or penalties or both that are directly related to compliance or non-compliance with wage or price norms.These will be referred to as incentive policies" (Cornwall 1983, pp. 249-250).In our opinion, the problem with this classification is that a lasting implementation of "consensus policies" is not possible because wages policies are a net cost for workers.In this type of incomes policies the only benefits for workers are those resulting from the better macroeconomic performance.However, and as explained above, these benefits, at best, can only be reaped in the long-run; and there always exist the possibility of an asymmetric distribution of these benefits between workers and business.Consequently, a permanent incomes policy can only be of an "incentive policies" type.
Therefore, a necessary condition for a permanent wage policy is an institutional framework that generates the compensation required for workers supporting permanent wage moderation.For Cornwall and Cornwall (2005), institutions (defined as the beliefs, customs, laws, rules and norms that guide the behavior of individuals and groups within society), fulfill two functions: legitimising power and providing mechanisms for conflict mediation.Once that an institution legitimises a power relationship, it simultaneously legitimises the economic outcomes (including income distribution) flowing from it.If we assume that power and conflict are inherent features of a market economy, the conflict concerned with price-wage setting can only be reconciled through a proper institutional environment that makes wage claims compatible with a certain macroeconomic outcome.Even more relevant, it is outcome that generates the incentives needed to offset the unbalanced impact of voluntary wages policies.

Lessons from Voluntary Wage Policies in Ireland and Spain
For David Natali and Philippe Pochet the new social pacts signed in some EU countries in the eighties and nineties "sought to accommodate the uncertainties associated with a difficult and unpredictable economic climate" (David Natali and Pochet 2009, p. 154).The new economic environment increased the uncertainty related to the implementation and effectiveness of the policy measures to solve the main economic problems of the Irish and Spanish economies: low competitiveness, inflation and mass unemployment.
The origin of this uncertainty can be found in the failure of previous measures implemented in the eighties.In the case of macroeconomic policies, monetary policies led in both economies to problems with the exchange rate.This led, first, to an appreciation of the currencies (thus affecting external competitiveness) and, later, to a strong devaluation of the Spanish peseta and the Irish pound with the crisis of the EMS in the early nineties.Fiscal policies also generated high public deficits, what had to be reverted through tight fiscal policies mainly based on expenditures cuts.In the case of supply-side or structural policies, past experiences with labour market reforms proved not only ineffective but also generating of collateral negative effects.These reforms focused in Ireland on the collective bargaining structure (promoting a move to a decentralization of collective bargaining) and in Spain on the legal regulation of employment contracts (promoting the use of fixed-term employment contracts).
Therefore, in Spain and Ireland both government authorities and social agents were uncertain about which economic policy strategy could effectively helped to solve the deep macroeconomic problems suffered by the two economies.This uncertainty was also fuelled by the new economic environment, which arose from the construction of the European monetary union.Maastricht Treaty involved from its very beginning the implementation of an orthodox strategy of macroeconomic policy not only during the transition phase but also once the EMU was created (as the Statutes of the ECB of the Stability and Growth Pact prove).Moreover, the EMU posed a new constraint: the removal of the exchange rate as an instrument to offset the negative impact from higher inflation than that registered in the other European economies.
Neither Irish nor Spanish trade unions showed a frontal rejection of this new economic and institutional environment.Actually, they saw the EMU not as a threat but as a challenge or opportunity.The reasons can be that they did not have an alternative and, also, because their respective weaknesses excluded the possibility of substantially reforming that new framework (a weakness also extensive to employers and, even, to public authorities).Nonetheless, although the introduction of a new macroeconomic constraint, the EMU one, which is a clear incentive for voluntary wage moderation (Anke Hassel 2006;Maarten Keune 2008;Natali and Pochet 2009).However, it does not explain why this wage moderation is institutionalized and not market-determined.If the argument was right and the removal of the exchange rate and the implementation of a deflationary strategy of macroeconomic policy make more clear for employers and unions the economic negative impact of excessive wage growth, then the internalisation of the effects of wage setting would be auto-PANOECONOMICUS, 2014, 1, Special Issue, pp.107-127 matic Therefore, an institutionalized framework of wage setting would not be necessary.Actually, there would be no need of compensations to wage moderation Thus, the EMU may have acted as a necessary condition for accepting a voluntary wage policy but not as a sufficient one.
Contrary to expectations, failure of past wage policies did not act as a disincentive for a new phase of social concentration.Ireland and Spain made an intense use of wage policies and social pacts in the seventies and early eighties (Fajertag andPochet 1997, 2000;Ferreiro andGómez 2006, 2008;Hassel 2006).In Ireland, during the seventies seven bipartite agreements of wage setting were signed between trade unions (ICTU) and employers associations (FUE), and in 1979 and 1980 two new tripartite agreements were also signed.In Spain, two bipartite agreements (trade unions and employers) in 1979 and 1983 were signed, and two tripartite agreements were approved in 1981 and 1984.
However, Irish and Spanish trade unions rejected during a decade the use of a voluntary wage policy.The reason is that in both countries trade unions considered past experience with wage policies a net cost.First, wage moderation did not reach the expected outcomes in inflation and unemployment.Second, compensations in terms of higher public expenditures disappeared when both governments had to implement tight fiscal policies.This was necessary as part of their joining to EMU in order to reduce the "excessive" levels of public deficits and stocks of public debt.Third, past agreements were not supported by the whole trade union movement, but only by some unions.Therefore, the acceptance-rejection of wages policies were mainly influenced by the struggle among individual trade unions.This struggle among trade unions was more relevant in Spain that in Ireland, because in Spain the rejection of a wage policy was not focused on some (industry) unions integrated in the trade union confederation, but whole union confederations, like Comisiones Obreras (CCOO), and actually onlye the socialist trade union, Unión General de Trabajadores (UGT) signed all incomes policies agreements in the decades of seventies and eighties.Finally, the implementation of a restrictive strategy of fiscal and monetary policies in both countries were fully rejected by the trade unions; thereby closing the door to a new experience of voluntary wage moderation as a component of a restrictive macroeconomic policy.
Despite past experiences, the response of trade unions in Ireland and Spain to the new economic situation was a change to a co-operative strategy based on dialogue and social concertation.With the explicit (Ireland) and implicit (Spain) support and/or intervention of public authorities, trade unions entered into a strategy of permanent dialogue on specific matters on which the measures that emerged from social dialogue could be easily implemented at the lowest (political) cost.Those measures related to collective bargaining, wage setting processes and labour market reforms.Thus, in Ireland in 1987 the model known as "social partnership" began with the signature of the Programme for Nacional Recovery, leading to eight social partnership programmes, the last one, "Towards 2016.Review and Transitional Agreement 2008-2009" signed in 2008.In Spain, the signature in 1997 of the first Acuerdo Interconfederal para la Negociación Colectiva (AINC) (Interconfederal Agreement of Collective Bargaining) opens the process of social dialogue and concertation, agree-PANOECONOMICUS, 2014, 1, Special Issue, pp.107-127 ments that were subsequently uninterruptedly signed or renewed until 2008, and that following to Oscar Molina and Martin Rhodes (2011) can be seen as an implicit incomes policy better than a series of ad hoc wage agreements.In both cases, wage policies, in the form of voluntary wage moderation with the explicit aim of reducing the rates of inflation and accelerating the employment creation, are a key component of this new strategy of economic policy.
Public authorities accepted as their own responsibility the creation of an environment that could favour the dialogue between social agents.This institutional environments is formed by three elements: the acceptance of the agreements signed by trade unions and employers; the commitment to not adopting measures in the matters of social dialogue without the previous consensus with social agents; and the awarding of economic compensations in exchange of the voluntary wage moderation.These compensations took the form of social expenditures and cuts in income tax.In the case of social expenditures, the commitment was not so much to increase their size (due to the adoption of a tight fiscal policy), but to maintain their current levels.Thus, the idea was that the fiscal adjustment required to reduce fiscal imbalances would not involve a fall of this spending.Consequently, there is a double exchange to offset the cost for unions of voluntary wage moderation.First, a political exchange, where some matters, mainly those related to the labour market regulation, are kept in the hands of social agents; and with public authorities refusing to adopt measures without the previous consensus of employers and workers.Second, an economic exchange where the maintenance of social expenditures and cuts in personal income taxes was offered to increase workers' disposable income.
In our opinion, the political exchange was more relevant than the economic one.Voluntary wage moderation and the guidelines of wage growth are, though important, one of the elements that form the content of social dialogue and concertation.Trade unions exchange wage moderation for policy and institutional influence (Michael Doherty 2011).Public authorities accept the basic principle of the autonomy of the social agents to make a decision about the changes and, even, the legal measures to be adopted in certain matters, mainly in those related to the labour market.The latter include social protection for workers, such as pensions, unemployment compensation schemes, employment contracts, structure and content of collective bargaining, and so on.The basic pattern of behaviour is that trade unions and employers decide about these matters, and the agreements reached being later ratified by government.In the case of tripartite bargaining, the government's commitment is to avoid bipartite bargaining, either with trade unions or with employers.The range of bargained matters has been extended.In Spain more than 40 bipartite agreements have been signed since 1994, and in Ireland, the last partnership programme Towards 2016 has a length of 141 pages.Here it is important to emphasize a difference between Spain and Ireland.In the Irish case the political exchange is embedded in the process of wage bargaining.Thus, the different partnership programmes include both the guidelines of wage growth and the different matters bargained in parallel to the wage growth approved.Issues included in the social partnership programmes are among others, fiscal policy, taxation, unemployment, monetary union, enterpriselevel partnership, welfare, social exclusion, literacy, drugs, etc. (Rory O'Donnell, PANOECONOMICUS, 2014, 1, Special Issue, pp.107-127 Maura Adshead, and Damian Thomas 2011).On the contrary, in Spain, the subsequent interconferederal agreements on collective bargaining focus on the matters of wage growth and the content of the collective bargaining.The political exchange in terms of the social protection policies, the reforms of the labour market, etc., are object of discussion and bargaining in other forums and agreements.
The main reasons for this political exchange are the own weakness of social agents and the public authorities (O'Donnell 2005;Ferreiro and Gómez 2008).In this sense, the political exchange reduces the uncertainty resulting from the incapacity of all the agents to impose their views.In Ireland wage policies were substituted in the early eighties by a return to a decentralized collective bargaining.However, the decentralisation in collective bargaining did not lead to the desired moderation of the real wages growth; and, though, inflation rates declined the rise in the real wage growth led to a huge increase in unemployment rates (O'Donnell and Colm O'Reardon 2002;O'Donnell 2005).In Spain, wage policies were substituted by a reform in the labour market that promoted the use of fixed-term employment contracts.This reform did not affect the wages of permanent workers.It led to a significant increase of temporary workers that generated related problems in terms of low productivity and a high pro-cyclical behaviour of labour market performance (Ferreiro and Serrano 2001).
Despite that the reforms that promoted decentralisation of collective bargaining and the external flexibilisation of labour market were against the workers' interests, the poor macroeconomic performance of those measures in both countries showed that the adoption of measures to introduce market mechanisms into wage formation was to a great extent offset by trade unions.Actually, as Table 1 showed, despite the high figures of unemployment in Ireland and in Spain, real wages kept increasing in positive terms.Public authorities could not implement measures with a frontal rejection from unions and employers.But trade unions (or employers) could not enter in a bilateral bargaining with the government that generated a result instantly rejected by employers (or trade unions).The only solution to this dilemma was to accept not a first-best (from the perspective of each agent) but a second-best solution: a bargained solution accepted by all parties.Although this solution was not necessarily the most effective one, it had the advantage that at least when implemented it generated positive outcomes.
The Irish and Spanish cases show that a powerful trade union movement is not a prerequisite for an effective implementation of wages policies.Actually, as in other European countries, the new phase of incomes policies in Ireland and Spain has taken place in a situation of weaker trade unions, at least in terms of union density (Hassel 2006).What is relevant to implementing a wage policy is a situation of balance of power among the involved parties.Actually, the acceptance of a wage policy in Ireland and Spain can be considered as a defensive strategy to avoid other kind of measures that could negatively affect the relative distribution of power of trade unions, employers and government.
The successful working of social pacts in Ireland and Spain can also explained by the general strategy of macroeconomic policy, which is not questioned by trade unions (Doherty 2011).Actually, and in opposition to the traditional view of incomes PANOECONOMICUS, 2014, 1, Special Issue, pp.107-127 policy, voluntary wage policy takes place with a restrictive macroeconomic policy.In the traditional view of Keynesian policies, the aim of wage policies is to avoid the inflationary pressures arising from expansionary demand-side policies.Wage policies could only take place with restrictive demand-side policies if the former were compulsory.On the contrary, in the Irish and Spanish cases, the fight against inflation is in the hands of demand-side policies, mainly monetary policies.The new role of wage policies is to avoid the negative impact of a restrictive strategy of macroeconomic policy on employment.Also, by accepting a low or even negative growth of real unit labour costs wages policy may contribute to the employment creation process.Obviously, this means an implicit acceptance of income redistribution against labour, with the result that the higher share of wages in national income is due to higher employment, not to real wages rising above productivity.By making employment the main objective of wage policies, the evaluation of the successfailure of voluntary wage policies will be made in terms of the employment creation process.
Moreover, since the main source of compensation to wage moderation does not come from higher (social) expenditures, the implementation of a tight fiscal (or monetary) policy, that is, a policy focused on the reduction of fiscal imbalances, is not detrimental to the acceptance of a voluntary wage moderation.
Thus, whilst in the traditional Keynesian view, wages policies were part of the "demand-side" policies, now in Ireland and Spain, they can be included in the catalogue of the "supply-side" policies.Actually, in both countries, wage moderation is part of more general and wider agreements (the social dialogue in Spain and the social partnership in Ireland): In these agreements, besides wage growth, social agents bargain different supply-side determinants of competitiveness, economic growth and employment creation Special attention is paid to the content and structure of collective bargaining (mechanisms of coordination, degree of centralization, contents of collective bargaining agreements) and aspects related to internal flexibility of workforce (mechanisms of flexibility, use of fixed-term employment contracts, etc.).

Summary and Conclusions
The recent experience of Ireland and Spain shows that voluntary wage policies can, in certain circumstances, work as an effective instrument of economic policy, leading to a good economic performance, at least, in terms of employment.However, contrary to the traditional view of these policies, the current objective has not been to reduce inflation but to avoid the negative impact of a restrictive strategy of macroeconomic policy on employment.This shows that wages policy can co-exist with a strategy of demand-side policies whose main objective is not full employment.
Although macroeconomic policy and wage bargaining must be articulated to reach the desired economic objectives, neither in Ireland or Spain is macroeconomic policy a matter of bargaining between trade unions and the government.Consequently, macroeconomic policy operates as a constraint accepted by trade unions in the wage setting process.Social agents, mainly workers, internalise in this new environment the negative effects of an excessive wage growth.However, the implementation of an orthodox macroeconomic policy does not imply that market forces set the growth of wages.On the contrary, wage bargaining remains institutionalized at the aggregate level.The reason is that, despite the weakness of Spanish and Irish trade unions, neither government or employers have enough power to impose a framework of wage setting in accordance with a process of labour flexibility, a deregulation leading to a full decentralisation of collective bargaining or, even, an individual wage setting process (single worker -employer).The generalized weakness of social agents and government generates a balance of power that implies that any measure not agreed has an uncertain outcome.Hence, the need of implementing minimum policies, that is, only those agreed by all the parties, But, voluntary wage moderation is a net cost for workers, mainly in situations, like those recent ones in Ireland and Spain, of economic expansions.For example, in Spain, wage policies do not lead to an increase in real wages.This makes that the internalisation of positive effects of wage moderation (in terms of employment creation) is not enough to guarantee the maintenance of wage policies, thereby requiring the introduction of compensations.Actually, the successful working of wage policies complicates their maintenance, because the very fall in unemployment rates removes the main compensation for wage moderation, which is the employment creation process.Once full employment is reached, workers will need another source of political-economic compensations to avoid undesirable wage increases that could lead to inflationary pressures.Although this is not still the case in Spain, in Ireland over the current decade the low unemployment rates have led to an acceleration in the growth of wages.More specifically, over the period 2004-2007 with an average unemployment rate of 4.5%, real wages rose by 3.4% each year on average.These compensations do not take the form of the traditional economic exchange between social expenditures and wage moderation, because the existence of tight fiscal policies excludes this exchange.Neither do they take the form of a political exchange between wage moderation and the design of the general strategy of macroeconomic policy (objectives, targets and instruments), simply because this strategy is given and it is non-bargainable; it merely works as a constraint.The political exchange comes now from the fact that labour issues are part of the range of matters bargained between trade unions and employers (and government): trade unions participate in the design of measures in this field and the government does not unilaterally adopt measures without the agreement of trade unions.
The lasting maintenance of wage policies in both countries is thus explained by the combination of this political exchange and the positive economic performance of this strategy of economic policy.In as much both elements remain, they involve the maintenance of the institutional environment that allows the generation of the needed compensations.The main elements that favour the acceptance by workers of voluntary wage moderation will help this policy to continue to be operated successfully in both countries.
It is important to notice that the recent crisis has put in difficulties the maintenance of the voluntary wage policies in both countries.In 2008, a new social partnership programme was approved, the "Towards 2016.In this sense, the more recent experience shows that Spain departs significantly from Ireland.The rejection to a voluntary wages policy in 2009 can be explained by the absence of compensations to a wage moderation demanded by the political authorities.However, in Spain, there was a fast return to the wage policy in 2010, a support to the wage moderation that can be explained as a defensive strategy by trade unions trying to reduce the impact of the crisis in terms of employment destruction and the implementation of radical reforms in the Spanish labour market.Like in previous years, the maintenance of this wage policy will depend on the future performance of the Spanish labour market.
Review and Transitional Agreement 2008-2009", with the objective of facing the economic and fiscal challenges that Ireland was suffering as result of the global financial crisis.However, the fiscal consolidation measures and the cut in public wages adopted in 2009 led to the suspension of the partnership.In a similar vein, in Spain the Acuerdo Interconfederal para la Negiación Colectiva of 2007 was extended to 2008.However, that agreement could not be reached for the year 2009.Nonetheless, in 2010 the employers associations and the trade unions UGT and CCOO reached a new agreement, the Acuerdo para el Empleo y la Negociación Colectiva (Agreement for the Employment and the Collective Bargaining), valid for the years 2010, 2011 and 2012.This agreement was renewed in 2012, the II Acuerdo para el Empleo y la Negociación Colectiva, valid for the years 2012, 2013 and 2014.

Table 1
Economic Performance of Ireland and Spain (%)