Labour markets and the crisis of the European Monetary Union

The single currency, ten years after

In 1999, the introduction of the Euro heralded the crowning achievement of post-war political and economic integration in Europe – at least for those who chose to partake. A single market was complemented by a single currency, which, in turn, formed the basis for a closer alignment of economic policies across the continent, and would ultimately lead to the formation of a pan-European social and political identity – the foundations for the European demos whose absence was decried by many political philosophers sympathetic to the European project. By 2009, ten years after the introduction of the euro and less than two years after the first signs of the crisis that enveloped the global financial system, the first cracks in this novel, unique, political-economic edifice began to show. Ireland’s prime minister had just mortgaged away the future of the country by underwriting the failure of the country’s banks in full, Portugal was facing a slow-motion crisis of confidence from financial markets and, perhaps most spectacularly of all, Greece was unable to roll over its sovereign debt without paying excessively high interest rates. In the months that followed, Spain, Italy, and, at some point, even France and Belgium, risked being dragged into the turmoil of the sovereign debt crisis, leading many observers to wonder publicly about the survival of the single currency.

A decade after its inception, therefore, the single currency faced an existential crisis, both political and economic. Governments of many – and by early 2014 possibly most – EMU member states had been rejected by their electorates, while most of their successors fared little better, facing massive protests and extremely low popularity ratings very early on in their terms. Populist parties both on the Left and the Right upset party systems in Italy, Greece, France, the Netherlands, and Belgium. A handful of euro-zone member states were forced to go, cap in hand, to the ECB, the European Commission and the IMF, begging for financial support. In practically every EMU member state, several banks that faced potential bankruptcy as a result of their exposure to shady private and public debt, had to be nationalized by their governments. Defying its own strict mandate the ECB all but promised unlimited support to the euro’s financial system in the summer of 2012 up to the point, it appeared, of effectively bailing out governments who were unable to borrow at reasonable rates in international capital markets. And in early 2013, the logical foundations of a single currency were shaken when Cyprus, an economy accounting for a tiny fraction of euro-zone GDP, introduced capital controls to stop a run on its banks: from then on a euro in a bank account in Cyprus was different from a euro in a German or French bank account.

Most observers, especially in policy-making circles in the Brussels-Frankfurt axis, think of the crisis of EMU primarily as a crisis of failed fiscal discipline in a monetary union. Once a country has become a member of EMU, its loose fiscal policy can no longer be checked by national or international institutions: the Stability and Growth Pact, EMU’s economic quasi-constitution, has turned out to be the paper tiger that many feared it was, and the result was fiscal incontinence everywhere. While a priori appealing, there is a serious problem with this argument: very few of the member states that have found themselves in fiscal problems after the crisis erupted in 2010 actually ran a primary deficit, and most did not even run a deficit beyond three per cent over the first nine years of the euro. Greece, held up as the most visible point of the iceberg, is the exception here, not the rule. Others see it primarily as a crisis of the financial sector that spilled over into the public sphere following the financial crisis of 2007-08, which exposed the weakness of a deregulated international financial system. As banks became weaker, governments were forced to take them over, thus adding to their debt burden. The upshot: banks that held sovereign bonds were weakened even more, thus producing a vicious spiral of weak banks and increasing government debt. That argument, though, does not help understand why countries such as the Netherlands and Belgium, with high private (NL) or public debt (BE) appear to be almost immune from the problems and pressures that southern European and other peripheral countries in EMU have faced since the onset of the crisis.

Different systems of employment and labour relations

There is a very different way of looking at this. The crisis of EMU since 2009 has laid bare problematic aspects of the interaction between employment relations, and in particular wage bargaining systems, on the one hand, and central banks on the other. Somewhat schematically, continental (western) Europe consists of two very different systems of employment and labour relations, roughly along the lines of how the Varieties of Capitalism approach to contemporary capitalism distinguishes between ‘coordinated market economies’ (CMEs) in the north-west of the continent (including Austria – geography is not the defining characteristic of this group or the others), and ‘Mixed Market Economies’ (MMEs) in the south, in the form of the now infamous GI(I)PS, Greece, Italy, (Ireland), Portugal, and Spain. The main difference between the two lies in the nature of the actors and the configuration of institutions and rules that they face. In CMEs, strong labour unions encounter strong employers associations, particularly in the export sector; as a result, they negotiate wage settlements which simultaneously safeguard real wages and profitability; and that is done through negotiating wage rates between a floor set by inflation and a wage ceiling set by labour productivity. Labour productivity, in turn, is high as a result of the micro-level arrangements for training and work organization in the northern CMEs. Strong systems of wage coordination then transmit these moderate wage rates to the rest of the economy. In MMEs, the situation is different. First of all, the state regularly has to step in to compensate for the lack of autonomous bargaining capacity among the key actors. Secondly, cross-industry wage coordination is considerably weaker than in the north of Europe, and as a result inter-sectoral wage drift is endemic. These differences in employment relations and wage-setting systems implied that, against the background of a relatively restrictive one-size-fits-all monetary policy in place since 1999, the north-west of the continent systematically improved its competitiveness, while the south lost competitiveness in parallel. Small differences between the two groups of countries at the start of EMU thus were accentuated and, against the background of low growth and an almost closed economy (the virtual economy known as EMU trades less than 10% outside the EU), the northern CMEs accumulated current account surpluses while the GIIPS ran into severe balance of payments problems in 2010 and 2011. The sovereign debt crises of 2010-13, which threatened the survival of the Euro-zone itself at several stages, simply reflected these structural imbalances: current account deficits are financed through debt, private and public. The problem with EMU, in other words, is one of current accounts, not fiscal deficits.

The analytics behind this argument are remarkably simple. Imagine, for ease of exposition, that EMU consists of two economies of equal size, called DE (i.e. Germany and its immediate neighbours in north-western Europe including Austria), and RE (for Rest of Europe). At some point after the start of EMU, DE’s inflation rate is, because of its more strongly coordinated wage-setting system, slightly below RE’s; they average two per cent, which is the ECB’s inflation target. Since the ECB sets its interest rate for all members to reflect the difference between the target and the actual (i.e. the aggregate/average) inflation rate of DE and RE, the real interest rate (the nominal interest rate that the ECB sets for all minus the country-specific inflation rate) is therefore lower in the country with high inflation (RE) and higher in the low-inflation country (DE).

This set-up has three effects, which are not very well understood. One, monetary policy is effectively pro-cyclical. The country with higher inflation faces a more accommodating monetary policy than it should, because the bank’s target is lower than its actual inflation rate and vice versa, the opposite of what would happen if monetary policy were decided for each country individually. The lower real interest rate that RE has faced during the first ten years of EMU has, with the exception of Italy, also fed into a path of higher growth in RE, fuelling (wage) inflation. At the same time, the tighter than necessary monetary policy has imposed further disinflation through wage moderation on DE. The very small differences in inflation that existed at the start of EMU thus have become more pronounced in the second round.

The differences in wage setting between DE and RE are crucial in these dynamics: not only did different wage-setting systems put DE and RE on different tracks from the start; the ability of DE to counter inflationary pressures through wage coordination around more slowly growing unit labour costs is almost perfectly mirrored by the inability of RE to do so. Since inflation is more of a problem in RE, the lack of capacity to disinflate implies that RE slowly but steadily lost competitiveness relative to DE – without being able to make up for that through trade outside EMU.

Historical roots of the problem

This drama of the euro has long historical roots. It started with the inception of the Deutschmark (DM) bloc in the first half of the 1980s, when all countries that decided to peg their currencies to the DM bloc found themselves forced to reorganize their wage-setting systems, often after significant social conflict. Wage rates in the sheltered sector in the economy – primarily in the highly organized public sector – were subjected to wage rates in the exposed (read: export) sector through voluntary wage coordination or through coercion imposed by a coalition of labour unions in the export sector, governments and central banks.

The outcome of this period of adjustment was a tightly organized system in which national central banks of the DM-bloc members were hierarchically linked to the Bundesbank, labour unions (and wages) in the exposed sector hierarchically linked to German wage setting, and public sector wages in each country hierarchically linked to exposed sector wages. The first of these linkages assured the credibility of the peg: national central banks made clear to domestic audiences that they would defend the currency, even if that entailed raising interest rates to a prohibitively high level. The composition of government in terms of Left and Right mattered little for this process: after Mitterrand’s U-turn in 1983, all EU (and prospective EMU) governments adopted low inflation as the key instrument to preserve the DM-centered exchange rate peg on which economic policy credibility now hinged. The second linkage, between the key German trade unions and their counterparts elsewhere, assured that the German set-up with a strong conservative central bank that disciplined excessive wages was transmitted to all other countries in the currency bloc. After some resistance in the early years of the decade, labour unions in the strong export sectors faced strong incentives to keep the public sector under control.

Wages outside Germany thus were kept under control through two mechanisms: one was direct wage shadowing, whereby wages outside Germany grew, adjusting for labour productivity, at a similar rate as German wages; the other was provided by credible conservative monetary policies as the back stop in case of excessive wage settlements, supported by an implicit coalition of governments and export unions.

The second stage, which covers the Maastricht convergence process in the 1990s, mirrored the first. The prospective EMU member states in the south of Europe went through massive, heroic adjustment programmes to meet the convergence criteria laid out in the Maastricht Treaty. But instead of arriving there after protracted social conflict, many of them adopted a path of social pacts and high-level concertation. Italy was the only one to adopt a ‘proper’ social pact involving all social partners, but Spain, Portugal and others at least attempted to find a broad agreement between social partners and government that encompassed all relevant areas and, when that failed, agreed to more topical reorganizations in some areas while leaving others to parliament. The end result was that by 1998 all EU member states that had expressed a preference to join EMU met the Maastricht criteria (bar Greece, who joined in 2001) and entered the single currency club.

The introduction of the single currency in 1999 dramatically changed the institutional framework of macro-economic policy, both within and between countries. First of all, it produced a pro-cyclical monetary regime. The single nominal interest rate, reflecting the ECB’s two per cent inflation rate target, translated into excessively accommodating real interest rates (the nominal interest rate minus the actual inflation rate) in countries with inflation above the two percent, and excessively tight monetary policy in countries with a low inflation rate. That fed into higher growth and higher inflation in the first group and lower growth in the second group, thus pushing both groups of countries in opposite directions: inflation rose in the high-inflation group in the first period and fell in the low-inflation group – thus fuelling asset price inflation in the first and stifling growth in the second group of countries.

While these perverse effects could easily be offset through fiscal policy, governments are on the whole quite reluctant to impose taxes, especially in times of fiscal surplus: fiscal tightening to counter monetary relaxation is thus very hard to implement. The Stability and Growth Pact (SGP), in addition, makes annual deficits above three per cent of GDP problematic: that raises the bar for counter-cyclical fiscal policy in a tight monetary regime. (The SGP, in fact, operates in a moderately pro-cyclical fashion as well, by rewarding countries with a surplus and punishing countries with a deficit, thus exacerbating the problems that pro-cyclical monetary policy produces.)

But the most important structural shift, though underappreciated in most analyses, is the implicit transfer of stewardship of the economy from domestic central banks that could respond to diverse conditions in each of the member states, to a single European central bank that steered the economy through euro zone-wide aggregates. Consequently, the domestic pressure exercised by the central bank on wage setters in each of the EMU member states effectively disappeared without being replaced by similar constraints imposed by Frankfurt. Many observers in the late 1990s thus feared a massive inflationary scramble as a result: since the ECB is unable to retaliate against one union in one country, excessive wage rates could no longer easily be punished, and a battle between the ECB and labour unions would erupt.

Different wage setting trajectories

This is not what happened. While wage inflation rates diverged between member states, EMU’s aggregate inflation rate remained low throughout the first decade, usually hovering between two and three per cent. Wage growth was, on the whole, moderate, and there were very few signs of the inflationary regime that many observers had feared. What the introduction of the single currency did reveal, however, was that wage setting in the member states were aggregations of two increasingly divergent trajectories: the exposed sector’s path, on the one hand, where markets had sufficient power to contain excessive wage demands, and the sheltered sector’s, on the other, where international competition (and in the case of the public sector any competition whatsoever) which restrains wage growth was absent. All other things equal, wage inflation was unlikely in the former, lest the export sector began to price itself out of the market and workers therefore out of a job, while it was, for the mirror reason of job stability, almost certain to emerge in the latter. The institution of EMU thus, somewhat perversely, reopened a cleavage within the labour unions that had been closed in the previous decades.

Here the systemic differences in the organization of labour markets between north western and southern Europe re-emerged: wage coordination across different sectors constrained the public sector in its wage setting in the former – mostly because shadowing wage rates in the leading manufacturing sector possibly secured the best medium-term wage deal for the public sector, but sometimes also because of coercion, as in Austria and Belgium, where institutional and legal constraints, such as labour law, budget rules or organizational power within the union confederation. In countries where the exporting manufacturing sector was not the leading trade union, however, or where public sector unions were capable of extricating themselves from the wage-setting system that revolved around the leading export-sector unions, wages in the public and in the manufacturing export sector diverged rapidly. This was the case in Ireland, Portugal, Spain, Italy, and Greece for much of the first decade of EMU up until the crisis of 2008. Since domestic wage inflation is, in effect, the weighted average of sheltered (including, and possibly dominated by, public) sector wage inflation and exposed (manufacturing and other export) sector wage inflation, inflationary pressures thus started to rise in these countries.

Rising wage inflation in the public sector is, in principle, relatively easy to compensate in the exposed (export) sector, as long as the productivity rate of the latter is high enough – which it is in much of the key manufacturing sectors – and wages grow at a moderate enough rate. But in some cases the export sector may have only a low potential to compensate, because it consists primarily of relatively low value-added sub-sectors, because the export sector is too small compared to the sheltered sector, or the export sector might simply set its own wages above productivity regardless of the consequences, thus exacerbating the inflationary pressures emanating from the sheltered private and public sectors. Under those circumstances, the ability to compensate for high wage inflation in the sheltered (public) sector is drastically limited, aggregate domestic wage inflation rises faster and higher, and the competitiveness of the export sector falls rapidly as a result of what is, in effect, an appreciation of the real exchange rate. That was also exactly what we witnessed in the EMU economies that faced important public debt problems in the 2010-13 period. Before the introduction of the Euro in 1999, manufacturing wages and public sector wages roughly followed the same pattern in all prospective member states. From 1999 onwards, however, the evolution of the two diverged sharply: manufacturing wage rates across the euro-zone remained tightly controlled (expressed in unit labour cost terms, they were negative, in fact – Johnston 2012), while public sector wages were on an upward trajectory until 2007.

The upshot of this analysis: the institution of the ECB – the key actor at the heart of EMU – has inadvertently contributed to outcomes that run counter to its own mandate: instead of taking away the punch bowl when the party got going, as Paul Volcker once famously described monetary policy-making, it added rum to the mix in the south to make it more potent, while never even bringing out the bowl in the north. The effect was predictably that southern Europe found itself in a haze fed by easy money, while the north was forced to tighten belts. Labour markets, and especially differences among countries in this area, were important arenas in this process, both historically and during the first decade of the euro. They determined the path along which the political-economic regime would evolve, and they became key adjustment tools in the single currency regime. By and large, the northern economies had developed institutions to handle what is now known as an ‘internal devaluation’, by keeping a tight lid on the growth of wages (adjusted for productivity). The southern economies, however, never really developed such labour market institutions, and thus found themselves unable to compensate wage inflation, particularly in the public sector.

But there is more: constructing such benign labour market institutions that control wage costs may, as France in the 1980s and Central Europe in the 1990s learned to their dismay, be very difficult, and probably impossible. Assume for a moment, as the Varieties of Capitalism school does, that labour market institutions of the northern type indeed depend on a tightly interlocked network of other, logically prior institutions such as labour unions and employers that understand the logic of collective bargaining, cooperation among companies to avoid unnecessary competition in the labour market, stable and robust skill production systems, and mechanisms that defuse social conflict in the workplace. Each of these elements has been several decades or even centuries in the making; building one of them requires building all of them. Posing the problem this way gives us a sense of the deep challenge that EMU faces: if the key mechanisms of adjustment in today’s EMU have such deep roots in history, we probably should be rethinking the arrangement from the ground up and search for a different monetary union than the one we have today.