Recently, our analysis has been questioned by Servaas Storm who has claimed that it is untenable to blame neo-mercantilist Germany for driving a wedge into the Eurozone. [i] It is shown below that Storm’s critique has a certain aplomb, but lacks substance.
Most macro-economists in Europe have probably accepted by now that persistent German wage moderation is a vital cause of the fundamental imbalances in the European Monetary Union (EMU). The writers of this note have shown that these imbalances are responsible for the Eurozone crisis in the broadest sense, since they have amounted to German dominance in trade flows (exports), export of German unemployment, weak investment and low productivity gains within the Union and, finally, deflation. [ii]
Storm’s position is clear and worth quoting at length:
“Secondly, as shown in Figure 1 [in Storm’s article], there is no clear sign of a nominal wage squeeze on German workers if we compare Germany to the Eurozone as a whole (but excluding Germany). German nominal wages increased relative to the Eurozone in the 1990s and the German relative nominal wage stayed more or less flat during the period 1999-2007 (there was a negligible decline of 0.7 percentage points over these eight years). It is nevertheless true that Germany’s unit labor cost declined relative to those of the rest of the Eurozone (as Figure 1 illustrates), but this was not a result of wage restraint: It was completely due to Germany’s outstanding productivity performance: during 1999-2007 average German labor productivity (per hour worked) increased by almost 8 percentage points compared to the rest of the Eurozone, which accounts fully for the decline in Germany’s relative unit labor costs by 7.8 percentage points over the same period. It was German engineering ingenuity, not nominal wage restraint or the Hartz “reforms,” which reduced its unit labor costs. Any talk of Germany deliberately undercutting its Eurozone neighbors is therefore beside the point.”
In sum, Storm admits that there has been nominal wage moderation in Germany, but considers this fact to be relatively unimportant. In his view, what has mattered in the Eurozone has been superior German productivity and correspondingly declining relative unit labor costs. Accusing Germany of deliberately undercutting its partners via domestic wage moderation is to miss the point. This is an astonishing position. Storm, who has authored a number of articles on the Eurozone, simply misunderstands the functioning of monetary unions and the consequences for their members.
A monetary union is in essence an agreement to have a common inflation target coupled with the willingness to transfer the power of monetary policy to a single central bank. As unit labor costs are closely correlated with inflation, a monetary union implies agreement that nominal wages in each member state would rise in accordance with national productivity gains plus the inflation target (in the case of the EMU just below 2%). If this rule was followed, differences in productivity among member states would of course still arise, but they would not imply divergences in national competitiveness within the monetary union. Competitiveness gaps would typically emerge from divergences in unit labor costs correlated to divergences in inflation.
Storm does not deny that a real depreciation has taken place in Germany, since he admits that German wages rose less than German productivity. But he states that German wages increased less relative to Germany’s partners and that German productivity increased strongly relative to Germany’s partners, thinking that this somehow proves his case. In fact this argument is nothing more than a rephrasing of our core statement that German nominal wages rose too little given the inflation target of 2%. It is remarkable that Storm does not see what the implications are. And that is without mentioning that it is simply not true that German productivity performance has been clearly superior to competing countries; indeed, in the case of France it is plain wrong.
There is, however, a broader import to the critique by Storm, which reveals some of the misunderstandings that have gradually accumulated in economic policy making and theory. Specifically, there is a profound misunderstandings regarding competition among nations in the field of trade and the implications for exchange rates. It is briefly shown in the following section that over- and undervaluation of implicit exchange rates in a monetary union cannot be the result of divergent growth in productivity, but only of nominal wages exceeding or falling behind national productivity increases. The only valid measure for nominal wage increases could be the commonly agreed inflation target.
The most important theoretical point in this respect is that competition among nations in the international markets is qualitatively different from competition among enterprises. With this in mind, the fate of fixed exchange rate systems as well as of monetary unions is determined by consistent adjustment of nominal wages to national productivity for each member state. In Europe, as long as Germany maintains current domestic policies and imposes austerity on the rest of its partners, the EMU boat will continue to head for the rocks at high speed.
Competitiveness of Countries and Competitiveness of Enterprises [iii]
During the years of “globalization” it has become commonplace to consider that nations compete in a manner similar to enterprises. More than that, it is often thought that nations should be made to compete in that way. The wealth of nations is supposed to depend on their ability to adjust effectively to the challenges created by open markets for goods and capital: nations with high capital endowment would come under pressure by trading partners with low labor standards. In particular the existence of a huge pool of available labor in developing countries – such as China and India – would presumably alter the capital/labor ratio of the entire globe in favour of capital, forcing the equilibration of low and high wages somewhere in the middle.
Reality appears to have confirmed this view, as wages in several high-wage countries of the North have indeed come under pressure, and labor has dramatically lost out to capital in appropriating the benefits of productivity growth. [iv] Wage shares have been falling and the promise made by the advocates of free markets that there would be full participation of all people in the progress of society is fast fading. However, the actual decline of wage shares does not imply that the forces driving this development are those to be found in the neoclassical theory of the labor market. Furthermore, and contrary to neoclassical thinking, the analysis of competition among enterprises does not apply to competition among countries – neither to those with independent currencies, nor to those in a currency union. Consider, first, the basic parameters of competition among enterprises, as established by both Keynesian and Marxist economics.
In a dynamic setting, the enterprises of a market economy compete through productivity gains. Supply side conditions are pretty much given for all since market forces tend to equalise the prices of intermediate goods, such as labor, and the cost of capital. Consequently, success or failure is determined by the specific value that is added at the enterprise level to the goods and services obtained as inputs in the open markets. Enterprises have to pay the price of labor determined in the market for different qualities and levels of skill as well as covering the cost of capital. Enterprises that are able to generate higher productivity through innovation and new products produce at lower unit labor cost than their competitors, thus offering their goods at lower prices, or making higher profits at given prices. If the former holds, enterprises gain market share; if the latter, enterprises may gain strategic long-term advantages through higher investment ratios. As long as prices for labor and other intermediary products are given, competing enterprises adjust by implementing the same or similar technology, or quit the race through bankruptcy.
In competition among countries, however, this mechanism does not apply because wages are normally set at the national level. Whether through limited mobility of labor at the international level, or through wages being set through nation-specific institutional procedures, countries are wage setters, not wage takers. If wages are centrally negotiated at the level of the nation state, or if labor is mobile only within the boundaries of a nation state, the so-called law of one price, i.e. a given rate for standard work, is broadly applicable. It follows immediately that stronger productivity growth at the level of a nation state would not increase the competitiveness of the nation’s enterprises against the rest of the world, since advantages in productivity would normally be reflected in higher nominal (and real) wages and unchanged growth of unit labor cost.
Even if, for whatever reason, this mechanism did not work a country with high productivity but very low wages and thus very low unit labor costs would still not automatically increase its national competitiveness, and nor the competitiveness of its enterprises. For, prices in that country would not necessarily be lower than in the rest of the world, if expressed in international currency. In a world of national currencies and national monetary policy, a country supplying its goods at much lower prices would gain market share in a number of fields, thus accumulating large trade and current account surpluses. Economic and political pressures to adjust wages and prices expressed in international currency would mount, and sooner or later the country would be forced indirectly to adjust its wages, as measured in international currency, through a revaluation of its national currency.
Nations would be able to open their borders to trade and capital flows, if they could be assured that their enterprises would have a reasonable chance in the global division of labor, and that they would not be in danger of permanently losing out against the rest of the world. This simple proposition underlies all international trade arrangements in the World Trade Organization and elsewhere. If, at the national level, the nominal remuneration of labor (i.e., of the internationally immobile factor) consistently exceeded its effectiveness (labor productivity) by a wider margin than in competing countries, the country would be getting in trouble as most of its enterprises would be facing difficulties. For, its enterprises would either ask for higher prices and accept permanent loss of market share, or take lower profits to avoid loss of market share.
A condition of this type – an overvaluation due to an appreciation of the “real exchange rate” – would not be sustainable for long. As a rule of thumb, if the accumulated overvaluation reached about 20% or so, a crisis would be unavoidable. A growing deficit on current account would be the most visible indicator of this pathological condition, not its cause. If exchange rates were adjustable, the solution would be simple: the currency of the country with the deficit would devalue, bringing nominal wages and nominal unit labor costs measured in international currency back to a competitive level. Devaluation would also lead to a relative fall in real wages, but that would be a by-product and an incidental part of the underlying process. Thus, in Europe, Italy and Britain faced precisely this problem as members of the European Monetary System in 1992; one opted in and the other opted out, but both devalued.
In a currency union member countries should agree – explicitly or implicitly – not to go down the inflationary route, i.e., to have nominal wages exceeding national productivity by more than an explicit inflation target. If they do decide to go down that route, they should all do it together. With an inflation target of close to 2% in the EMU, the implicit understanding is that nominal wages would not rise by more than national productivity growth plus two per cent. This means that each country could and should enjoy the benefits of its own productivity increases, be that 1%, as in Germany, or 2%, as in Greece. These benefits would be growth in real wages, or shorter working hours; perhaps it could be a combination of the two. Such a policy would not at all prevent – indeed it would encourage – each country in the union from taking measures and introducing reforms to improve its productivity performance.
The most important point is that in a monetary union deviations from the common inflation target are equally dangerous whether nominal unit labor costs are rising faster or slower than the target. The former creates inflationary dangers for the union as a whole, while the latter creates deflationary dangers. If one member deviated either upward or downward from the target, an externally unsustainable situation would result, which would also be reflected in deviations in long-term interest rates among the members of the union.
Since the time of David Ricardo political economy has known that competition among countries cannot and should not be analysed similarly to competition among enterprises. In Ricardo’s terms, the law of labor value does not hold among countries and the law of comparative advantage prevails. For a monetary union such as the EMU, the principle of differentiating between countries and enterprises means that national competitiveness depends on divergences from the inflation target, and hence on unit labor costs.
Germans ought to know better than all others about the difficulties caused by wage divergences in a currency union. The deviation of East German wages as measured in international currency, following the German Monetary Union of 1990, destroyed East German industry and forced a transfer union. Unfortunately, for the EU and the EMU the option of a transfer union is simply not available. As long as Germany persists with its policy of wage moderation, the only future for the EMU is collapse.
[ii] See (Flassbeck and Lapavitsas, 2013 and 2015). See also Flassbeck (2007).
[iii] For a fuller analysis of this issue see Flassbeck and Lapavitsas (2015).
[iv] See Lapavitsas (20013, ch. 7).
First published in ineteconomics.org
View this article in .pdf form