VoxEU Column Europe's nations and regions

For Greece, a “fiscal devaluation” is a better solution than a “temporary holiday” from the Eurozone

Martin Feldstein argued last week that Greece should take “a temporary leave of absence with the right and the obligation to return at a more competitive exchange rate.” In this column, Argentina’s highly regarded former Minister of the Economy and a co-author argue that the idea won’t work. A better solution would be to adjust the Greek tax system.

Martin Feldstein argues that, to mitigate the costs associated with fiscal adjustment, Greece should ask its Eurozone partners for “a temporary leave of absence with the right and the obligation to return at a more competitive exchange rate” (Feldstein 2010). More specifically, he suggests that Greece reintroduce the drachma at par with the euro, devalues it by 23%, and returns to the Eurozone a few years later with a more competitive exchange rate of 1.3 drachmas per euro. His proposal does not include a redenomination of financial contracts from euros to drachmas (as this would be the same as defaulting on those contracts) but, simply, a temporary change in the unit of value used to set prices and wages in the economy.

The objective is to reduce unit labour costs to expand exports and reduce imports (assuming that the pass-through effect of the devaluation to local prices is nil to low) and, in this way, stimulate the economy in order to offset the contraction due to the inevitable fiscal adjustment needed to improve Greece’s creditworthiness (this, in turn, assumes that the expenditure-switching effect of the devaluation outweighs the contractionary effects due to other reasons, such as balance sheet effects and the increase in import prices).

Eurozone as Bretton Woods?

In essence, what Marty is proposing is to re-create within the Eurozone the same mechanism that existed under the Bretton Woods system, which allowed countries facing severe external and fiscal imbalances to adjust their fixed exchange rates after consulting with the IMF and the other countries that participated in the agreement.

In this sense, the proposition has some logic but, unfortunately, it opens the door for credibility issues in other countries of the Eurozone, which makes it politically difficult to accept by all.

We do, however, totally agree with Marty that, unless something is done to improve external competitiveness, fiscal adjustment alone will not do the job of bringing the Greek economy back on its feet. In Argentina, a similar monetary arrangement as the one Greece has today prevented the economy from adjusting more smoothly in 2001 when, faced with a confidence crisis and little international support, the government was forced to implement a drastic fiscal adjustment program. Social tensions caused by this and other measures, including the temporary implementation of capital controls, led to the fall of the government and the adoption, by a new administration, of terribly misguided policies, the price of which Argentina is still paying today. One of these policies was the forcible conversion to pesos of financial contracts denominated in US dollars, something that (as Marty correctly perceives) Greece should be loath to do.

Would devaluation improve competitiveness?

The point where we beg to differ with Marty is his idea that exchange rate devaluation is necessary to restore competitiveness. Actually, the same effect can be achieved through other means, which, at least in the case of Greece today, are less disruptive, for example, by changing the tax structure without affecting tax collection.

In our opinion, one of the reasons why labour costs are high in Greece is social security taxes. These amounts to 44% of gross wages, of which 28% is paid by the employer and 16% by the employee. Together, these taxes raise €33 billion for the social security system or slightly less than 14% of GDP. In contrast, the value added tax, whose base is broader, collects less than €32 billion. And this despite the fact that payroll tax evasion (particularly among the self employed, but also among firms) is notorious in Greece, which is not surprising given the high level of these taxes.

Of course, the reason why there is less collection of VAT than of social security taxes despite the broader base and the relatively lower tax evasion (it is, generally, harder to evade VAT than to evade payroll taxes) is that the effective VAT rate is relatively low. At issue is the fact that, in Greece, there is not one but several VAT rates, which vary across products and regions. Thus, while the general rate is 19%, the one that applies to basic necessities, such as food and medicines, is 9% and that for books and newspapers is 4.5%. In addition, some Greek islands enjoy lower VAT rates than the rest of the country. Besides introducing distortions at the microeconomic level, a non-uniform tax rate reduces tax collection for a given level of the average tax rate by inducing tax arbitrage. Conservatively, one could say that if the Greek government applied a uniform rate of 15% across all sectors, it would collect about the same amount of money it collects today under a differentiated rate system with a top rate of 19%.

Using the tax system instead

Given this information, it appears to be possible, in principle, to redesign the Greek tax system in such as way as to achieve the same reduction in euro labour costs that Marty has in mind without requiring Greece to either exit the Eurozone or sacrifice tax collection. This can be done by:

  • raising the VAT rate to 25%;
  • applying it uniformly across all goods and services produced in the economy; and
  • either eliminating payroll taxes altogether or allowing firms to expense them fully against the VAT (of these two options, we prefer the last one since it guarantees the continued funding of the social security system while rewarding the firms that comply with their payroll tax obligations).

The table below shows how we get to this conclusion. Increasing the VAT rate to 25% and making it uniform would raise VAT collection by €21 billion (from €31.5 to €52.5 billion). This is the same amount that would be lost by allowing firms to credit their payroll tax payments (namely, the 28% of wages that employers contribute to social security) against the VAT. While the effect on tax collection is neutral, the effect on the economy is not. Under our proposal, total labour costs (the sum of gross wages and payroll taxes) would decline from €96 billion to €75 billion. The difference of €21 billion is, in percentage terms, similar to the one Marty has in mind. However, absent the pass-through and contractionary effects of devaluation, the impact on competitiveness should be higher, perhaps significantly so. Naturally, some firms would be benefitted more than others. But, unless one wanted to reward tax evaders, there would be nothing to lament.

As a final note, let us point out that this measure, along with several others, including a 13% nominal reduction in public sector wages and an orderly restructuring of the public sector debt held by domestic creditors were successfully implemented in Argentina throughout 2001.

If they were not enough to stave off the monetary and debt crisis that occurred at the end of that year was because – after having promised financial support in exchange for the implementation of those measures – the IMF suddenly decided to pull the plug on Argentina by announcing, in November 2001, that it would not disburse the funds that it had promised, triggering a massive run on bank deposits and sovereign bonds that precipitated the crisis. We hope that Germany, France, and the other Eurozone countries will not make the same mistake with Greece that the IMF made with Argentina.

Table 1. How a "fiscal devaluation" works, (in billions of euros)

  Current Proposed
   Revenue  Rate  Base  Labour cost  Rate Revenue  Labour cost 
VAT gross   31.5  15%  210.0    25%  52.5  
Payroll credit  –          21.0  
VAT net  31.5          31.5  
Payroll tax  21.0  28%  75.0  96.0  28%  21.0  75.0
Total  52.5          52.5  



Feldstein, Martin (2010), “Let Greece take a holiday from the Eurozone,” Financial Times, 17 February.