VoxEU Column EU institutions EU policies

Why Europe needs two euros, not one

As the Eurozone cautiously implements stabilising reforms, Germany is forced to go further with concessions than it would prefer. This column suggests that it would be beneficial for discontented members to consider the formation of a second monetary union. The second euro can be constructed better than the first, bringing the discontented members exchange-rate adjustments relative to Germany, and avoiding competitive devaluations.

One basic feature of the sickly situation in the Eurozone today is that the system does not clearly bear any essential flaw from the standpoint of Germany. All things considered, the country has not done badly since the Great Recession of 2008-2010. And as the Eurozone moves forward gingerly with necessary reforms in order to avoid a break-up of the system, it is evident that Germany is constantly under pressure to go further with concessions than it would prefer.

In these circumstances, it may be time for radically new thinking. I would like to propose that the discontented members would do well to consider seriously the formation of a second euro. Such proposals have been made (Dobbs and Spence 2012), but the lingering unsatisfactory macroeconomic conditions seems to call for a rethink.

Three advantages of a second euro

A second euro would have three fundamental advantages.

  • First, the second euro could be constructed in a better manner than the previous euro from the discontented members’ point of view. There is no need to repeat the errors of the past.
  • The second argument comes in two parts.

First, the creation of a separate monetary union by the discontented members would bring them exchange rate adjustment relative to Germany, which is the single most important exchange rate adjustment that they need.

Second, by forming a second monetary union rather than reverting to separate currencies, they would still avoid the problem of competitive devaluations that hounded the earlier EU after the breakdown of Bretton-Woods and up to the appearance of the Maastricht Treaty as a possibility on the horizon in 1986 (when the Single European Act came).

  • Thirdly, there are strong indications that the discontented members of the Eurozone will get a worse deal from the movement towards reforms of the system that is now proceeding with grudging German approval than they would by forming a second euro.

Every German concession faces major political opposition at home, to say nothing of an occasional legal challenge. Furthermore, these roadblocks are probably in Germany’s true interests.

Of course, these three basic arguments for a second common European currency pave the road to one more.

  • A second euro could serve as a bargaining tool in negotiations with Germany.

But I would like to rest the case for a second euro strictly on its own merits and, therefore, strictly focus on the first three arguments.

One issue I shall disregard is the costs of the transition, which prominently include the interpretation of outstanding debts in euros. It is important, however, that the status quo could fall apart in the Eurozone in any event, and if only for this reason, it is a good idea to entertain the best of the alternatives. Furthermore, if the only serious argument against two euros is really that ‘it is better to stick to a bad marriage than to go through the costs of a divorce,’ it would seem good to know it and to behave accordingly.

First advantage of a second common currency

Any optimal design of a new social regime with broad distributive consequences should be done in ‘the veil of ignorance’ about where the chips will fall in the future, and from this point of view, the Eurozone contained two spectacular flaws.

  • First, a monetary union should provide monetary control by the central bank. But the rules of the Maastricht Treaty violated this condition.

They made monetary control by the ECB contingent on the willingness of commercial banks to lend to individual firms and households. The ECB was not free to buy and sell in open market operations. The term ‘open market operations’ surrounding the ECB describes certain kinds of refinancing transactions with banks and, since 2010 – first, under the ‘Securities Market Program’ and, more recently, under the program of ‘Outright Monetary Transactions’ – purchases of bonds of member government that are under financial stress with effects on the monetary base that must be sterilised. These are not open market operations in the true sense. It is a fact that a number of the central banks in Europe that preceded the ECB operated for decades in the manner that the ECB was allowed to do without any problem, and under ordinary circumstances the ECB would have been able to do the same. However, constitutions must allow for conditions that are only likely to happen once every half a century or more.

In the face of the situation that arose in 2010-2012, where member banks of the Eurozone preferred to rebuild their balance sheets than to lend to their customers, and where the official intervention rate on the interbank market approached the zero interest rate floor, once the ECB had exploited a few loopholes, the new constitution left it powerless to increase the money supply at all without engaging in legally questionable actions or emergency measures that required bending the rules.

  • Second, in a monetary union any threat to the payment system resulting from insolvency of the banking sector in any political region of the system is a collective problem to be shared by all. This condition was also not met.

Unlike the situation in the US following the savings and loan crisis in the 1980s and 1990s, the Irish paid for the insolvency of their banking sector, the Spanish paid for theirs, and in Cyprus, it even looked for a while that small bank depositors would have to contribute heavily to pay for theirs. Help from the rest of the Eurozone in these cases came late, haltingly, and moderately, and came to the national governments who avoided the meltdown of their banks (on terms that Spain has never accepted).

Repairing both these problems, of course, is simple. As regards to open market operations, there must be some designated securities that the central bank can buy and sell. To avoid partiality to any private interests, the purchasable securities should be public or semi-public, so that the beneficiaries are essentially taxpayers. To avoid partiality to any country, an international mix of debts is necessary. Avoiding moral hazard is also easy. The central bank should only be able to buy securities that have aged sufficiently. For example, suppose that the required ageing is three years. Then, the outstanding stock of government securities that would be eligible for purchase by the central bank would be enormous and, yet, no new issues could be monetised until the lapse of a significant time.

The euro-nomics group (Brunnermeier et al. 2012) has recently made an interesting suggestion that would permit the central bank to buy a ready-made assortment of securities with different maturity dates satisfying my condition. The group proposes creating a new intermediary (which they call the ‘European Debt Agency’) that would hold an imposed combination of government securities (in fixed, irreversible weights) against which the intermediary would issue new securities (which they term ‘European Safe Bonds’) to the public that the central bank could subsequently buy and sell. This would help but it is not essential.1

Joint responsibility for the solvency of the banking sector in any political region of the monetary union also requires some preconditions that are fully manageable. If risks of insolvency are to be borne collectively, evidently all the member banks must be subject to uniform prudential rules and a collective supervisory authority.

Further, in case of a meltdown of the financial sector in a member country, the costs of compensating the depositors or creating new bank capital, or both, must be borne collectively. There has been a move toward some form of banking union (29 June, 2012) and therefore the satisfaction of the first precondition, but little advance toward the second one, which is equally basic (for a good, detailed discussion of the required degree of fiscal union, see Pisani-Ferry et al. 2012). Of major note, the needed degree of fiscal union is quite limited. Even joint insurance of all demand deposits may be going too far. Something like the Single Bank Resolution Fund (as distinct from the Single Resolution Mechanism) that has been mentioned but is nowhere in sight, could well suffice.

Second advantage

A repeated question about the Eurozone is whether the members form an optimal currency area. But another question – which is actually closer to Mundell’s original contribution (Mundell 1961) – is whether the right number of currencies in the Eurozone is as high as 18, the number of member countries in the system. If the right number is neither 1 nor 18, then 2 may be far, far better than either extreme.

Let me begin by recalling the reasons why 18 would be too many. First, some of the members are probably small enough to have no scope for using monetary or exchange rate policy as a tool of economic stabilisation over the business cycle (McKinnon 1963). Next, the 18 members do form an economically integrated group of geographical neighbours, and therefore their mutual efforts to use monetary and exchange rate policy to their advantage could easily lead them to enter into non-cooperative games with costly Nash consequences. Finally, national monetary policy can mean Treasury-dominated monetary policy, which can lead to very poor outcomes apart from strategic games with any foreigners, near and far.

If 18 is wrong, there are a couple of reasons to veer toward 2, if not all the way. First, studies of the question whether the Eurozone is an optimal currency area generally tended to show no more than one major possible fault line in a single monetary policy for the 12 initial members (including Greece, which entered in 2001, but excluding the UK and Denmark) – roughly between north and south (see, e.g., Bayoumi and Eichengreen 1992). Subsequent events since the crisis of 2008-2010 have underlined this fault line. In particular, the under-pricing of German goods relative to southern European ones in the current debacle in the Eurozone is too well-known to require new development. The recessionary and deflationary consequences also need not be rehearsed. My main point, though, is that judging from past studies of synchronisation of business cycles and asymmetries in economic performance in the Eurozone (and therefore excluding the UK and Denmark), two currencies may be enough – one including Germany, the Netherlands, and Austria, and another comprising the countries bordering the Mediterranean and stretching to Portugal.

Next, history provides an important reason for 2 instead of 18 in the case that 1 is wrong. The only experience with freely floating exchange rates in the EC – the EU’s predecessor – was short-lived and unfortunate. It followed the breakdown of Bretton-Woods in 1971. Soon after this experience, the members of the EC tried to return to some semblance of exchange rate stability, and in any scenario of a break-up of the Eurozone, it is difficult to imagine that the members would tolerate general floating. The strategic game aspects, if nothing else, weigh too heavily against this alternative. Yet, the past also teaches us that no system of fixed exchange rates ever worked in the EC. The main effort to bolster exchange rate stability after the ineffectual ‘snake,’ the European Monetary System (EMS), led to major realignments every nine months in 1983-1985. It is highly questionable that the EMS was dynamically stable and would have survived much longer despite major capital controls, had not the prospect of Eurozone come to save it. Thus, a further reason to envisage a second euro is that neither floating nor fixed exchange rates would work in the event of a break-up of the Eurozone. On the other hand, having three major currencies in the EU seems tolerable. The floating of the British pound relative to the euro has raised no difficulty, even though it means having two major currencies in the system. The Common Agricultural Policy also now depends heavily on direct subsidies rather than price controls, so that exchange rate movements cannot wreak the havoc in agriculture that was possible in the 1970s and 1980s. Two euros would be a far cry from return to separate liras, pesetas, francs, escudos, drachmas, etc.

Third advantage

We are now living through a veritable tragicomedy in the Eurozone. Faced with a situation where Greek, Spanish, Portuguese, and Italian governments were forced to borrow at unsustainable interest rate premiums, the ECB was compelled to step in to lower the interest rates in order to prevent defaults that would have spread to the banking system and forced some countries out of the Eurozone. But at every step along the way, the question raised in Germany, and that has now come up twice before the German Constitutional Court, is whether the ECB is acting within its mandate. And the tragicomic aspect is that the ECB probably is not and the Court of Karlsruhe thinks so but simply forestalls crises for political reasons. However, in its most recent decision (7 February 2014), the Court has left little room for future manoeuvre by explicitly dismissing the ECB’s argument that the Outright Monetary Transactions program is legitimate on grounds that it ensures the proper functioning of monetary policy and avoids the breakup of the Eurozone. Therefore, the scope for expansionary monetary policy in the future in the Eurozone is now more circumscribed than ever.2

Is this then a game that the large section of the Eurozone membership with grave need for expansionary monetary policy has an interest in continuing to play? It would be easy to write a new set of rules that would impose tough inflation targets on another central bank, yet still allow expansionary monetary policy in deep recessions, and provide the central bank the ability to engage in selective aid measures in circumstances that threaten the very survival of the monetary union without inviting such circumstances to arise. Very significantly, no joint bonds backed by ‘joint and several’ guarantees of the member governments would be called for. Let us remember too that the Eurozone courted some of its current problems by announcing early on that the whole system hinged on the Stability and Growth Pact despite the no-bail-out rule, instead of warning markets that government defaults remained possible (despite the Pact), and the organisation would only intervene to insure the safety of the banking sector and the payment system.

The prospects for banking union are admittedly better than those for adequate monetary control. Yet, as mentioned before, the fiscal implications of banking union have not received the attention they need. The Single Bank Resolution Fund that has been mentioned will require a separate treaty and remains only a vague possibility.3 However, without a mechanism assuring the collective bearing of the costs of a major insolvency, there is no reason to expect that we can avoid a repetition of the experience since 2010 with ‘the diabolic loop’, as the euro-nomics group (Brunnermeier et al. 2012) terms it between bank risk and government bond risk (Ireland, Spain, and Greece, see also Shambraugh 2012.)


Germany was unquestionably the leader in the pre-euro European Monetary System. It was never clear when Eurozone got started why the country agreed to renounce its monetary independence. The question ‘What’s in it for Germany?’ never got a satisfactory answer. But, in compensation, the country did obtain virtually free rein to set the rules in the Eurozone. At first supporters of a single currency (including the present author) vaguely expected that the constitutional problems that might arise would be ironed out as they appeared. But this overly sanguine expectation has proven false. In these circumstances, isn’t it time to give some serious thought to a second euro?


Bayoumi, T and B Eichengreen (1992), “Shocking aspects of European monetary unification”, CEPR Working Paper 643, May.

Brunnermeier, M, L Garicano, P Lane, M Pagano, R Reis, T Santos, D Thesmar, S van Nieuwerburgh and D Vayanos (2012), “European Safe Bonds (ESBies),” Princeton University, 2 April.

McKinnon, R (1963), “Optimum currency areas”, The American Economic Review, 53, September, 717-725.

Mundell, R (1961), “A theory of optimum currency areas”, The American Economic Review, 51, pp. 657-664.

Pisani-Ferry, J, A Sapir, N Véron and G Wolff (2012), “What kind of European Banking Union”, Bruegel, 25 June.

Posen, A and N Véron (2014), “Europe’s half a banking union”, Europe’s World, Summer, pp. 10-18.

Shambaugh, J (2012), “The euro’s three crises”, Brookings Papers on Economic Activity, Spring, pp. 157-226.

Sinn, H W (2014), “Responsibility of states and central banks in the Euro crisis”, CESifo Forum, 15 (1), Spring.


1 The euro-nomics group’s concern is also broader than mine. The members want to create safe assets rather than merely define a set of member government securities that could be purchased by the central bank without causing any moral hazard. Therefore, they define an appropriate level and combination of purchasable government securities differently than I do.

2 See Sinn’s revealing report to the German Constitutional Court, (Sinn 2014), which raises legal or political questions about every least fiscal implication of the OMT program for Germany, including, for example, the potential future effects on its part of seignorage revenues.

3 For a more optimistic appraisal, see Posen and Véron (2014).

3,465 Reads