In conformity with their rights and obligations related to EU membership, all acceding countries have now submitted to the officials of EU both a prospective date to join EMU and a strategy in order to reach it. “Estonia and Cyprus aim at introducing the euro early, in 2006 or 2007. The other acceding countries favour a later adoption of the euro. Hungary, Latvia, Poland and Slovakia mention 2008 or 2009 as target years for joining the euro area, while the Czech Republic targets euro adoption by 2009-2010.” (Italianer, 2003). Indeed, by acceding to EU, they enter as Member States with a derogation which provides that, sooner or later, they will have to join EMU. It should be recalled that in the context of EU accession negotiations, they have not tried to negotiate an “opting-out” clause, like the United Kingdom or Denmark. Adopting the euro as legal tender is truly the ultimate goal of acceding countries and, in that sense, accession to EU is only an intermediate step on the way.

In broad terms, euroisation denotes the use of a foreign currency (i.e. the euro) in the domestic monetary system of a non-issuer country. “Euroisation” is then just an adaptation of the term “dollarisation”, used to define the use of the dollar in non-issuer countries, essentially Latin America economies.

The idea of an early introduction of the euro was first put forward in December 1996 when the Slovenian Prime Minister said that “Slovenia will do its best to join monetary union as soon as possible, perhaps even before full EU membership” [9]. This idea was dropped fairly soon, however, after the EU had signalled that it would not appreciate such an approach.

This section surveys the main costs and benefits related to euroisation or, more generally, to the adoption of the euro. Indeed, some of them result from using the euro, regardless of the way in which it is introduced in the economies. Nevertheless, an accurate analysis reveals some differences between unilateral and regulated introduction of the euro.

The costs arising specifically from unilateral euroisation are generally categorised as resulting from (i) a “money supply shock” due to insufficient reserves in euro; (ii) a loss of seigniorage revenues; (iii) the absence of a lender of last resort in the event of a banking crisis. Finally, the use of the euro, however it is introduced, induces (iv) inadequate responses of monetary and exchange rate policies to smooth asymmetric shocks.

From a pure macroeconomic viewpoint, one necessary and sufficient condition for a successful unilateral euroisation is a sufficient level of official foreign exchange reserves to replace the national currency. However, this technical condition is not exempt from problems: which monetary aggregate concept to use? In the literature, no consensus emerges. According to Wójcik (2001), the main technical precondition is that official international reserves cover the monetary base (currency in circulation including vault cash and commercial bank reserves at the central bank, i.e. M0). By contrast, Gabrish (2002) argues that broad concepts (i.e. M3) should be used because the demand for money transactions is contemporaneously not precisely identifiable. The problem is that the central banks of candidate countries report only M1 and M2 data: a potentially large part of private savings in banks (for instance in M3) is then unknown. Alongside the official foreign exchange reserves, the deposits denominated in foreign currency are often considered as another precondition for a successful euroisation (Habib, 2002).

One cost of unilateral euroisation (and also the more easily quantifiable) is the loss of seigniorage revenues that accrue from the issue of a legal tender. Indeed, circulating cash represents a non-interest obligation of the central bank, matched on its balance sheet by interest-bearing reserve assets that are an ongoing source of revenue. Unilateral euroisation automatically terminates that revenue. This contrasts with a regulated introduction of the euro since EMU members participate in the sharing of seigniorage revenues that derive from issuing the euro.

According to the flow approach, losses of seigniorage revenues due to unilateral euroisation are evaluated for the current period to be in the range of 1-2% of GDP per year, depending on methodologies and/or countries (Habib, 2001; Hochreiter and Rovelli (2002); Backé and Wójcik, 2002 for a empirical survey) [12]. Such losses are generally judged quite low. For comparison, the current income seigniorage incomes in Western industrialised countries are put at just below 1% of GDP.

Empirical studies based on stock data show that full membership in EMU would result in positive net gains for all candidate countries except Malta and Cyprus (Feist, 2001; Fisher et al., 2002). Indeed, most accession countries are gainers in the redistribution of seigniorage wealth while Germany, Austria and Spain are unambiguously losers. Most accession countries enjoy per-capita gains of about 200-300 euros. The highest per-capita gain is for Romania and the smallest for the Czech Republic [13]. For comparison, the per-capita loss is 400-500 euros for Germany, Austria and Spain.

In the event of unilateral euroisation, domestic monetary authorities can no longer act as a lender of last resort. Costs arise since a euroising country formally gives up a central bank capable of discounting freely in times of financial crisis. This makes a priori a difference from a regulated introduction of the euro because in that case the ECB is legally the lender of last resort [15]. Nevertheless, Cohen (2003) argues that the loss of a lender of last resort can be quite easily offset on a unilateral basis. Euroisation “reduces the overall need for international reserves, since a share of external transactions that previously required foreign exchange can now be treated as the equivalent of domestic transactions. A portion of the central bank’s assets therefore could be dedicated instead to a public stabilization fund to help out domestic financial institutions under stress” (Cohen, 2003). Moreover, the absence of a lender of last resort is somewhat mitigated by a high participation of foreign institutions which may lend to their affiliates in troubled times (Backé and Wójcik, 2002).

Adopting the euro induces de facto the renunciation of independent monetary and exchange rate policies. Costs then arise from the incapacity to smooth asymmetrical shocks by using these policies when prices and wages are sluggish [19]. These costs relating more generally to the fixity of the exchange rate are typically in line with the traditional Optimum Currency Area theory. According to this theory, a low degree of asymmetrical shocks is then a sufficient condition for a successful fixity of the exchange rate.

The potential benefits resulting from the use of the euro are traditionally categorised as follows: (i) greater price stability resulting from importing credibility from the ECB; (ii) lower interest rates resulting from elimination of the exchange rate risk premium; (iii) lower transaction costs; (iv) lower exposure to speculative attacks.

Adopting the euro will result in greater price stability. The argument goes as follows: euroising countries will “import” credibility from the ECB, which has a stronger commitment to price stability than domestic central banks of acceding countries.

Adopting the euro (i.e. fixing the exchange rate against euro) will induce a fall in the interest rate due to the diminishing of the exchange rate risks. Roughly, the sole important exchange rate risk remaining will be relative to the US dollar. A lower risk premium, by bringing about lower interest rates, will foster investment and then economic growth. Moreover, it will reduce the service of external and public debt (Habib, 2001).

For acceding countries, “an early adoption of the euro would ward off the danger of currency crises during the approach to EMU” (Habib, 2002). Indeed, the sustainability of ERM-II might be difficult in the context of a fully liberalised capital account, even if sound economic policies are implemented. In terms of exchange rate arrangements, only the so-called “two corner solutions”, with either irrevocably fixed or independently floating exchange rates, are often viewed as viable options in a world of free capital mobility. In this regard, two country cases are particularly illustrative: Greece and Hungary.

Lower transaction costs due to the use of a common currency arise mainly from the elimination of exchange rate risk. Such benefits are both direct and indirect. The direct benefit is that real resources that would have been devoted to managing this risk can now be channelled to other productive uses. The indirect benefit is that trade between countries using the same currency increases. Indeed, there is some empirical evidence that trade expands between members of a currency union (Frankel and Rose, 2002; Glick and Rose, 2002) and we take it as axiomatic that trade enhances wealth. Note however that the increase in trade resulting from currency union has been probably over-estimated, as demonstrated by very recent empirical work. More than the adoption of euro, membership of the EU, by suppressing all tariff and non-tariff barriers, is proved to expand trade between members (Box 2).

Balancing economic costs and benefits of the two options is a very intricate exercise, as stressed by Backé et Wójcik (2002). However, the following general conclusion can be drawn from an overview of the empirical literature. The loss of seigniorage incomes (especially, the once-and-for-all cost) and the reduction in risk premia are the two main factors in the economic cost-benefit analysis of euroisation. As previously underlined by Habib (2001), “potential benefits from euroisation– stemming from the reduction of risk premia– could compensate for these costs”. The difficulty is that, while seigniorage incomes are rather easy to assess, estimating the decrease of risk premia is challenging.

This section is devoted to examining how EU officials can actively discourage acceding countries from adopting the euro without respecting the “institutional path” defined in the EU legislation. Finally, it appears that the credibility of (implicit) sanctions threatened by the EU officials depends crucially on the costs and benefits of euroisation in acceding countries for present EU members.

First at all, note that, to our knowledge, there is no explicit mention in the existing EU Treaties of sanctions to be taken against a country euroising unilaterally. Moreover, the draft Constitution for Europe– which will encompass all the previous Treaties– does not specify sanctions that EU could impose on a country that does not respect the “institutional path for adopting the euro”. For instance, the “exit” option is only on the initiative of the Member State itself. Nothing is specified on the possibility that a provision could be activated to force a Member State to withdraw if it does not respect the “institutional path” for adopting the euro stressed by all EU officials. Thus, from a legal viewpoint, there is a void and sanctions can be only implicit or indirect, via the activation of other provisions of the Treaties. One consequence is that, due to the intricacy of EU legislation, implicit sanctions may be difficult both to define and to implement, finally losing their credibility. Consider the following example. Given the current openness of world trade– both in goods and in assets– the euro area can hardly prevent its currency from crossing the borders of a potentially euroising country, except by resorting to capital controls. The problem is that, legally speaking, capital controls against one EU Member State (i.e. the potentially euroising country) are contrary to another pillar of EU legislation, namely the free movement of capital. Of course, present members can always refer to some European “court” (e.g. the European Commission), arguing that “euroisation” is contrary to the obligations related to EU membership, but at the risk of a long and unpredictable legal procedure.

It is important that the EU should have some sort of credible threat in order to prevent another country from adopting unilaterally the euro and “a credible threat is one which, if carried out, would harm the adopting country but would not hurt the issuing country” (Altig, 2002). To take again the issue of capital controls– but ignoring the legal problems, now assumed to have been solved– it seems to us rather difficult to imagine that EU members would not be hurt by such a decision. Capital controls would necessarily have feedback effects on other sectors of the economy, disrupting trade and then production in both euroising and present EU countries. It can be argued at first sight that, due to the small economic size of acceding countries relative to the present EU, the adverse feedback on existing EU members would be quite limited. Candidate countries as a whole accounted for 16% of total extra-EU exports and for 11.5% of total extra-EU imports in 2000 (latest available homogenous disaggregated data). Four countries (the Czech Republic, Poland, Hungary and Turkey) account individually for at least 20% of EU exports to candidate countries and 15% of EU imports from candidate countries [37]. Germany accounts for more than 40% of total extra-EU imports and exports. According to these trade data, the “problem” of euroising countries is then restricted to the four aforementioned countries and, among present members, is largely a matter for Germany. Nevertheless, going beyond the trade level, what we need to consider are the characteristics of trade between CEECs and all present members. Indeed, owing to the impetus of European Association agreements and a legislation favourable to FDI during the 1990s, the EU multinationals are currently very present in CEECs, especially for operations involving fragmentation of their productive processes [38]. As a result, EU multinationals account for a large part of both imports and exports in the acceding countries. Cutting EU multinationals off from these trade activities could be costly, if it is FDI in CEECs driven by cost considerations that makes it possible to maintain the whole chain of production, including activities (stages) in the multinational’s home country. More generally, both the private and public sectors of present EU countries have strong interests in CEECs, including financial ones. For instance, a large share of the CEECs’ external debt is purchased by EU investors (ECB, 2003b). In our view, therefore, current high degree of economic and financial integration with acceding countries explains why EU officials have low credibility in their (implicit) sanctions. Since the degree of integration is expected to increase further with the formal accession to EU, the credibility of sanctions is expected to decrease even further.

What response should the EU officials adopt faced with the wish of some countries to euroise? In the absence of credible sanctions, either a passive acceptance (like that of the United States) or a consensual euroisation are the two possible EU attitudes. At first sight, consensual euroisation would be a better option than unilateral euroisation, with advantages for both euroising countries and present members of EU.