Journal of Economic Perspectives—Volume 11, Number 4—Fall 1997—Pages 3–22

EMU: Why and How It Might Happen

Charles Wyplosz

T
he adoption of a single currency has long been a holy grail for Europe.
Since the late 1950s, various plans had been devised and shelved, as Mun-
dell (1993) describes in a brief and insightful history. But in a few sharp
steps between 1988 and 1991, bewildered Europeans saw their governments agree
to what is now known as the Maastricht Treaty.
The story begins auspiciously in 1986. The European Community emerges
from a decade-long period of little institutional progress, high inflation and rising
unemployment following the oil shocks. This is the year when three new countries
(Greece, Spain and Portugal) join the European Community and when the Single
European Act (frequently dubbed "1992," the year when it came into effect) is
adopted as an extension of the founding Treaty of Rome. The aim of the Single
Act is to plug the loopholes which limited the full mobility of people, goods and
capital within Europe. In the process, all restrictions to capital movements were
eliminated.1
This last innocuous-seeming step made a move to monetary union unavoidable.
The reason is a straightforward implication of the Mundell-Fleming textbook model
of an open economy, known in Europe as the "impossible trilogy" principle.2 This

1
Oddly, the implementation date for this part of the act was set on July 1, 1990, a year and a half ahead
of the other provisions. Recent European Community members—Greece, Ireland, Portugal and Spain
— were given grace periods.
2
The implications for Europe of this general principle, also known in Europe as the inconsistent trinity,
were first articulated by Padoa-Schioppa (1985). For a textbook presentation of the Mundell-Fleming
model, see, for example, Burda and Wyplosz (1997).

• Charles Wyplosz is Professor of Economics, Graduate Institute of International Studies,

Geneva, Switzerland, and Research Fellow, Center for Economic Policy Research, London,
United Kingdom. His e-mail address is (wyplosz@[Link]).
4 Journal of Economic Perspectives

principle asserts that only two of the three following features are mutually compat-
ible: full capital mobility, independence of monetary policy, and a fixed exchange
rate. The problem arises because, under full capital mobility, a nation's domestic
interest rate is tied to the world interest rate (at least for a country too small to
influence worldwide financial conditions). More precisely, any difference between
the domestic and world interest rate is equal to the expected rate of depreciation
of the exchange rate; that is, if interest rates are 5 percent in the domestic market
and 3 percent in global markets, this must reflect that global currency markets
expect the currency to depreciate by 2 percent this year. This is known as the
interest parity condition: it implies that integrated financial markets equalize ex-
pected asset returns, and so assets denominated in a currency expected to depre-
ciate must offer an exactly compensating higher yield.
A country that wants to conduct an independent monetary policy, raising or
lowering interest rates for the purpose of its domestic economy, must allow its
exchange rate to fluctuate in the market. Conversely, a country confronted with
full capital mobility that wants to fix its exchange rate must set its domestic interest
rate to be exactly equal to the rate in the country to which it pegs its currency; since
monetary policy is now determined abroad, the country has effectively lost monetary
policy independence.3 The alternative option of letting exchange rates float was
never acceptable to Europeans. The perception is that markets are too integrated
to allow for sizable relative price changes. The exchange rate and trade wars from
before World War II are still remembered as an example of a jack that must abso-
lutely be kept in the box.
By the time it was decided to free capital flows, the European Monetary System
(EMS) had been in place for nearly ten years. Most European Community members
had agreed in early 1979 to set up a system of fixed bilateral exchange rates with
fluctuation bands of ±2.25 percent around the declared central parity (±6 percent
for Italy and, briefly, the United Kingdom). Member central banks were committed
to intervene jointly to defend the parities, in principle with no limit. When it was
felt that existing parities had to be changed, the decision had to be taken by con-
sensus. By the late 1980s, the EMS was commonly hailed as a major success, credited
with the relative stability of intra-European real exchange rates during the turbulent
post-Bretton Woods period (Begg and Wyplosz, 1993). This is illustrated in

3
In algebraic terms, the interest parity condition, where i is the domestic interest rate, i* is the global
rate, e is the expected rate of depreciation of the exchange rate (in logs), and t is an index of time
periods, is:

il– i*l = E l ( e l + 1 ) – el.

A small country which pegs its exchange rate El(el+1) = el = e0, where e0 is the peg, can no longer choose
the level of its own interest rate. Only by letting the exchange rate fluctuate can it control the interest
rate, and then el becomes endogenous in the interest parity equation. This reasoning ignores risk aversion
which gives rise to a risk premium term. Among developed countries the risk premium is known to be
small and volatile.
 Charles Wyplosz 5

Figure 1
Bilateral Exchange Rates

Figure 1 which presents bilateral exchange rates deflated by consumer price indices
(the following conclusions emerge irrespective of the choice of country pairs and
price indices). Contrast the left panel which shows intra-European real exchange
rates with the right panel which shows real exchange rates vis-á-vis the U.S. dollar.
Currencies with normal allowed fluctuations, like the French franc and the
deutsche mark, displayed remarkably low volatility in comparison to floating
exchange rates. Even where larger margins were allowed, like in Italy, quarter-to-
quarter real exchange rates are still less volatile than with floating rates. There is
no economic reason for the real exchange rate to be constant in the long-term, of
course. However, among the OECD countries, multi-year fluctuations around the
long-term trend suggest that most of the observed changes are temporary and do
not correspond to structural shifts.
Perhaps blinded by the success of the EMS, leading European policymakers
did not perceive that the freeing of capital flows meant the end of monetary policy
independence in all but one EMS country. By the late 1980s it had become obvious
that the Bundesbank, Germany's central bank, was setting monetary policy for Eu-
rope as a whole. One reason for this evolution was relative economic size (further
increased by unification following the fall of the Berlin Wall in late 1989). In ad-
dition, the Bundesbank had acquired a strong reputation for fighting inflation and
keeping its currency strong. For countries where inflation was the number one
target, adopting tough monetary conditions under the Bundesbank leadership was
in fact welcomed. Small countries, like the Netherlands, had already given up mon-
6 Journal of Economic Perspectives

etary independence. Among the larger ones, the United Kingdom was outside the
fixed exchange rate mechanism and therefore could retain monetary policy
independence.
However, other larger European nations like France, Italy, and Spain, gradually
realized that they had lost control of their domestic monetary policy. They con-
cluded that the only way through which they could regain some influence over
their monetary policies was to create a broader European monetary institution
which would supersede the Bundesbank, and in which they would have a voice.
Naturally, since Germany was being asked to sacrifice one of its most valued insti-
tutions for the sake of Europe, it was going to ask a lot in return. In particular,
Germany was bound to require that this new European monetary institution offer
strong guarantees of price stability. From the very beginning, Europe's future cur-
rency would have to be as strong as the deutsche mark. This would mean explicit
institutional safeguards and exacting startup conditions. The negotiations leading
to the Maastricht Treaty would bear the birthmark of this situation: what Germany
asks, Germany gets, provided that it gives up the Bundesbank.

The Maastricht Treaty

The Maastricht Treaty updates and incorporates the 1957 Treaty of Rome, the
founding act of the European Community, and incorporates the Single European
Act implemented in 1992 (free movement of goods, people, and capital). The treaty
has been formally ratified by all member countries. With the Maastricht Treaty,
Europe ceases to be called the European Economic Community and becomes in-
stead the European Union or EU, which involves both economic and political un-
ion. The economic component of the treaty mainly involves the adoption of a single
currency. The political component has been left rather vague, hinting at an evo-
lution towards joint defense and foreign affairs.
The treaty includes a detailed timetable for the adoption of a single currency.
It sets in motion a gradual convergence process, espousing the view that the adop-
tion of a common currency is just the cherry on the sundae, the last step in a process
through which national currencies become indistinguishable from the deutsche
mark. It is formally structured around three stages (Thygesen, 1993). The first stage
began in 1992 with the formal ratification of the treaty. During the second stage,
started in January 1994, national central banks must be given formal independence
and cease to grant direct loans to their nation's treasuries. The shift to the second
stage also coincides with the establishment of the European Monetary Institute
(EMI), with two main functions. One is to prepare the creation of the European
Central Bank, whose statutes and mission are actually laid out in the Maastricht
Treaty. The other function of the EMI is to oversee the "convergence criteria"
which will be used to decide which countries are ready to enter the monetary union,
marking the beginning of Stage III. This may happen as soon as a sufficient number
of countries meet the convergence criteria, and must happen by January 1, 1999.
 EMU: Why It Might Happen 7

The first formal review which took place in December 1996 concluded that a ma-
jority of countries did not satisfy the criteria.
What are these criteria? The underlying notion is that unless countries enter
the single currency with similar inflation rates and fiscal positions, the single cur-
rency will be unsustainable. Three conditions deal with monetary convergence.
First, the inflation rate of any country joining the single currency must be within
1.5 percentage points of the average of the three lowest rates in Europe. Second,
the long-term interest rate in a country joining the single currency must not exceed
by more than 2 percentage points the interest rates observed in the three countries
with the lowest inflation rates, on the grounds that high long-term rates reflect high
expected inflation. Third, the exchange rate must have remained within the normal
bands of the existing EMS "without severe tensions" for at least two years. Two
other criteria concern fiscal policy. They set ceilings on the ratios of debt/GDP (60
percent) and deficit/GDP (3 percent) ratios. At the time of the signing of the
Maastricht Treaty in 1991, only Luxembourg—which does not have a currency of
its own—could meet the five criteria.
Yet the wording of the treaty leaves some room for flexibility. For example, the
60 percent ceiling can be interpreted as a target if "the ratio is sufficiently dimin-
ishing and approaching the reference value at a satisfactory pace" (art. 104c, b).
In addition, compliance will be decided by the heads of state upon receiving reports
from the European Commission and the EMI, and a recommendation (not a de-
cision) by the Commission, which is notoriously supportive of economic and mon-
etary union.
The Maastricht Treaty had left a number of issues pending. Most of them
concern the political side, but some also concern the actual operation of the mon-
etary union. The "excessive deficit procedure" issue has been settled in June 1997.
This procedure makes permanent one of the entry convergence criteria, the
3 percent deficit/GDP ceiling. It defines the "exceptional conditions" under which
a country may be temporarily allowed to breach the ceiling, and it specifies how
noncompliant countries will face first private, and then public reprimands, before
being fined. Progress has also been made on symbolic matters: the new currency's
name will be "euro" and the European Central Bank will be established in Frank-
furt, Germany.
On all of these issues, the German view has prevailed. The excessive deficit
procedure is the one presented by Germany, and initially rejected by a vast majority
of countries as excessively restrictive. The name of the currency itself reflects the
German rejection of ECU, acronym for European Currency Unit and the name of
an ancient French currency, although it is explicitly referred to in the Maastricht
Treaty. German influence has not only affected the currency name and location; it
is also Germany that insisted on the long transition process and the controversial
convergence criteria. Moreover, the statutes and objectives of the European Central
Bank remarkably resemble those of the Bundesbank: strong independence from
government, responsibility clearly limited to price stability, no explicit involvement
in bank supervision, and no lender-of-last-resort function.
8 Journal of Economic Perspectives

Germany will again prevail when it comes to selecting the countries which
qualify for membership in the single currency. That decision will be taken by the
heads of state in spring 1998, with voting weights determined by country size (a
combination of population and GDP). Once chosen as fit, a country must join the
Economic and Monetary Union (EMU), even if it does not wish to, with the excep-
tion of Denmark and the United Kingdom who made opting out a condition of
ratifying the treaty. Thus the Maastricht Treaty envisions a "two-speed" Europe,
with a "core" of EMU members and a "periphery" of countries either rejected or
opting out. Much of the debate revolves around the initial list of members. Will
EMU start as a narrow deutsche mark zone (Germany, France, Belgium, Luxem-
bourg, the Netherlands, Austria, Ireland)? Will the "Club Med" countries (Italy,
Spain, Portugal) also join, despite a reputation for tolerating inflation and deficits?
Will the Nordic countries (Sweden, Denmark, Finland) want to join? The UK has
already let it be known that it will stay out and Greece is not really trying.

Is Europe an Optimal Currency Area?

The decision to adopt a single currency is the outcome of constrained opti-

mization. The constraint is the impossible trilogy: given the freedom of capital flows,
the choice is between freely floating exchange rates and monetary union. The as-
sessment is that monetary union dominates a free float. This assessment is based
on the experience with floating exchange rates since 1973: wide and long-lasting
fluctuations (20 to 50 percent over three to five years) are just not compatible with
fully open markets and the complete removal of border posts. While that assessment
is open to debate (but seldom challenged so far), the discussion on the intrinsic
desirability of the monetary union is moot as long as it ignores the constraint.
Yet, it is probably unavoidable that the question be asked whether EMU is
welfare-increasing per se.4 This question has led to a revival of the literature on
optimum currency areas following seminal works by Mundell (1961), McKinnon
(1963) and Kenen (1969). Recent efforts have gone into providing formal models
which confirmed the early insights (Bayoumi, 1994; Ricci, 1996). However, most of
the work has attempted to size Europe up against the Mundell-McKinnon-Kenen
criteria. By these particular standards, the case for Europe as an optimal currency
area is lukewarm at best.
The (unconstrained) optimum currency area literature establishes the condi-
tions under which two or more countries could share the same currency without
seriously adverse consequences. It assumes that the nominal exchange rate has real
effects; otherwise, there is no cost in a nation's giving up its own currency. In

4
Some studies have attempted to measure directly the costs and benefits from EMU. Bean (1992) con-
cludes that these attempts have failed to come up with tangible answers. The recent report by the Swedish
Government Commission on the EMU (1997) provides an excellent and exhaustive review; it concurs
that current knowledge prevents any sharp conclusion one way or the other.
 Charles Wyplosz 9

particular, the exchange rate is a policy instrument which can affect relative prices
such as the real wage paid by producers, the ratio of traded to nontraded goods
prices, or the ratio of export to import goods prices. As one example of where this
tool could be useful, consider the case where some exogenous shock requires that
relative domestic to foreign prices change. Such an adjustment can plausibly be
made easier and faster through the exchange rate, rather than by changing nominal
prices throughout the economy or through migration of the factors of production
from one sector to another.
The three criteria proposed in the literature are those features which make
adjustment through exchange rates less effective or less compelling. One criterion
is openness to mutual trade; greater openness means that most prices are being
determined on markets at the area level, which reduces the ability of the exchange
rate to alter significant relative prices. A second criterion is diversification of indi-
vidual economies; a more diversified economy is less likely to suffer country-specific
shocks, which makes its own exchange rate a less useful tool. Finally, the third
criterion is mobility of inputs across the area, especially labor. Greater mobility
allows an economy to deal with asymmetric shocks through migration, lessening
the need for adjustment through exchange rate changes.
On the openness criterion, Europe scores rather well. Measuring openness by
looking at exports as a share of GDP, the United States and Japan score 11 percent
and 9 percent, respectively. Larger European economies like Germany, Italy,
France, and the United Kingdom all have export/GDP ratios above 20 percent, and
smaller EU economies like Ireland and Belgium have export/GDP ratios above 70
percent. It makes sense that the smallest European countries are traditionally warm
supporters of monetary union. Because of their extreme openness to foreign trade,
relative prices in their economy are set on world markets, and the exchange rate is
a less useful policy tool.
As to the second criterion, European economies are found usually to be well-
diversified. Countries with important endowments in natural resources, like the
Netherlands and the United Kingdom with their oil and gas resources, stand apart,
but only slighdy so. A wide body of research looks at the risk of country-specific
(asymmetric) shocks. One set of studies investigates co-movements of key macro-
economic variables like GDP, unemployment, inflation, or the current account
balance across European countries (Cohen and Wyplosz, 1989; Weber, 1990).
Other studies compare shocks across regions with shocks across countries (de
Grauwe and Vanhaverbeke, 1993; von Hagen and Neumann, 1994). The general
message is that there is more co-movement in macroeconomic variables among
European countries than between individual European countries and the United
States or Japan. Further studies attempt to separate out domestic from external
shocks, and demand from supply shocks. The underlying argument is that demand
shocks are at least partly due to divergence in monetary policy which will be less
prevalent in EMU—so attention should focus on supply shocks. Bayoumi and Ei-
chengreen (1993), for example, find more asymmetric supply shocks across Europe
than across the United States, although they identify a more coherent group of
10 Journal of Economic Perspectives

core countries around Germany. They suggest that Europe is less-suited to be an

optimum currency area than the United States.
Work on the labor mobility criterion clearly suggests that Europe is not an
optimum currency area. For example, looking at the United States as a prototype
monetary union, Blanchard and Katz (1992) find that when a particular region is
hit by an adverse shock, a large proportion of the subsequent drop in employment
is matched by labor migration. Applying the same approach to European regions,
Decressin and Fatas (1995) find less labor mobility and longer-term effect on re-
gional unemployment, confirming a similar finding by Eichengreen (1991). Two
caveats are in order, however. First, the evidence is that the lack of labor mobility
is not a national but a regional phenomenon in Europe (Eichengreen, 1993). It
affects regions within existing nations of Europe, and there is no reason why mon-
etary union would make things worse. Second, both the occurrence of shocks and
labor mobility may change as economic integration proceeds. Krugman (1993)
conjectures that economic integration leads to increased regional specialization. In
that case, the situation will worsen as the incidence of asymmetric shocks will in-
crease. Frankel and Rose (1996) empirically reject Krugman's conjecture as they
find that integration leads to more diversification. In that case, the criteria for an
optimum currency area are endogenous. It then comes as no surprise that the
United States, which has shared the same currency for a century, appears better
suited for a single currency than does Europe.
In the end, we need not be impressed by the result that Europe is not as much
an (unconstrained) optimum currency area as the United States. The choice is not
between EMU and heaven. It is between EMU and freely-floating exchange rates,
with possibly poorly coordinated monetary policies, within an area gradually be-
coming as tightly integrated as the United States. Would the United States have
passed the currency area tests a century ago? And had it failed, all things considered,
was it a mistake for the country to adopt a single currency?

Convergence: Will Tough Criteria Backfire?

One striking feature of the Maastricht Treaty is that it anticipates a long eight-
year phase from the passage of the treaty in 1991 to the deadline for a single
currency by 1999. This long phase-in was the result of a conflict between two com-
peting views.
One view argued that monetary union would be sustainable only if those coun-
tries that joined had first achieved a low level of inflation and had resolved fiscal
imbalances. This position is commonly referred to as the "economist's view," al-
though it does not seem to have been fully articulated in the professional literature.
However, it was popular among the monetary authorities; for example, the Bun-
desbank championed it under the name of "coronation approach," seeing the shift
to monetary union as the last step of successful efforts to eradicate inflationary
 EMU: Why It Might Happen 11

behavior. Economic and monetary union was to be born in a land dedicated to a

culture of price stability.
The opposing view, generally referred to as the "monetarists' view," had the
favor of most academic economists.5 Their argument was that the creation of a new
currency with its own independent central bank would radically alter the wage and
price mechanisms, inflation trends, and the incentives of national governments
when they decide on fiscal policies. In this view, which is rooted in the Lucas cri-
tique, pre-monetary union behavior of both the public and private sectors is a bad
predictor of their behavior once the single central bank is in place. Instead, what
is needed in the monetarist view are solid institutions, chiefly central bank inde-
pendence. Other convergence criteria create pain with no assured gain.
Predictably, the "economist" view favored by central bankers won out over the
"monetarist" views of academic economists. It is impossible to say what would have
happened if EMU had started fairly promptly after ratification of the Maastricht
Treaty in 1991. However, what is known is that the period dedicated to convergence
has been especially agitated. Even before the Maastricht Treaty could be ratified, a
series of exchange rate crises forced Italy and the United Kingdom out of the EMS.
After severe currency realignments, the "normal" ±2.25 percent bands of fluctu-
ation were widened to ±15 percent, effectively marking the end of the system as
initially designed and intended. By mid-1997, about one year before the scheduled
selection of the countries which will start EMU, there is considerable pressure to
postpone the starting date. The surrounding debate well illustrates the view that
there is never likely to be a time when all countries can meet the exacting conver-
gence criteria.
Of the criteria set in Maastricht, those mandating inflation convergence have
proven relatively easy to achieve.6 However, the budgetary criteria—that the
debt/GDP must not be above 60 percent nor the deficit/GDP exceed 3 percent—
are more challenging, as Table 1 documents. Why after such a long period of
convergence are the budget criteria still some way off? Part of the problem is that
the tight monetary policies aimed at meeting the inflation criteria have helped
create a slow-growth climate for Europe in the 1990s, with double-digit unemploy-
ment rates and no net job creation since the beginning of the decade. While this
effort has made it possible to achieve inflation convergence, it has also reduced tax
revenues, causing deficits that will not go away and forcing governments to adopt

5
For a statement of the "monetarist" view, see Begg et al. (1991). The rationale of the terminology of
"economists" versus "monetarists" is unclear. It goes back to earlier debates on economic and monetary
union in the 1970s, well summarized in Mundell (1993).
6
However, the jury is still out for the criterion concerning the long-run interest rate, which is not to
exceed the average of the three lowest-inflation countries by more than 2 percentage points. Since long
rates incorporate market expectations of inflation, they are affected by the probability of joining the
monetary union. This opens up the possibility of multiple equilibrium: if the markets believe that a
country will not join, they may expect monetary policy relaxation and rising inflation, and set high
interest rates which indeed rule out EMU membership. Conversely, an expectation that a country will
join may bring down long-term rates, thus allowing the country to meet this criteria for convergence.
12 Journal of Economic Perspectives

Table 1
The Maastricht Budget Criteria as of Mid-1997
(percent of GDP)

further policies of fiscal contraction. This vicious cycle is jeopardizing monetary

union both by making the fiscal targets more difficult to achieve and by undermin-
ing public support. The situation is now a gamble: either a country reaches EMU
and is able to relax after having indeed put its fiscal house in order, or it fails entry
(or EMU does not take place at all) because excessively restrictive economic policies
have deepened the budget deficit.

Monetary Union and Fiscal Discipline

The inclusion of restrictions on fiscal policy in a treaty which, after all, aims at
monetary union, is a source of considerable debate. Before the Maastricht Treaty,
most academic analyses emphasized that national fiscal policy would have to be-
come more active to compensate for the loss of the exchange rate instrument.7 The
opposite approach, that monetary union requires fiscal policy restraint, is grounded
in the view that excessive budget deficits may lead to eventual monetization of the
debt (Sargent and Wallace, 1981). Monetary authorities were clearly concerned by

7
For example, see the papers by Begg, Masson and Melitz, and Wyplosz in European Economy, Special
Edition No. 1, 1991.
 Charles Wyplosz 13

high debts in some countries, especially in Italy, whose public debt represents some
18 percent of Europe's GDP. They feared that an explicit or implicit lender-of-last-
resort function might force the European Central Bank to step in and indirectly
monetize a country's public debt if banks faced a financial crisis in the wake of a
default. This concern is reflected in the budgetary criteria for EMU membership
and in the "excessive deficit" procedures designed to enforce fiscal rectitude once
in the monetary union.
While it is difficult to disagree with the view that fiscal policy ought not to
jeopardize monetary and financial stability, how to provide the incentives for ap-
propriate fiscal policy is open to debate. The debate implicitly revolves around one's
view of the ability of fiscal policy to play a macroeconomic stabilizing role. It also
hinges on the ability to define at the time a deficit is enacted that it is "excessive."
In principle, the proper answer must be in terms of "sustainability," since by def-
inition, unsustainable debt buildup will eventually have to be reversed. Fiscal policy
sustainability is often associated with stationarity of the debt, usually defined as a
stable debt/GDP ratio. In fact, the proper definition of sustainability would hold
only that the state will remain solvent, a definition that emphasizes the future be-
havior of fiscal authorities. By emphasizing future behavior, this view of sustainabi-
lity also implies that information from the past does not reveal what a country will
do after it is inside EMU, and that rules for fiscal rectitude must affect future fiscal
policies. A workable definition of sustainability along these lines is a tall order.
The Maastricht approach, relying on arbitrary quantitative limits, is quite un-
sophisticated.8 The 3 percent annual debt/GDP rule corresponds to what is called
the "golden rule" in Germany: governments may only borrow to pay for investment
spending, and it turns out that governments usually dedicate about 3 percent of
GDP to such spending. Even if one ignores doubts about the 3 percent estimate
itself, the rule is naive at best; it ignores socially productive spending like education
which is classified as consumption, while it may include ill-designed investment
spending. The 60 percent debt/GDP rule was chosen because it was the average of
EU countries when the Maastricht Treaty was being negotiated, with not even the
pretense of any deeper economic justification.
Yet Europe is not alone in adopting quantitative limits for fiscal policy. How
does it work elsewhere, where a unique central bank coexists along with several
fiscal authorities? In the United States, for example, states must operate under
balanced budgets, borrowing money only by issuing bonds for explicit capital
projects. But the comparison must be handled quite carefully. In true federations,
the central government is as large as the lower-level governments, and is in charge
of macroeconomic stabilization. In Europe, in contrast, the equivalent of a central
government is the European Commission, which is not allowed to run deficits and
whose spending represents a mere 2 percent of the Europe Union's gross domestic
product.

8
For a critique of the entry criteria ceilings, see Begg et al. (1991) and Buiter et al. (1993).
14 Journal of Economic Perspectives

The size and role of a powerful central government matters for two main rea-
sons. First, several studies have shown that in federal states, the center smooths out
income fluctuations through redistribution from regions in good economic shape
to regions undergoing a recession. This function operates automatically through
the federal budget, the result of a combination of welfare support and income taxes
(Sachs and Sala-i-Martin, 1992; Bayoumi and Masson, 1995; Pisani-Ferry et al.,
1995). In this setup, it can make sense to limit the stabilization role of sub-central
authorities. Second, quantitative fiscal restraints at some levels of government can
actually encourage the buildup of debts at other levels, according to evidence from
von Hagen and Eichengreen (1996). The problem occurs when fiscally irrespon-
sible lower-level governments refuse to borrow and can bait the federal authorities
into rescuing them. In Europe, a central government with powerful redistribution
and stabilization authority is not likely within the foreseeable future. Consequently,
Europe needs national-level stabilization policies much more than individual U.S.
states do, and there is no risk that national governments will conduct irresponsible
fiscal policies in an attempt to extract transfers from a penniless center.
Are there less coarse methods than quantitative limits of providing govern-
ments with effective incentives against fiscal irresponsibility? One attractive ap-
proach would be to rely on financial markets to impose discipline. In a single cur-
rency area, interest rates no longer reflect a country's sovereign risk. Instead, they
reflect the risk category of borrowers, be they fiscal authorities (a municipality in
the United States, a province in Canada, or a government in Europe) or private
borrowers. To the extent that markets price risk correctly, the demand for public
debt of various governments could act as both a barometer and a constraint. If a
country lets its debt grow and there is an enhanced risk of default, markets should
react by downgrading their evaluation and by increasing the interest rate at which
new debt is being financed, until fiscal authorities see it to be in their best interest
to curtail the deficit.
However, history suggests skepticism about the ability of markets to impose
discipline in this way. For one, markets tend to throw good money after bad for a
time (Eichengreen and Portes, 1987). When markets do react, it is often too late
and too violently. They abruptly cut financing, making it impossible for the gov-
ernment to borrow further and bankrupting large bondholders, among them com-
mercial banks and other financial institutions. This leads to a scenario where central
banks may feel compelled to monetize (part of) the debt.9
This is presumably why the Maastricht Treaty includes a no-bailout clause
which explicitly forbids the rescue of one government either by its fellow members
or by community institutions, including the European Central Bank. In this way,
fiscal misbehavior becomes a strictly national issue with no union-wide implication
and fiscal restraint is unnecessary. Yet Germany has argued that the no-bailout
clause cannot be fully credible, that any rule can always be circumvented.

9
That this mechanism bears strong resemblance to the events that provoked hyperinflation in Germany
in 1922–23 is not irrelevant for an understanding of the Maastricht Treaty.
 EMU: Why It Might Happen 15

In the end, the explicit fiscal restraints embodied in the excessive deficit pro-
cedure can be seen as insurance against a remote risk that European institutions
would be compelled to monetize some nation's out-of-control debts. This insurance
scheme may turn out to be very costly in terms of the ability to run countercyclical
policies.

EMU and the Rest of the World

The potential for the euro to replace the U.S. dollar as the world's premier
currency is one of the understated motivations of EMU. In part, the desire is a
symbolic one; it is the belief that "great powers have great currencies" (Mundell,
1993, p. 9). In part, it is a hope to reap seigniorage, although U.S. benefits from
seigniorage are worth only about 0.2 percent of GDP (Alogoskoufis and Portes,
1992). The usual criteria for becoming the world's lead currency are measures like
size (GDP or the share of world trade). By these measures, the prospects for the
euro to challenge the dollar are favorable but not overwhelming. For example,
Europe's international trade with non-European nations will not exceed by much
Germany's current level of foreign trade—once intra-European trade is netted out
(Hartmann, 1996). Also, history teaches that it takes time for a reserve currency
to change (Eichengreen, 1989; Mundell, 1993). To overcome its handicap relative
to the incumbent U.S. dollar, the euro must discover some absolute
advantage.
One potential advantage is likely to be greater price stability. As a currency
expected to follow a long-run trend of appreciation, the euro will be a currency
that stores value better than the alternatives. This prediction derives from the con-
stitution of the European Central Bank, which makes it more independent and
more focused on price stability than the U.S. Federal Reserve. If anything, the
constitution is even stricter than that of the Bundesbank, so that Europe's economy
will be more stable than Germany's (Masson and Turtelboom, 1997). A counter-
argument is based on politico-economic considerations. The board of the European
Central Bank will be composed of representatives of all member countries. With
the one-man one-vote principle, Germany's weight will be no larger than that of
Belgium or Italy. The constituencies of the European Central Bank will not share
the German allergy to even moderate inflation.10 In theory the outcome may differ
from the wishes of the median European voter, and the bias can go in either di-
rection (Alesina and Grilli, 1992). Ultimately, this counterargument is not fully
convincing.

10
This is another reason why the Bundesbank has advocated a long convergence process: to provide for
a period of deep conversion to a "stability culture." In a perceptive comment on Buiter et al. (1993,
p. 97), Frankel interprets the Maastricht convergence process as a "test of will," referring to Buddhist
traditions: "A meditating neophyte is supposed to learn to refrain from responding to a flea by scratching
it, just as a political region is supposed to learn to refrain from responding to a local downtick in demand
by lowering interest rates."
16 Journal of Economic Perspectives

A second potential advantage for the euro could be the depth and cost-
efficiency of financial markets. The market for the euro and euro-denominated
assets could be the world's largest, depending on whether the city of London shifts
to the euro. Yet the location and prominence of markets relies increasingly less on
regional considerations and more on the regulatory environment. Europe will have
to fight its own heavy-handed approach and powerful lobbies if it wants the euro
to become the world's currency.
Thus, the best bet is that, for a long while at least, the dollar's supremacy will
remain. Still, the creation of the euro is bound to affect international monetary
relations. Will it lead to more or less instability on exchange markets? Two argu-
ments suggest more instability. First, if the U.S. dollar has been acting as a market
leader on exchange rate markets, the shift to a situation of bargaining between
more equal partners is likely to create greater volatility. Second, while the fairly
open economies of Europe are now keenly interested in stabilizing world curren-
cies, a euro zone would join the United States and Japan as giant economies less
inclined to give up domestic policy objectives for the sake of exchange rate coor-
dination. However, the opposite view is that moving from G-7 to G-3 should make
it easier to negotiate methods for reducing volatility in exchange rates (Alogos-
koufis and Portes, 1992; Goodhart, 1993; Kenen, 1996). In the end, little should
change when the European Central Bank steps in the shoes of the Bundesbank as
the master of the EMS exchange rate.
Finally, what will be the impact of economic and monetary union on the In-
ternational Monetary Fund? One view is: nothing much. Each country will retain
its existing role. In its annual review exercise, the IMF will have to take account of
the fact that monetary policy is no longer a national responsibility, but that is already
the case for other monetary unions in Africa and the Caribbean. However, a more
entertaining scenario, if unlikely, envisions EMU countries merging as a single IMF
member. Not only would Europe cast the largest number of votes and challenge
U.S. dominance, but it could invoke the agreements' article that states "the prin-
cipal office of the Fund shall be in the territory of the member having the largest
quota" and request that the IMF move from Washington to Madrid, Frankfurt, Paris
or Amsterdam.

The Early Steps: What to Watch For

The Treaty of Maastricht sets a clear timetable: a single currency will come into
being no later than 1999. It may seem that all that remains is to watch the count-
down before lift-off. Nothing is further from the truth. Power in the boosters is not
assured; last minute checks reveal a number of blinking red lights; and politico-
economic pressures are building up to dangerous levels. Public support for the
euro is lukewarm at best. It is largely incomprehensible. As a symbol of national
belonging, it is desirable to some and threatening to others. As the time to launch
 Charles Wyplosz 17

draws near, popular anxiety is tending to rise. In virtually every country, politicians
are making capital out of their opposition to monetary union.
Must EMU start by January 1999? Several loopholes exist for sidestepping the
deadline. First, it is understood that monetary union will not exist without both
Germany and France. This gives each of these countries veto power that they can
exercise by missing the convergence target. In fact, it appears that both are likely
to miss the targets narrowly, which will inevitably lead to negotiations about their
situation. Second, certain provisions of the treaty could be twisted to postpone
startup beyond 1999, although it would be a farfetched interpretation of the treaty.
By June 1998, Europe's heads of state must agree on the list of the passengers
of the first mission. Many countries will not fulfill the formal criteria, so the decision
will have a degree of arbitrariness relying on flexibility in the precise wording of
the treaty. In anticipation, adversaries of economic and monetary union are calling
for a postponement. In fact, any delay would feed doubts that convergence can be
achieved and reduce chances of success. In that case, speculators could well unleash
new attacks on exchange rates, which might make any transition to a single currency
even more difficult to achieve.11
Immediately after the list of members is drawn up, final preparations will start.
At least one unresolved issue has been identified. Legal restrictions imply that the
rate at which currencies will be converted into euro on January 1, 1999, must be
those observed at the closing of markets on December 31, 1998. This creates the risk
of major exchange market instability in the time leading up to that date, as traders
will need to form a view of what the authorities are trying to achieve. Moreover, at a
time of high unemployment and with policy settings driven by the need to meet
contractionary convergence criteria, some countries may be tempted to secure a
temporary competitive advantage by entering monetary union with an undervalued
currency. Solutions for tying down the issue ahead of time still remain to be adopted
(Begg et al., 1997; Obstfeld, 1997).
According to the Maastricht Treaty, the European Central Bank will come into
existence soon after July 1998. It will have to coexist for six months with national
central banks due to become its subsidiaries. From January 1999, the European
Central Bank will operate only in euros, as will the financial markets. At the retail
level, national currencies will continue to circulate and remain sole legal tender
until July 2002, but will be legally considered as (horrendous six-digit) fractions of
the euro. The euro itself will be finally introduced for retail transactions in January
2002 (probably), opening up a switchover period of six months. Thereafter national
currencies will be redeemed in euros for periods to be set by national legislation.
The three-year overlap is bound to raise endless practical issues, not the least of
which is that it may be difficult for governments and citizens to realize that the

11
In any case, speculative attacks are expected against those countries which are not admitted to the
single currency. Such attacks could be minimized if information is gradually leaked to the markets well
in advance and if new dates for entry are clearly set along with a clear signal that the next decision will
be positive.
18 Journal of Economic Perspectives

European Central Bank is solely in charge after 1999, and all surviving currencies
are mere subdivisions of the euro with a fixed and irrevocably set conversion rate.

Conclusion

Currencies and nations normally coincide. Europe is set to attempt an original

experiment. Is it going to work? Is it even going to happen? The fact that a year
before lift-off, doubts remain about the likelihood that EMU will start, or will start
on time, is testimony to the fact that there can be no firm answers to these questions.
Yet some simple observations can take us a long way.
The Maastricht Treaty is the fundamental act on which Europe rests. It is an
international treaty, formally ratified by all European Union countries, and it su-
persedes national legislation. Giving up EMU would throw up more than just mon-
etary union. It would create a situation of deep political crisis with unpredictable
consequences. For that reason alone, the bet is that EMU will be on, on time.
Is the logic behind monetary union only political? Quite the contrary. The
political aim of a single currency has been pursued relentlessly by its advocates since
the late 1950s; several explicit attempts failed because economic conditions were
not ripe. The Maastricht Treaty only came about because the lifting of capital con-
trols had reduced the alternate options to just two unpalatable extremes: either
allow exchange rates to float freely or accept the complete domination of Ger-
many's Bundesbank over Europe's monetary policy.
Freely floating exchange rates are not compatible with a completely borderless
economic area. They carry the germs of protectionist pressure and financial insta-
bility which threaten economic integration. As for dominance by the Bundesbank,
it has been largely beneficial over the last decade, chiefly because inflation has been
eliminated. Yet there have been costs: lasting double-digit unemployment, major
policy mistakes that led to the currency crises of 1992–93, and continuing disagree-
ments over the objectives of the Bundesbank. The current situation is not sustain-
able because it entails a fundamental contradiction. On one hand, the Bundesbank
derives its leadership from a reputation of undeterred commitment to price stability
in Germany. On the other hand, long-lasting leadership requires that all of Europe's
economic conditions be taken into account, which is against the Bundesbank's
constitutional duty to Germany. Tinkering with the Bundesbank's constitution is
not only politically impossible, but doing so would also undermine its credibility
and its ability to lead. In this setting, EMU emerges as the best possible economic
solution.
Assessing the costs and benefits of monetary union quantitatively is both frus-
trating and useless. It is frustrating because, frankly, as economists we are unable
to compute them with any precision, and we owe it to the profession to admit so
in public. Our understanding of monetary and exchange rate policy is regrettably
limited, and the lack of a precedent leaves us with more conjectures than certainties.
Moreover, quantitative estimates are useless unless they are sized up against the
 EMU: Why It Might Happen 19

costs and benefits of the relevant alternatives, which is equally beyond our current
ability. The best that can be done in this situation is to gain an understanding of
where the costs and benefits are likely to reside.
The direct benefits come in the form of reduced transaction costs and reduced
uncertainty, possibly including additional transparency in competition. Such effects
are likely to be small, but not trivial. Direct benefits also include lower real interest
rates for countries where a sizable currency risk premium exists. Indirect benefits
come from the institutional arrangements that accompany EMU. The broadening
of central bank independence from political control would not have happened
without EMU, and with it comes the realization that international competition is
not achieved through lobbying for exchange rate manipulation.
More ambiguous is the role of the fiscal restraints, both the entry conditions
and the excess deficit procedure. In most countries, these restraints have promoted
long-needed efforts at coming to grip with unsustainable deficits. At the same time,
the insistence on price stability along with the adoption of rigid and arbitrary cri-
teria of fiscal rectitude have already played a role in deepening and lengthening
Europe's phase of slow growth, with huge costs in terms of unemployment and
social suffering. The risk now is of more of the same in the early EMU years. As
already noted, these costs are the consequence of EMU's parenthood: Germany
could not be expected to give up its famed deutsche mark without extensive guar-
antees. These demands could not be turned down and have probably become ex-
cessive. However, once monetary union exists, many arrangements can be changed.
Right now, Europeans are biting the bullet and looking beyond the 1999 horizon.

• For useful comments and advice, and without any implication, thanks to the editors, Alan
Krueger, Brad De Long, and Timothy Taylor, as well as to Benoît Coeuré, Barry Eichengreen,
Hans Genberg, Paul de Grauwe, Paul Masson, Jacques Melitz, Maury Obstfeld and Richard
Portes.

References
Alesina, Alberto and Vittorio Grilli, T h e Eu- Currencies in a Tripolar World," In Canzoneri,
ropean Central Bank: Reshaping Monetary Poli- Matthew B., Vittorio Grilli, and Paul R. Masson,
tics in Europe." In Canzoneri, Matthew B., Vit- eds., Establishing a Central Bank: Issues in Europe
torio Grilli, and Paul R. Masson, eds., Establishing and Lessons from the U.S. Cambridge: Cambridge
a Central Bank: Issues in Europe and Lessons from the University Press, 1992, 273–302.
U.S. Cambridge: Cambridge University Press, Bayoumi, Tamim, "A Formal Model of Opti-
1992, 49–85. mum Currency Areas," International Monetary
Alogoskoufis, George, and Richard Portes, Fund Staff Papers, December 1994, 41, 537–54.
"European Monetary Union and International Bayoumi, Tamim, and Barry Eichengreen,
20 Journal of Economic Perspectives

"Shocking Aspects of European Monetary Un- Mark P. Taylor, eds., Policy Issues on the Operation
ion," In F. Torres, Francisco and Francesco Gia- of Currency Unions. Cambridge: Cambridge Uni-
vazzi, eds., The Transition to Economic and Monetary versity Press, 1993, 111–29.
Union. Cambridge: Cambridge University Press, Eichengreen, Barry, "Hegemonic Stability
1993, 193–240. Theories of the International Monetary System."
Bayoumi, Tamim, Morris Goldstein, and Geof- In Cooper, Richard N., et al. (eds.) Can nations
frey Woglom, "Do Credit Markets Discipline Sov- agree? Issues in International Economic Cooperation.
ereign Borrowers? Evidence from the U.S. Studies in International Economics series. Washing-
States," Journal of Money, Credit, and Banking, No- ton, DC: Brookings Institution, 1989, 255–98.
vember 1995, 27, 1046–59. Eichengreen, Barry, "One Money for Europe?
Bayoumi, Tamim and Paul R. Masson, "Fiscal Lessons from the U.S. Currency Union," Eco-
Flows in the United States and Canada: Lessons nomic Policy, April 1990, 10, 119–86.
for Monetary Union in Europe," European Eco- Eichengreen, Barry, "Is Europe an Optimum
nomic Review, February 1995, 39, 253–74. Currency Area?," Cambridge, Mass: National Bu-
Bean, Charles R., "Economic and Monetary reau of Economic Research Working Paper 3579,
Union in Europe," Journal of Economic Perspectives, January 1991.
Fall 1992, 6, 31–52. Eichengreen, Barry, "Labor markets and Eu-
Begg, David, and Charles Wyplosz, "The Eu- ropean Monetary Unification." In Masson, Paul
ropean Monetary System: Recent Intellectual R. and Mark P. Taylor, eds., Policy Issues on the
History," in: The Monetary Future of Europe. Lon- Operation of Currency Unions. Cambridge: Cam-
don: CEPR, 1993. bridge University Press, 1993, 130–62.
Begg, David, Jean-Pierre Danthine, Francesco Eichengreen, Barry, and Richard Portes, "The
Giavazzi, and Charles Wyplosz, "The East, the Anatomy of Financial Crises," In Portes, Richard
West, and the Deutschmark," Monitoring Euro- and Alexander K. Swoboda, eds., Threats to Inter-
pean Integration, 1990, 1. national Financial Stability. Cambridge: Cam-
Begg, David, Francesco Giavazzi, Luigi Spav- bridge University Press, 1987, 10–58.
enta, and Charles Wyplosz, "European Monetary Eichengreen, Barry, and Charles Wyplosz,
Union—The Macro Issues," Monitoring European "The Unstable EMS," Brookings Papers on Eco-
Integration, 1991, 2, 3–68. nomic Activity, 1993, 1, 51–144.
Begg, David, Francesco Giavazzi, Jurgen von Folkerts-Landau, David, and Peter M. Garber,
Hagen, and Charles Wyplosz, "EMU, Getting the "The ECB: A Bank or a Monetary Policy Rule?"
End-Game Right," Monitoring European Integra- In Canzoneri, Matthew B., Vittorio Grilli, and
tion, 1997, 7, 1–75. Paul R. Masson, eds., Establishing a Central Bank:
Blanchard, Olivier J. and Lawrence F. Katz, Issues in Europe and Lessons from the U.S. Cam-
"Regional Evolutions," Brookings Papers on Eco- bridge: Cambridge University Press, 1992, 86–
nomic Activity, 1992, 1, 1–75. 123.
Buiter, Willem, Giancarlo Corsetti, and Nour- Frankel, Jeffrey A., and Andrew Rose, "The
iel Roubini, "Excessive Deficits: Sense and Non- Endogeneity of Optimum Currency Areas," Lon-
sense in the Treaty of Maastricht," Economic Pol- don: Centre for Economic Policy Research Dis-
icy, April 1993, 16, 57–100. cussion Paper No. 1473, September 1996.
Burda, Michael and Charles Wyplosz, Macro- Goodhart, Charles A. E. "The European Sys-
economics, A European Text, 2nd ed., Oxford, UK tem of Central Banks after Maastricht." In Mas-
Oxford University Press, 1993. son, Paul R. and Mark P. Taylor, eds., Policy Issues
Cohen, Daniel and Charles Wyplosz, "The Eu- on the Operation of Currency Unions. Cambridge:
ropean Monetary System: An Agnostic Evalua- Cambridge University Press, 1993, 215–39.
tion." In Bryant, Ralph, David Currie, Jacob Grilli, Vittorio, Donato Masciandaro, and
Frenkel, Paul R. Masson, and Richard Portes, Guido Tabellini, "Political and Monetary Insti-
eds., Macroeconomic Policies in an Interdependent tutions and Public Financial Policies in the In-
World. Washington: International Monetary dustrial Countries," Economic Policy, October
Fund, 1989, 311–37. 1991, 13, 341–92.
Decressin, Jorg, and Antonio Fatas, "Regional Hartmann, Philipp, "The Future of the Euro
Labor Market Dynamics in Europe and Implica- as an International Currency, A Transactions
tions for EMU," European Economic Review, De- Perspective,", unpublished paper, 1996, Lon-
cember 1995, 39, 1627–55. don: London School of Economics.
de Grauwe, Paul and Wim Vanhaverbeke, "Is Kenen, Peter B., "The Theory of Optimum
Europe an Optimum Currency Area?: Evidence Currency Areas." In Mundell, Robert A., and Al-
from Regional Data." In Masson, Paul R. and exander A. Swoboda, eds., Monetary Problems ofthe
 Charles Wyplosz 21

International Economy. Chicago: Chicago Univer- Sachs, Jeffrey D., and Xavier Sala-i-Martin,
sity Press, 1969, 41–60. "Fiscal Federalism and Optimum Currency
Kenen, Peter B., ''Sorting Out Some EMU Is- Areas: Evidence for Europe from the United
sues," Reprints in International Finance, No. 29, De- States." In Canzoneri, Matthew B., Vittorio Grilli,
cember 1996, International Finance Section, and Paul R. Masson, eds., Establishing a central
Princeton University. bank: Issues in Europe and lessonsfrom the U.S..Cam-
Krugman, Paul, "Lessons of Massachusetts for bridge: Cambridge University Press, 1992, 195-
EMU," In Torres, Francisco and Francesco Gia- 219.
vazzi, eds., The Transition to Economic and Monetary Sargent, Thomas J. and Neil Wallace, "Some
Union. Cambridge: Cambridge University Press, Unpleasant Monetary Arithmetic," Federal Reserve
1993, 241–69. Bank of Minneapolis Quarterly Review, 1981, 5, 1–
McKinnon, Ronald, "Optimum Currency Areas," 17.
American Economic Review, 1963, 53, 717–725. Swedish Government Commission on the
Masson, Paul R. and Mark P. Taylor, "Cur- EMU, EMU—A Swedish Perspective, Amsterdam:
rency Unions: A Survey of the Issues.'' In Masson, Kluwer, 1997.
Paul R, and Mark P. Taylor, eds., Policy Issues on Thygesen, Niels, "Economic and Monetary
the Operation of Currency Unions. Cambridge: Cam- Union: Critical Notes on the Maastricht Treaty
bridge University Press, 1993, 3–54, Revisions," In Torres, Francisco and Francesco
Masson, Paul R. and Bart Turtelboom, "Char- Giavazzi, eds., The Transition to Economic and Mon-
acteristics of the Euro, the Demand for Re- etary Union. Cambridge: Cambridge University
serves, and Policy Coordination Under EMU," Press, 1993, 9–45.
unpublished paper, 1997, Washington, DC: In- von Hagen, Jürgen, "Monetary Union and Fis-
ternational Monetary Fund. cal Union: a Perspective from Fiscal Federalism."
Mundell, Robert A., "A Theory of Optimum In Masson, Paul R. and Mark P. Taylor, eds., Pol-
Currency Area," American Economic Review, Sep-
icy Issues on the Operation of Currency Unions. Cam-
tember 1961, 50, 657–665.
bridge: Cambridge University Press, 1993, 264–
Mundell, Robert A, "EMU and the Interna-
96.
tional Monetary System," In The Monetary Future
von Hagen, Jürgen and Barry Eichengreen,
of Europe. London: Centre for Economic Policy
"Federalism, Fiscal Restraints, and European
Research, 1993.
Monetary Union," American Economic Review, Pa-
Obstfeld, Maurice, "A Strategy for Launching
pers and Proceedings, May 1996, 86, 134–38.
the Euro," unpublished paper, March 1997, Uni-
von Hagen, Jürgen and Ian J. Harden, "Na-
versity of California, Berkeley.
Padoa Schioppa, Tommaso, "Squaring the tional Budget Processes and Fiscal Perfor-
Circle, or the Conundrum of International Mon- mance," European Economy, 1994, No. 3, 311–418.
etary Reform," Catalyst, a Journal of Policy Debate, von Hagen, Jürgen, and Manfred J. M. Neu-
Spring 1985, 1:1. mann, '' Real Exchange Rates within and between
Pisani-Ferry, Jean, Alexander Italianer, and Currency Areas: How Far Away Is EMU?," Review
Roland Lescure, "Stabilization Properties of of Economics and Statistics, May 1994, 76, 236–44.
Budgetary Systems: A Simulation Analysis," Eu- Weber, Axel A., "EMU and Asymmetries and
ropean Economy, 1995, 5, 511–38. Adjustment Problems in the EMS: Some Em-
Ricci, Luca, "A Model of an Optimum Cur- pirical Evidence," London: Centre for Eco-
rency Area," Working Paper 97/76, Washington, nomic Policy Research Discussion Paper: No.
DC: International Monetary Fund, 1997. 448, August 1990.
22 Journal of Economic Perspectives