How should the European Central Bank (ECB) operate and how should its
monetary policy be coordinated with outside countries after the birth of
the euro? Both questions remain open. At the Verona meeting, it was
agreed that a new EMS will link some of the outside countries to the
euro. But participation in this new EMS is optional, and some likely
outsiders have already declared that they do not intend to join.
Moreover, the details of how such a system would work remain to be
decided. Some of the most controversial aspects are whether the burden
of defending the exchange rate will also fall on the ECB, the width of
the exchange rate band, and the procedure for realignments.
Consensus on these difficult questions will not come easily because
there is an unavoidable contradiction between the inflation objective
given to the ECB by the Maastricht Treaty and stabilizing exchange
rates. Fixing any exchange rate is a daunting task unless there is
unlimited bilateral support by both the ECB and outsiders’ national
central banks. But unlimited support by the ECB would undermine the
credibility of its anti-inflationary policy and is ruled out by the
Treaty.
For this reason, we argue that a system of mandatory inflation
targets is a better answer than any other mechanism. It is a symmetric
arrangement and one that facilitates holding the ECB accountable for its
actions. Moreover, a regime with strict and symmetric inflation targets
helps solve the problems of monetary cohabitation in three distinct
ways:
- It removes the incentive to engage systematically and deliberately
in competitive devaluations.
- A target regime restores domestic credibility to a low inflation
policy. This in turn makes monetary cohabitation easier because it
reduces the volatility of speculative shocks to the exchange rate.
- Inflation targeting forces monetary policy to respond to shocks in
a way that stabilizes the exchange rate.
But how could a European system of inflation targets be implemented
in practice? All EU members would have to participate. They would
jointly have to announce precise quantitative targets for a well-defined
measure of inflation, though not necessarily the same targets.
Their inflation performance would be monitored by a European
institution, most naturally by the European System of Central Banks
(that is, the ECB plus the national central banks not participating in
the single currency). The penalties for missing the target would also be
recommended by a European institution. For this, the most natural
candidate is the Council of Ministers, since this is the only body that
comes close to being a European principal for the ECB.
In this respect, the arrangement would be similar to the ‘excessive
deficit’ procedure described in the Maastricht Treaty. Unlike the
excessive deficit procedure, however, the penalties for missing the
announced target would have to fall on the central banks, and not on the
country involved. A mild penalty would be public blame; a harsher
penalty would be a recommendation to fire the governor, though in this
case, national legislation may be required to make the sanction legal.
To what extent is a system of generalized inflation targets
consistent with the Maastricht Treaty? In some ways, the Treaty provides
a useful underpinning. Except for the opt-out countries, central banks
are required by the Treaty to be independent and to pursue price
stability as a goal. Moreover, inflation is one of the convergence
criteria, and this imposes an implicit penalty for the countries that
would like to join the euro but have excessively high inflation compared
to the other EU countries.
The idea of imposing accountability for inflation differences is thus
already an important ingredient in the Treaty. The system we advocate
can therefore be thought of as strengthening some provisions in the
Treaty. It makes inflation convergence an overriding objective of
monetary policy, and it forces each central bank to give that objective
operational and quantitative contents and to announce them clearly in
advance. It also gives ‘institutional’ content to the exercise of
accountability by making inflation targeting a requirement for all
central banks, including the ECB and those opting out of the single
currency, by spelling out more clearly who is responsible for hitting
the target and by prescribing penalties for poor performance.
The Maastricht Treaty also contains two references to the exchange
rate. It says that the exchange rate is a matter of ‘common
interest’, and exchange rate stability (in the sense of no
realignments) is one of the convergence criteria. But under the current
EMS, with fluctuation bands of plus or minus 15%, the exchange rate
criterion is almost meaningless. We think it should remain meaningless:
any attempt at implementing this part of the Treaty by giving the
exchange rate criterion a stricter interpretation should be avoided as
much as is legally possible.
Naturally, even with a system of generalized inflation targets, the
exchange rate would remain a matter of ‘common interest’. Indeed, a
major reason for institutionalizing inflation targets at the EU level
rather than at the national level is precisely because of the desirable
repercussions for the exchange rate. But exchange rate stability ought
to be the result of successful monetary policies not the explicit target
for these policies.
Torsten Persson and Guido Tabellini Persson is Professor of
Economics at the Institute for International Economic Studies in
Stockholm; Tabellini is Professor of Economics at the Università
Bocconi in Milan and Director of IGIER in Milan. Both are Research
Fellows in CEPR’s International Macroeconomics programme.