government's budget, it is generally much larger than the contribution of seigniorage to the
budget d, hence b > d.8 Thus, even in this case that the central bank simply executes the
wishes of the government and adopts its (or their) preferences, this result will hold. Only for
unreasonable values for the inflation aversion of the government this condition will not be
fulfilled.
The intuition for this result is that single governments do no longer fully internalize
the inflationary impact a distortive individual fiscal policy has on the common rate of
inflation. Before monetary union governments had an incentive to pursue a disciplined
fiscal policy to avoid an increase in the rate of inflation, and to increase the output gap. At
the same time, it forced the government to implement structural reforms to make up for
lower fiscal revenue and inflation. The trade off between inflation, fiscal revenue and
reforms changes with monetary union since the government needs no longer fear that an
expansive fiscal policy has a strong inflationary impact. The disciplinary influence on the
policy choices from the government's inflation aversion is reduced. This effect is stronger
the more members the monetary union has.
The discussion of the symmetric case has shown that a monetary among such
countries would have a strong impact on their fiscal policy and their reform efforts.
Cooperation among them would basically restore the national case. Viewed from this
perspective, rules for cooperation are an attempt to internalize negative spillovers that arise
through the introduction of a common central bank and a justification for fiscal policy rules
in the monetary union.
8 It is clear that b>1 because the effects from fiscal policy on output are normalized to
unity (eq. 1). If fiscal policy is distortive, b<1 would imply that the revenue from taxes is
lower than the negative effects it has. This is not very reasonable, not even for highly
distortive fiscal systems. The contribution of seigniorage to the overall budget instead is
certainly much smaller than unity. Masson and Patillo (2001b) estimate values between 0.6
percent of GDP (for WAEMU) and 3.9 percent (for Liberia).
12