D ∞ t+i-1
(3) pγt =∑≈=0(∏ t ρk ) st+i
where st = ^PY + 1^^+~~pγ~^ is the budget surplus inclusive of seigniorage as a ratio to
GDP.
The solvency condition can be satisfied in many ways. First, budget discipline may ensure
that future budget surpluses St+i are such that they match the actual public sector debt. Second,
fiscal discipline can be weak and recourse to seigniorage is needed to deliver the needed
sequence of {st+i}. This is the channel through which, historically, fiscal indiscipline has
repeatedly delivered inflationary episodes. This is why the Maastricht Treaty explicitly rules
out any financing of budgets through seigniorage, both on a routine basis and in an
emergency situation, the latter case being dealt with through the no-bailout clause. This is
also why the independence of the ECB is guaranteed by the treaty.
The third case is the relevant one. If the sequence of {st+i} violates the solvency condition, it
is the price level Pt on the left-hand side of (3) that becomes the variable of adjustment. This
is the case of fiscal dominance where the budgetary authorities can impose their will and
carry out unsustainable budget deficits. Monetary dominance is the opposite case, when
neither seigniorage nor the price level are made to be endogenous in (3) so that the variable of
adjustment is the disciplined sequence of {st+i}. The task of the SGP can be seen as imposing
fiscal dominance so that control of the price level is not lost and without having to call upon
seigniorage as mandated by the treaty.
3. What Do National Governments Do with their Fiscal Policies?
3.1. Policy effects
Unsurprisingly, the question of the usefulness of fiscal policy as a macroeconomic tool is
highly controversial. At the theoretical level, the debate pits (neo)Classical against
(neo)Keynesian macroeconomists. The former asserts that, one way or another, consumers
and firms view public debts as their own liability; accordingly, they reduce their expenditures
whenever the debt increases or diminishes less than previously expected. The latter relies on
price stickiness, borrowing constraints and/or other market imperfections to find that fiscal
policy can affect output.4 In view of such conflicting theoretical results, the verdict should
come from empirical studies.
Empirically, too, the issue is controversial. Some authors find that fiscal policy affects output,
even though the multipliers are small and have possibly declined in recent years (Blanchard
and Perotti, 2002; Perotti, 2004, 2007; Favero and Giavazzi, 2007; Romer and Romer, 2007).
Others find that consumption moves in an offsetting direction, although the offset effect is
partial, which leaves a small output effect (Ramey, 2006).
Why do different authors reach different results? All the above papers use VAR estimates to
pinpoint the relationship between output and fiscal policy. All of them also use the cyclically-
adjusted balance as a measure of the fiscal policy stance.5 They differ in the way they identify
the VARs. Those who find a positive effect of fiscal shocks on consumption, and therefore
output, typically use quarterly variables and make the assumption that there is no
contemporaneous effect from cyclical conditions to policy discretionary actions.
4 For a brief review of the arguments, see Ramey (2006) and Perotti (2004).
5 A different literature looks at the automatic stabilizers. Buti et al. (2003) claim that a heavy tax
burden may well reduce the effectiveness of the stabilizers.