should also include discretionary fiscal measures that have been a normal feature of a
country’s stabilisation policy (Boije, 2004).
On closer inspection, however, the cyclically adjusted budget balance contains several
components that capture different dimensions of fiscal policy, such as a core balance
describing the underlying fiscal position; a component reflecting discretionary fiscal
policy responses to the business cycle that can move either pro- or counter-cyclically with
the output gap; and a residual component capturing all remaining shocks to the fiscal
position, reflecting transitory changes in the fiscal position due to non-stabilisation-
oriented discretionary policy and/or macroeconomic shocks.50 Disregarding these latter
aspects could provide an explanation for the sometimes quite substantial variations of
cyclically adjusted balances during the cycle.
Following an approach suggested by Jaeger (1998) and expanded by Brandner and
Diebalek (2000), we track fiscal policy behaviour over time by decomposing the observed
budget balance (as a percentage of GDP) into four unobserved components: (1) a core
balance, (2) an automatic or built-in fiscal stabiliser component, (3) a component
reflecting discretionary fiscal policy responses to the business cycle, and (4) a component
reflecting all other transitory shocks to the fiscal position.
By means of an unobserved components (UC) model, we provide an estimate of a core
balance for Austria. For this purpose we analyse the relationship between the budget
balance and the cyclical development of the Austrian economy by looking at the impact
of both automatic stabilisers and discretionary policies aimed at output stabilisation -
with particular attention to the latter.51 By doing this, we can assess whether fiscal policy
in a broader sense was pro- or counter-cyclical or reacted asymmetrically in up- or
downturns. Moreover, by looking at disaggregated data, we can answer the question
whether the pro-cyclicality / counter-cyclicality was related primarily to the expenditure
or the revenue side.
In section 4.2 we discuss some related literature before we move on to explain the
methodology chosen in section 4.3. Section 4.4 is devoted to the discussion of the main
results of our study; in section 4.5 we draw some conclusions.
4.2 Related Literature
The behaviour of fiscal policy over the business cycle has received increasing attention
from researchers in recent years. The conventional wisdom is that fiscal policy should be
counter-cyclical, stabilising economic growth around potential. In a recession, this would
call for higher deficits, while in a boom a contractionary budget would help dampen
cyclical upswings and prevent the economy from overheating. This “ideal”
50
51
Gall and Perotti (2003) conceptually split the cyclically adjusted budget balance into a “systematic”
or “endogenous” component (a component that reflects changes in structural spending or revenues in
a systematic way in response to changes in the actual or expected cyclical conditions of the
economy; corresponding to γ in section 4.3) and in a “non-systematic” or “exogenous” component
(that captures changes in the budget variables that do not correspond to systematic responses in
cyclical conditions, but are instead the consequence of exogenous political processes of
extraordinary non-economic circumstances; corresponding to what we name core balance in this
paper).
Further research will focus on the analysis of the “driving forces” of the core balance.
100