In this dissertation I develop an empirical model for fiscal policy in Norway. An important feature of the methodology is to derive and model the changes in the primary budget balance that comes from changes in the economy (induced changes) and changes in fiscal policy (discretionary changes). This decomposition of the changes in the budget balance draws heavily on Braconier and Holden (1999) and (2001). The motivation for this work, besides a wish to understand the workings and nature of fiscal policy, is to endogenise fiscal policy for implementation in empirical macroeconomic models. My approach is to identify and explain the variations in a fiscal policy indicator. The traditional indicators of fiscal policy in Norway are the Non-oil Cyclically Adjusted net of Interest Payments (FIN) produced by The Norwegian Ministry of Finance, and the Cyclically Adjusted Budget Balance (CAB) produced by the OECD. In section 2 I deßne induced and discretionary changes in the budget balance and thereby derive the Braconier-Holden Indicator (BHI). Their approach is to decompose the changes in the budget balance into what they call induced and discretionary changes (where the latter is interpreted as the ßscal policy indicator). A description of induced and discretionary changes in ßscal policy can be summarized as follows:
i) Induced changes: Changes in the budget balance following from àuctuations in economic variables such as variations in tax bases, changes in unemployment and potential growth. Variation in these variables induce changes in tax revenues and public expenditures.
ii) Discretionary changes: Changes in the budget balance due to ßscal policy decisions. Discretionary changes are either caused by changes in the tax components share of its tax base, such as diÞerent tax rates or introduction of new taxes. It can also follow from decisions regarding the expenditure side of the budget such as public consumption and investments or transfers to the private sector.