How do we measure the fiscal stance? Over the past twenty years the underlying fiscal policy orientation of Belgium has tended to be looser ex post than what real-time[...]

The term fiscal stance, which refers to the degree of fiscal support provided to the economy, is widely used, yet there is no generally accepted definition. International institutions tend to consider the change in discretionary fiscal policy (e.g. measured by the change in the structural primary balance) as the fiscal stance. However, an accurate assessment of the orientation of fiscal policy should not only consider the change in support provided by the government but also the level of that support (e.g. measured by the level of the structural primary balance). For instance, in reaction to the Covid-19 pandemic, the Belgian government took exceptional policy measures amounting to over 3 % of GDP in 2020 which resulted in an unseen loosening in the fiscal stance. While crisis-related fiscal measures were withdrawn over the following years, implying a tightening of the fiscal stance, the level of support remained loose (with significant primary deficits being observed). Fiscal policy thus continued to support the economy, albeit to a lesser degree.

There are various methods to estimate the discretionary part of fiscal policy. A commonly used indicator is the structural primary balance as a percentage of GDP which is obtained using a top-down method. This method removes transitory budgetary developments from the budget balance, such as the impact of the business cycle through the effects of automatic stabilisers. However, a broader indicator of fiscal support, that does not exclude these transitory components, is also insightful. In Belgium, in particular, automatic stabilisers play an important role in cushioning economic shocks, given the large size of the Belgian government relative to GDP. We use the primary budget balance as a percentage of GDP (or the change therein) as a proxy to assess the overall fiscal stance (or the change therein). By multiplying the budgetary semi-elasticity (which captures the average response of the budget balance to fluctuations in economic activity) by the output gap (which measures the position of the economy in the business cycle), we calculate a proxy for the contribution of automatic stabilisers. In 2020, the stimulus provided by automatic stabilisers was indeed considerable: the Belgian primary deficit amounted to 7 % of GDP, signaling an exceptionally loose fiscal stance, with half of the support stemming from the cyclical component. Once the output gap started to close, the stimulus stemming from the automatic stabilisers quickly unwound, illustrating their timely and temporary nature.

It is important to note, however, that estimates of the output gap are subject to significant revisions which renders the estimates of the structural balance obtained via the top-down method unstable. In the case of Belgium, over the past twenty years, real-time estimates of the output gap consistently turned out to be too negative. As a result, the cyclical component has been systematically overestimated in real time and the structural balance underestimated. While revisions of the change in the structural balance are more limited, they add up over time, meaning that the underlying fiscal policy orientation in Belgium has tended to be assessed as looser ex post than real-time estimates suggested.

To overcome the weaknesses of the top-down method, the change in discretionary fiscal support can also be calculated directly. The bottom-up method aggregates the budgetary impact of individual revenue and expenditure measures, derived from government documents or other verifiable sources. However, in the case of Belgium, the bottom-up method appears to have underestimated structural spending (at least up to the Covid-19 pandemic) as it does not capture structural developments that occur on top of discretionary expenditure measures, such as increased ageing-related costs. Therefore, a hybrid method - which combines a bottom-up approach on the revenue side with a top-down approach on the spending side - appears more accurate for assessing the change in discretionary fiscal policy in Belgium.

As the hybrid indicator yields a more robust estimate of the underlying budgetary position than the structural primary balance, it is more useful for fiscal surveillance. Indeed, in the new European fiscal framework, the European Commission replaced the structural balance indicator with an expenditure growth benchmark (which corresponds to a hybrid indicator). Our analysis confirms, however, that, in the case of Belgium, on average the difference between the structural indicators derived from the hybrid and the top-down method is relatively small.