MONETARY POLICY REPORT | WINTER 2021 (box 2)
(authors: Robert Ambriško, Dana Hájková, Petr Král, Stanislav Tvrz)
The fiscal impulse is an integral part of the Monetary Department’s analytical and forecasting work when preparing the CNB’s macroeconomic forecast. As shown in section II (The real economy and the labour market), fiscal policy can play an important role in shaping economic activity and inflation. Expenditure on government fiscal measures is extremely high in some periods. This was the case both last year and this year. We use the estimated fiscal impulse to quantify the effect of fiscal policy in the given year on economic activity over the business cycle.
The fiscal impulse can be estimated using two methods, described in more detail below. In the CNB forecast, we work primarily with a “bottom-up” method. The overall consistency of the fiscal forecast is then tested using a “top-down” method.
The primary method for calculating the fiscal impulse is the bottom-up approach, which is derived from fiscal discretion (see the definition on the right). Fiscal discretion is expressed in per cent of GDP and represents the sum of the budgetary impacts of discretionary fiscal measures on the revenue and expenditure sides of public budgets (see Table 1). Examples of discretionary fiscal measures include changes in taxes, new social benefits, and anti-crisis programmes to support employers (Antivirus) and the self-employed (one-off benefits).
Table 1 – Discretionary fiscal measures entering the bottom-up calculation of the fiscal impulse
% of GDP; impact on change in general government balance
| 2018 | 2019 | 2020 | 2021 | 2022 | |
|---|---|---|---|---|---|
| TOTAL REVENUE MEASURES | 0,1 | -0,1 | -0,6 | -1,5 | -0,3 |
| Increase in excise duties | 0,0 | 0,0 | 0,1 | 0,0 | 0,0 |
| Changes in VAT | 0,0 | 0,0 | -0,1 | -0,1 | 0,0 |
| VAT control statements | 0,1 | ||||
| Electronic sales registration | 0,1 | ||||
| Abolition of property transfer tax | -0,2 | ||||
| Abolition of social contributions for self-employed | -0,3 | 0,2 | |||
| Abolition of soc.contributions for employers (Antivirus C) | -0,2 | 0,2 | |||
| Abolition of super-gross wage | -1,4 | ||||
| Increase in the tax deductible bonus for individuals | -0,2 | -0,2 | |||
| Changes in depreciation | -0,2 | -0,1 | |||
| Other changes in taxes and soc.contributions | 0,0 | -0,1 | 0,1 | -0,1 | -0,1 |
| TOTAL EXPENDITURE MEASURES | -1,4 | -0,8 | -2,6 | 1,2 | 0,8 |
| Extra valorisation of pensions | 0,0 | -0,3 | -0,3 | 0,1 | 0,1 |
| Changes in social benefits | -0,2 | 0,0 | -0,2 | 0,0 | 0,0 |
| Compensation of government employees | -0,4 | -0,3 | -0,3 | 0,3 | 0,2 |
| Government investment, domestic | -0,4 | -0,2 | 0,1 | 0,0 | 0,0 |
| Government investment, EU funded | -0,3 | 0,0 | -0,1 | -0,1 | -0,2 |
| Increase in attendance allowance | -0,2 | 0,2 | 0,0 | ||
| Employment support programme (Antivirus A+B) | -0,5 | 0,4 | 0,1 | ||
| One-off benefits | -0,5 | 0,4 | 0,0 | ||
| Loss carryback | -0,3 | 0,0 | 0,3 | ||
| COVID subsidies to affected industries | -0,3 | 0,0 | 0,3 | ||
| Other changes in expenditure | 0,0 | 0,0 | 0,0 | 0,0 | 0,0 |
| TOTAL MEASURES | -1,3 | -0,9 | -3,2 | -0,3 | 0,5 |
Note: A minus sign indicates lower revenue/higher expenditure and a positive sign higher revenue/lower expenditure. Non-wage consumption expenditure does not appear on the expenditure side, as it is taken into account in the model forecast through the direct channel of government consumption.
To assess the effect of discretionary fiscal measures on GDP and its outlook correctly, we first need to decide whether they are permanent or temporary. Permanent measures have a permanent effect on public budgets but only affect year-on-year growth in economic activity in their first year. Their effect on economic growth in the following years is neutral. If permanent measures are adopted in the course of the year and not at its start, they also have a limited impact on economic activity in the following year or part thereof. Following the initial shift in aggregate demand, to which the price level responds in the initial phase, permanent measures do not affect subsequent inflation either. The impacts on inflation are thus only temporary.
By contrast, temporary fiscal policy measures, which affect the revenue and expenditure sides of public budgets only for a time, affect economic growth both when introduced and subsequently when discontinued in the opposite direction. The impact of temporary measures on inflation when they are introduced is thus similar as in the previous case. However, one should bear in mind that the effect they have when they end is opposite to their initial effect.
The effect of discretionary fiscal policy measures is incorporated into the CNB forecast through two channels. The first is the path of government consumption itself, which – as a direct component of domestic demand and GDP – is contained in the fiscal block of the g3+ forecasting model. In addition, the contributions of other discretionary fiscal measures are taken into account expertly through an estimate of the size and structure of the fiscal impulse using the bottom-up method. The contributions of individual fiscal measures to real GDP growth i.e. the fiscal impulse, are subsequently calculated as the product of the fiscal discretion identified (entering with the opposite sign by convention) and the fiscal multiplier. In the past, we worked with the single aggregate fiscal multiplier of 0.6. In the current forecast, the fiscal multiplier for changes in labour taxation has been lowered.
For illustration, we calculate the impacts of selected discretionary fiscal measures on real GDP growth. The abolition of property transfer tax last year permanently reduced government revenue by around 0.2% of GDP (see Table 1). After multiplication by the fiscal multiplier (0.6), this implies a positive contribution to real GDP growth of approximately 0.1 pp in 2020. Last year, the payment of one-off benefits amounting to 0.5% of GDP contributed around 0.3 pp to real GDP growth. This support is temporary, so the contribution of one-off benefits to GDP growth this year will be opposite (negative).
For forecasting purposes, the total fiscal impulse calculated using the bottom-up method is divided into the individual components of domestic demand: private consumption, private investment and government investment, which we subdivide into a domestic component and a component stemming from the drawdown of EU funds (see Table 2).
Table 2 – The contribution of discretionary fiscal measures to GDP growth is divided into the individual components of domestic demand
Contributions to GDP growth in pp
| 2018 | 2019 | 2020 | 2021 | 2022 | |
|---|---|---|---|---|---|
| FISCAL IMPULSE | 0,8 | 0,5 | 1,9 | -0,3 | -0,3 |
| of which impact through: | |||||
| private consumption | 0,3 | 0,4 | 1,7 | -0,3 | -0,3 |
| private investment | 0,0 | 0,0 | 0,2 | -0,1 | -0,1 |
| govt. investment, domestic | 0,3 | 0,1 | -0,1 | 0,0 | 0,0 |
| govt. investment, EU funded | 0,2 | 0,0 | 0,1 | 0,1 | 0,1 |
The top-down approach to calculating the fiscal impulse can be used to test the overall consistency of the forecast and to verify the primary bottom-up method ex post. This approach captures the effect of fiscal policy on economic activity through the fiscal stance[1] i.e. the year-on-year change in the ratio of the general government structural balance to GDP in percentage points (see Table 3). A positive fiscal stance means fiscal restriction and a negative one fiscal expansion. By multiplying the fiscal stance (which enters the calculation with the opposite sign by convention) by the aggregate fiscal multiplier (0.6), we get an alternative estimate of the effect of fiscal discretion on real GDP.
Table 3 – The fiscal forecast containing the estimate of the structural balance and the fiscal stance entering the top-down calculation of the fiscal impulse
% of nominal GDP unless otherwise indicated
| 2018 | 2019 | 2020 | 2021 | 2022 | |
|---|---|---|---|---|---|
| Government revenue | 41,5 | 41,6 | 41,4 | 40,3 | 40,0 |
| Government expenditure | 40,6 | 41,3 | 47,5 | 46,8 | 45,6 |
| of which: interest payments | 0,7 | 0,7 | 0,7 | 0,7 | 0,7 |
| GOVERNMENT BUDGET BALANCE | 0,9 | 0,3 | -6,1 | -6,5 | -5,6 |
| of which: | |||||
| primary balancea) | 1,7 | 1,0 | -5,4 | -5,8 | -4,9 |
| one-off measuresb) | 0,0 | 0,2 | 0,2 | 0,2 | 0,1 |
| ADJUSTED BUDGET BALANCEc) | 0,9 | 0,0 | -6,3 | -6,7 | -5,7 |
| Cyclical component (disaggreg. method)d) | 1,5 | 1,5 | -0,9 | -0,9 | -0,5 |
| Structural balance (disaggreg. method)d) | -0,6 | -1,5 | -5,4 | -5,8 | -5,2 |
| Fiscal stance in pp (disaggreg. method)e) | -1,1 | -0,9 | -3,9 | -0,4 | 0,6 |
| Cyclical component (aggregated method)d) | 0,7 | 0,6 | -0,6 | -0,6 | -0,1 |
| Structural balance (aggregated method)d) | 0,2 | -0,6 | -5,7 | -6,0 | -5,6 |
| Fiscal stance in pp (aggregated method)e) | -0,4 | -0,8 | -5,1 | -0,3 | 0,5 |
| GOVERNMENT DEBT | 32,1 | 30,2 | 38,4 | 43,3 | 46,7 |
Note:
a) government budget balance minus interest payments
b) This item consists of expected revenue from primary sales of emission permits, expenditure on the (New) Green Savings Programme, guarantees and revenue from the sale of frequency bands to mobile operators.
c) adjusted for one-off measures; CNB estimate
d) CNB estimate; the disaggregated method is based on the evolution of the individual tax bases in the business cycle; the aggregated method defines the position of the cycle on the basis of the output gap only.
e) y-o-y change in structural balance.
The above two methods can be compared using the example of the situation in 2020. The top-down estimate of the fiscal impulse for this year is 2.3–3.1 pp to real GDP growth (depending on the method used to cyclically adjust the government balance). In the assessment of the overall effect of fiscal policy on economic activity, this is qualitatively in line with the bottom-up estimate. The slightly higher quantification using the top-down method points to probable additional flexible government expenditure going beyond specific changes to the legislation.
The government budget balance and government debt are often taken into account when assessing fiscal policy. We should therefore mention their link with the fiscal impulse. The fiscal impulse is reflected in the government deficit and debt through individual discretionary fiscal measures. Expansionary discretionary measures increase the government deficit and debt over their entire duration. As explained above, however, they contribute positively to GDP growth only temporarily and only when introduced (the fiscal impulse they generate is positive only in their first year). By contrast, restrictive discretionary measures reduce the government deficit and debt over their entire duration but contribute negatively to GDP growth only when introduced (the fiscal impulse they generate is negative only in their first year).
Comparing the total fiscal impulse (the contribution of fiscal policy to GDP growth) in individual years with the path of the government deficit and debt expressed as a share of nominal GDP is therefore not in itself very informative. There are other methodological differences between these variables besides the above ones stemming from different definitions[2] and the general government deficit and debt are also affected by a whole range of other factors besides fiscal discretion.[3]
The fiscal impulse is a variable which reflects the effect of fiscal policy in the given year on economic activity over the business cycle. It is calculated using the volume of fiscal discretion and the fiscal multiplier.
Fiscal discretion refers to year-on-year changes in government revenue and expenditure (made by the government and/or due to legislative amendments) that are derived from specific fiscal measures and affect government finances beyond the impacts of the business cycle (i.e. beyond automatic fiscal stabilisers).
The fiscal multiplier measures the degree to which fiscal discretion affects real GDP.
The bottom-up fiscal impulse method is based on the sum of the budgetary impacts of the government’s individual discretionary fiscal measures.
The top-down fiscal impulse method uses aggregated data – the structural government budget balance, i.e. the budget balance adjusted for the effects of the business cycle and extraordinary one-off measures.
[1] The fiscal stance can be calculated using two methods: an aggregated one (using an estimate of the output gap) and a disaggregated one (based on the evolution of the individual tax bases over the business cycle).
[2] Change in drawdown of EU funds is fully reflected in the fiscal impulse. However, it affects public budgets and government debt only up to the amount of the government’s co-financing of European projects.
[3] The phase of the business cycle, inflation, nominal GDP growth, formation of government financial reserves, and interest paid on government debt.