The macroeconomic effects of a carbon tax in the Netherlands Íde Kearney, th September 08. This note reports estimates of the economic impact of introducing a carbon tax of 50 per ton of CO in the Netherlands. In Section I we use the global macroeconomic model NiGEM to estimate the macroeconomic impact of the higher export prices in NL and its competitor countries consequent on such a tax. We estimate the impact on NL GDP under three different assumptions about how widely the tax is implemented, namely in NL only, in Northern Europe only and EU-wide. In Section II we use the DNB macroeconomic model DELFI to simulate the impact of such a tax on the economy under different assumptions about the uses made of the revenue generated. I. ENERGY TAX INTRODUCED IN NL ONLY, IN NORTHERN EUROPE OR EU-WIDE This section reports the results of three separate scenarios run in the global model NiGEM. These estimate the impact on GDP in the Netherlands (NL) of higher export prices following the imposition of a 50 per ton of CO tax in ) the Netherlands alone, ) Northwestern European countries (excluding UK and Ireland), and ) all EU countries respectively. Figure I. shows the estimated percentage change in export prices in the Netherlands and all relevant competitor countries from the IO model in all three cases. These are summarized in Table I.: an NL-only CO tax leads to a.% increase in NL export prices, while an EU-wide CO tax leads to a.5% increase in NL export prices. The estimated change in export prices (Figure I.) is introduced in NiGEM as an exogenous shock to export prices for the Netherlands and all of its competitor countries in three separate scenarios. Figure I. shows the impact on NL GDP for the first five years (thereafter the effects start to stabilize). As the tax is more widely introduced, the increase in export prices both in NL and abroad becomes stronger which means that world trade suffers more. For the NL-only scenario, the impact after five years is to reduce NL GDP by 0.8% while for the EU-wide scenario the impact is to reduce NL GDP by 0.5%. Table I.: IO model results for 50/ ton CO: effect on NL prices, percentage change from base Price exports Relative export price Netherlands: All sectors.. North-west Europe: All sectors. 0.9 EU: All sectors.5 0.7 NIESR, 08, National Institute Global Econometric Model https://nimodel.niesr.ac.uk/. EXIOBASE data. Price exports minus weighted sum of competitor export prices.
Australia Belgium Bulgaria Brazil China Canada Czech Republic Denmark Estonia Finland France Germany Greece Hungary Indonesia India Ireland Italy Japan Lithuania Latvia Mexico Netherlands Norway Austria Poland Portugal Romania Russian Federation South Africa Sweden South Korea Slovenia Spain Slovak Republic Switzerland Turkey Taiwan United Kingdom United States RoW Europe Cyprus Croatia Luxembourg Malta RoW Middle East RoW Africa RoW America RoW Asia and Pacific Figure I.: the % change in country export prices from the IO model Export price % change ( 50 tax CO ).5.5.5.5 0.5 0 EU NW Europe NL Figure I.: GDP effects 0.00 Impact on GDP in the Netherlands % difference from base -0.0-0.0-0.0-0.0-0.50-0.60 08 09 00 0 0 NL only North West Europe All EU countries These results are run in the default forward mode in NiGEM. The results in backward looking mode are of a similar magnitude.
II. AN ENERGY TAX IN NL ALONE THREE DIFFERENT REVENUE SCENARIOS This section reports the results of introducing a 50 per ton tax on CO emissions in the business sector in the Netherlands. These results are estimated using DELFI 5. Three scenarios are presented. In the first the carbon tax is deficit reducing. In the second the revenue from the carbon tax is used to reduce labour taxes. In the third the revenues from the CO tax are used to reduce taxes on the corporate sector. The shock is implemented as follows:. We introduce a 6.9% increase in the price of energy inputs to the business sector. The size of this shock is based on estimates from the detailed sectoral I-O model.. We introduce a 50 carbon tax per ton of CO as an indirect tax 67. Scenario : Government collects energy taxes and uses them to reduce debt. The effect on employment and output is negative. Higher energy prices increase the cost price for firms, reduce the profit margin and increase prices. The price of exports increases reducing the demand for exports abroad. Lower profitability and lower GDP reduce business investment, this is further accentuated by the accelerator effect in the investment equation. Higher consumer prices and lower employment reduce personal disposable income. This reduces consumption, housing investment and house prices. There is a positive effect on the government balance but improvements in the public debt position do not have positive feedback effects in DELFI. The improvement in the public finances erodes relatively rapidly, where the government balance is percentage point of GDP higher in the first year, this has fallen to 0.65 pp in year. Scenario : Government uses higher energy tax revenues to lower labour taxes. In this scenario labour taxes are reduced to offset the higher energy taxes so that the carbon tax is revenue neutral on impact. Labour taxes are lowered in each year by an amount equal to the revenues generated from the CO tax in Scenario. This is equivalent to approximately 8 bn per annum. Within a year period there is a positive impact on GDP. Lower labour taxes stimulate consumption, housing investment and house prices. This positive development in domestic demand serves to offset the long-run deterioration in the Netherland s competitive position which is reflected in higher export 5 https://www.dnb.nl/en/onderzoek-/econometric-models-of-dnb/delfi-a-model-of-the-dutch-economy/index.jsp 6 To compute the fiscal impact of this carbon tax we use estimates for 0 suggesting that a 50 per ton CO tax would yield 8 billion in revenues annually. This implies a CO emission factor (emissions/gdp) of 0.0008 in 0. We assume that this emissions factor will not change over time; over an 8 year horizon this is a relatively innocuous assumption since technologies do not change rapidly in response to price changes. 7 Energy tax revenues affect indirect taxes and pass-through into prices. The DELFI price block includes an implicit indirect tax rate in the equations determining consumer prices, investment prices, government consumption deflator and export prices. This means that, with the exception of the energy sector, there is an immediate and full pass-through of a higher carbon tax on prices. In the case of prices in the energy sector, pass-through of the carbon tax operates through its effect on production costs.
prices. The government balance initially worsens by 0.6pp of GDP in year, however this deterioration gradually corrects itself as GDP and employment recover. The long-run response of wages in Scenarios and : In both scenarios the contractual wage falls in the long-run. However the underlying reason for this is very different. In scenario it is the weakness of the economy and the labour market that drives down wages, while in scenario it is the reduction in the tax wedge that drives down the pre-tax wage. Scenario : Government collects energy taxes and uses them to reduce corporate taxes. In this scenario the corporate tax rate (PIVPB) is calibrated to ensure that total corporation tax paid by firms falls by an amount equal to the higher energy taxes generated from the CO tax in Scenario. The effect on employment and output is negative. Higher energy prices lead to higher export prices. This reduces the demand for exports abroad. The very large reduction in the corporation tax rate (from 5% to 5%) leads to a very big reduction in the cost of capital (-7%). This increases the relative cost of labour and further reduces the demand for labour. Higher consumer prices and lower employment reduce personal disposable income. This reduces consumption, housing investment and house prices. The increase in net profit margins for firms and the very large fall in the cost of capital serve to partially offset these negative effects. There is a strong positive effect on investment which builds over time so that by year the negative impact on GDP begins to reverse. Furthermore, the increase in investment helps to build capacity in the economy, with a higher level of potential output. Nevertheless, the estimates suggest that over an eight year horizon the net impact on GDP and employment would be negative. Overview of the results. Scenario is effectively a technical exercise, where the revenues generated from a carbon tax are treated as deadweight. This is because DELFI does not include any reaction function where lower government debt has positive feedbacks on consumer or firm behaviour. In that sense Scenario can be viewed as a worst-case scenario benchmark. More interesting from a policy point of view is the contrast between the results of Scenarios and. Reducing labour taxes gives immediate positive results, it leads to higher disposable income and higher employment while reducing corporate taxes reduces disposable income and employment. Scenario helps to stimulate investment and potential output, but the relative burden placed on the price of labour overshadows this effect over an eight year horizon. The results suggest that a policy mix of both labour and corporate tax declines would be less distortionary. Within this the corporation tax reduction could be set up to discriminate between firms with environmentally-friendly policies, or could be directly linked to incentives for targetted R&D research.
Table II.: DELFI scenarios: 50 per ton tax on CO, 6.9% increase in energy prices Y Y Y Y Y5 Y6 Y7 Y8 Scenario : Revenue used to reduce debt GDP -0. -0. -0.70-0.86-0.9-0.86-0.75-0.67 Private consumption -0.7-0.8 -.9 -.7 -.09 -.8 -.8 -.0 Other private investment -0. -.50 -.9 -.66 -.97 -.08-0. -0.7 Exports -0. -0. -0. -0. -0.5-0. -0.5-0.9 Imports -0. -0. -0.7-0.9-0.96-0.9-0.9-0.97 HICP..0.5.50...0 0.86 Cost price (including energy) 0.90.7.5..5 0.89 0.5 0.6 Wages (contractual) private sector 0. 0.5 0.5 0. 0.07-0. -0. -0.5 Government balance (%GDP).05 0.87 0.7 0.65 0.60 0.6 0.6 0.68 Unemployment rate 0.0 0. 0.0 0.5 0.59 0.6 0.59 0.5 Scenario : Revenue used to reduce labour taxes. GDP -0. 0.5 0. 0.7 0.50 0.6 0.5 0.8 Private consumption 0.0 0.80.5.06...0.99 Other private investment -0.8-0.85 -.00-0.8-0.5-0.8-0. -0. Exports -0. -0. -0. -0.0-0.5-0.8-0.50-0.50 Imports 0.00 0.0 0. 0.8 0. 0.7 0.0 0.0 HICP...0.55.60.6.60.8 Cost price (including energy) 0.9.9.7.7.5.7 0.8 0.9 Wages (contractual) private sector 0.5 0.56 0.60 0. 0.7-0. -0.0-0.69 Government balance (%GDP) -0.08-0.6-0. -0. 0.7 0.6 0.9 0. Unemployment rate 0.0 0.08-0.05-0. -0.7-0.6-0.9-0.5 Scenario : Revenue used to reduce corporate taxes. GDP -0.0-0.8-0.8-0.7-0.6-0. -0.7-0. Private consumption -0.8-0.85 -.0 -. -.56 -.6 -.69 -.76 Other private investment 0.5 0.7.0...0.0. Exports -0.09-0.07-0. -0. -0. -0.0-0.06-0.0 Imports -0.05-0.0-0. -0. -0. -0.0 0.00-0.0 HICP.0.5.85.07.6.7..06 Cost price (including energy) -.68 -.9-0.7-0.87-0.9 -.0 -.8 -.55 Wages (contractual) private sector 0.5 0.60 0.7 0.75 0.7 0.70 0.6 0.5 Government balance (%GDP) 0.57 0.0-0.8-0.7-0. -0.5-0.8-0.90 Unemployment rate 0.0 0.8 0.8 0. 0.0 0. 0. 0. 5