Product Tax Modelling - using the dynamic interindustry model INFORGE

For many years, empirical research in economics concentrated predominantly on the effectiveness of monetary policy and less on fiscal policy. But developments like the Balance Budget Amendment in the USA, the single currency union in Europe or the world economic crisis in 2009 has raised the question on fiscal policy and its potential to initiate changes in economic activity. Especially in Germany, the application of simulation models to measure fiscal policy does not have a long tradition. A first review of the existing literature reveals that the qualitative effects correspond widely, whereas the quantitative effects alter considerable due to diverging modelling frameworks and different datasets. In this paper a dynamic input-output model which follows an evolutionary modelling approach is used to analyse fiscal policy instruments. The paper concentrates on the modelling of taxes on products which are a major contributor to state income and thus an important tool for conducting fiscal policy. This paper therefore wants to contribute to the growing literature on fiscal policy analysis by introducing and applying an alternative modelling approach. Two objectives are pursued in this paper. Firstly, the implementation of taxes on products in the modelling framework of the used model is described. This implies the identification of the location of product taxes as well as the setting of the database and the computing of regression equations for product taxes. Product taxes are further differentiated into value-added taxes, import taxes and other product taxes such as excise duties. The different tax types are considered in the choice of regression approach. The advantages of this detailed modelling approach to product taxes are an explicit differentiation of product taxes by tax types, by goods and by components of demand which enables a closer imitation of economic behaviour and allows the application of fiscal policy instruments in a more detailed way. Secondly, the application of the new modelling approach for product taxes. A simulation on an alternative value-added tax rate system is computed: the simulation follows the suggestion of the latest report of the German Council of Economic Experts that proposes the abolishment of the reduced tax rate and suggests the introduction of a normal tax rate by 16.5%. Most of the empirical studies applied for fiscal policy analysis use autoregressive simulation (VAR-) models. To a smaller scale, micro-simulation models (MSM) are also applied for the investigation of fiscal policy. Other empirical studies on the impact of fiscal policies are based on the application of computable general equilibrium models (CGE). In this paper an alternative modelling approach to analyse fiscal policy shocks has been chosen. A dynamic inter-industry model has been applied which uses regression analysis to describe economic behaviour of different economic agents (consumers, producers, exporters, importers, investors). Interindustrial relations are explicitly used and change over time. Accounting consistency is assured in all time; on the production side as well as on the demand side. The bottom-up approach is characterized by a high desaggregation on sectoral level, enabling a detailed modelling by industries and goods. The integrated structure of the model allows a complex and simultaneous solution due to the absolute accounting consistency. Input-output tables are fully implemented in the national accounts allowing linkages between inter-industrial interdependencies, distribution of income, the redistribution effects of the state and the usage of income for goods. Production is determined by the demand side of the economy via the Leontief-equation, all determinants of demand depend on relative prices which are again a function of firm’s unit costs and import prices. The inter-industry forecasting model for Germany – INFORGE – has been developed by the Institute of Economic Structures Research (GWS) and can be applied for economic forecasts as well as for simulation analysis. In this paper, its simulation qualities on product taxes are tested and demonstrated in one simulation example. The quantitative results of the simulation on the economy are compared with a status-quo scenario. The results are yet to be computed with the updated version of INFORGE 2010 but are to be expected to be finalized by March 2011. Overall, it is expected that the simulation (abolition of the reduced value added tax rate and introduction of a uniform tax rate of 16.5% on all products) leads to a decline in GDP compared to the baseline. In the short-run, a variation in value added tax rates affects domestic prices. The introduction of a uniform VAT rate raises the price levels of all price indices, whereas the deviation to the baseline is expected to be the highest for consumption prices. The results probably show that the impact of the abolition of a reduced VAT rate (in Germany from 7% to 16.5%) is higher than the reduction of the normal VAT rate (in Germany from 19% to 16.5%). The background is that the reduced value added tax rate is levied on everyday products such as food or beverages. An increase of the reduced value added tax rate is likely to have a greater impact on prices and total demand than the reduction of the normal value added tax rate. It is expected that one major consequence of an increase in the price level is an immediate decline in all components of total demand which leads to lower output compared to the baseline. At the same time, it is expected that a uniform VAT rate increases the volume of product tax revenues. Under the assumption, that most of the state income is used for consolidation purposes, net lending and net borrowing are likely to turn positive. With a time lag of approximately one year, the effect of uniform VAT rates will reach the labour market. Anticipated price increases react on wages and salaries which increase faster than in the baseline. Higher unit labour costs effect the price-setting behaviour of the firms and their demand for labour. Prices are further increased and the number of unemployed people rises. A higher unemployment rate puts pressure on the social security system which results in an increase in monetary benefits. The cumulating effects slow down after a further year of adjustment. National output drops back to the growth path of the baseline.