اثرات رفاه مالیات مسکن در اقتصاد تحریف شده: تجزیه و تحلیل تعادل عمومی

Efficient capital taxation has been one of the most important objectives for large tax reforms implemented in several countries during the last decades. The Norwegian Tax reform of 1992 took a large step towards tax neutrality between the different capital types and uses. However, housing capital is still an exception. In this paper the welfare effects of imposing a neutral system of housing taxation are analyzed by using an intertemporal CGE model for the Norwegian economy. The tax reform implies a substantial increase in the tax revenue from housing taxation, and the welfare effects of different rebating alternatives are considered.

Efficient capital taxation has been one of the most important objectives for large tax reforms implemented in several countries during the last decades.1 The Norwegian Tax reform of 1992 took a large step towards tax neutrality between different capital types and uses. However, housing capital is still an exception. The marginal effective tax rate on housing is substantially lower than the marginal effective tax rates on other capital types and uses. The current system of housing taxation is characterized by a low imputed value of the house for taxation purposes combined with a low imputed rate of return. Together, this generates a low marginal effective tax rate on housing capital.2 The lenient taxation of housing capital has been criticized from an efficiency point of view.3 Too much capital is allocated to housing compared with other types and uses of capital, giving an efficiency loss for the economy. Therefore, increased housing taxation may have a positive welfare effect. From a public finance perspective there may exist welfare improving tax reforms which include increased taxation of housing and where some or all of the additional tax revenue is used for reducing other distortionary taxes. The last years focus on internationalization and mobile tax bases, introduces another argument for increased taxation of housing capital: Housing capital is not very mobile between countries or other tax jurisdictions. Increasing the tax burden on housing may be an additional source of financing public spending if other tax objects more easily are moved to low tax countries, and traditional tax bases are eroded.4 As pointed out by Goulder and Thalmann (1993) it is possible to have efficient capital allocation along different dimensions. Efficiency along one dimension may imply a lower degree of efficiency along another one. Two dimensions of efficient capital allocation are respectively intratemporal and intertemporal efficient capital allocation. When effective tax rates are equal across types and end uses of capital, net marginal products of capital tend to be equal, and total output and welfare are larger than when net marginal products differ. This is static or intratemporal efficiency in capital taxation. The gain from intratemporal efficiency must be compared to efficiency along the intertemporal dimension. To obtain the ‘modified Golden rule’ allocation of consumption and total savings, the optimal marginal effective tax rate on all savings is zero, see e.g. Atkinson and Sandmo (1980) and Judd (1999).5 This is intertemporal tax neutrality for a closed economy. In a small open economy with interest rate given in the world financial market, capital income taxation will also in this case drive a wedge between private return to capital and social return to capital, and savings are too low. The tax reforms of the last decades may have put too much attention to static neutrality in capital taxation and too little attention to intertemporal neutrality, Slemrod (1990a). Goulder and Thalmann (1993) showed that the US Tax Reform of 1986 which leveled the tax rates across the types and uses of capital (including a small increase in the taxation of housing), increased the intertemporal non-neutrality of the tax system since the average marginal tax rate on capital was higher after the tax reform. 6 The presence of other distortions generated by the tax system must also be taken into account.7 The tax system generates inefficiency in the labor market through both direct and indirect labor income taxation. The US Tax Reform of 1986 increased the intertemporal non-neutrality of the tax system, but this negative contribution to welfare was more than outweighed by the intratemporal efficiency gain of leveling the capital tax rates and reducing the labor income tax, Goulder and Thalmann (1993). Holmøy and Vennemo (1995) by using a static general equilibrium model with exogenous labor supply for the Norwegian economy, calculate that the Norwegian tax reform of 1992 extracted 40% of the potential intratemporal welfare gain of a total neutral system of capital taxation. However, the analysis did not analyze the effects of neutral housing taxation separately, and did not assess the importance of intertemporal and labor market efficiency effects. This paper analyzes the welfare effects of imposing tax neutrality between housing capital and financial capital. This implies a substantial increase in the marginal effective tax rate on housing capital.8 Since there is a high degree of neutrality between real capital except housing and financial capital in the current tax system, the housing tax reform will also lead to approximate neutrality between the different types and uses of real capital. The housing tax reform will generate a substantial increase in the tax revenue, and different rebating alternatives for the additional tax revenue as lump sum rebating or reductions in other distortionary taxes, are also considered. The housing tax reform is analyzed by using an intertemporal disaggregated general equilibrium model for the Norwegian economy. The model, which distinguishes between many industries, sectors, assets and input factors, is well designed for analyzing differences in capital taxation. Investments are determined by producers maximizing total discounted value of each firm, while the consumers maximize total discounted utility, giving an optimal consumption savings path. The economy is considered as a small open economy, which implies that domestic investments are not equal to domestic savings. As in Goulder and Thalmann (1993) the consumer demand for housing services is derived from utility maximization. The analysis then takes care of housing capital both as an investment good and as a consumption good. The model also incorporates imperfect competition (the large group case of monopolistic competition, Helpman and Krugman, 1985), which is absent in other analyses of capital taxation reforms. A neutral housing tax reform gives, as expected, an intratemporal efficiency gain. But with lump sum rebating of the additional tax revenue this efficiency gain is more than outweighed by the efficiency loss in the labor market and the intertemporal efficiency loss. The efficiency loss in the labor market is explained by lower labor supply following a substantial fall in the real wage rate caused by a higher price on housing services. The intertemporal efficiency loss is generated by lower total savings since total wealth (sum of real and financial capital) is reduced by the tax reform. This effect turns out to be small. The paper analyzes whether there are potential welfare improving tax reforms in the economy by using the tax revenue generated by the neutral housing tax reform to reduce either the marginal labor income tax or the marginal capital income tax. Lower marginal labor income tax outweighs the efficiency loss in the labor market generated by the housing tax reform. The main contributor to the welfare effect of lower marginal capital income tax is the love-of-variety effect. This effect is generated by the model of monopolistic competition and implies that variety is valued in its own right. Lower marginal capital income tax leads to an increase in the number of varieties, having a positive effect on welfare. The paper is organized as follows: in Section 2 an outline of the numerical model is provided. Section 3 provides data for the current system of capital taxation and discusses the conditions for neutrality. Section 4 specifies the policy alternatives, while Section 5 discusses a priori different sources of welfare effects. Section 6 presents the numerical results, while Section 7 concludes.

Efficient capital taxation has been one of the most important objectives for large tax reforms implemented in several countries during recent decades. The Norwegian Tax reform of 1992 took a large step towards tax neutrality between different capital types and uses. However, housing capital is still an exception. The marginal effective tax rate on housing is substantially lower than the marginal effective tax rates on other capital types and uses. The lenient taxation of housing capital has been criticized from an efficiency point of view. Too much capital is allocated to housing compared to other kinds of capital, giving an efficiency loss for the economy. Therefore, increased housing taxation may have a positive welfare effect. From a public finance perspective there may exist welfare improving tax reforms which include increased taxation of housing and where some or all of the additional tax revenue is used for reducing other distortionary taxes. Increasing the tax burden on housing may also be an additional source of financing public spending if other tax objects more easily are moved to low tax countries, and traditional tax bases are eroded. This paper analyzes the welfare effects of imposing tax neutrality between housing capital and financial capital. This implies a substantial increase in the marginal effective tax rate on housing capital. Since there is a high degree of neutrality between real capital except housing and financial capital in the current tax system, the housing tax reform also gives approximate neutrality between the different types and uses of real capital. The housing tax reform is analyzed by using an intertemporal disaggregated general equilibrium model for the Norwegian economy. A neutral housing tax reform gives an intratemporal efficiency gain. But with lump sum rebating of the additional tax revenue this efficiency gain is more than outweighed by the efficiency loss in the labor market and the intertemporal efficiency loss. The efficiency loss in the labor market is explained by lower employment following a substantial fall in the real wage rate caused by a higher price on housing services. The intertemporal efficiency loss is generated by lower total savings. This effect turns out to be small. There are potentials for welfare improving tax reforms in the economy if the tax revenue generated by the neutral housing tax reform is used to reduce either the marginal labor income tax or the marginal capital income tax. Lower marginal labor income tax outweighs the efficiency loss in the labor market generated by the housing tax reform. The main contributor to the welfare effect of lower marginal capital income tax is the love-of-variety effect that is generated by the model of monopolistic competition. The welfare effects of the different tax reforms are quite small. But in a world of increasing international mobility of input factors, firms and commodities, the welfare gains of reducing either the marginal labor income tax or the marginal capital income tax, may be underestimated in the model simulations. The analyses then support the argument that the directions of new tax reforms must incorporate higher tax burden on less mobile tax bases as housing combined with lower tax burden on more mobile or elastic tax bases.