Taxing the added value is not a good idea edit

July 21, 2006

In his seasonal greetings to the press French president Jacques Chirac proposed to widen the tax base for employers' contributions to social security from wages to value added. The idea of cutting employers' payroll contributions to foster employment is not distinctively French. Instead the idea of financing it by widening the tax base from wages to value added certainly is: In general, governments tend to finance reductions in employers' social security contributions through the general tax system. This is the case of the German coalition government that plans to finance a reduction in employers' contributions to unemployment insurance from 6.5% to 4.5% through receipts from the general tax system.

-->To illustrate how the introduction of a value added employers' contribution (VAC) to social security modifies the implicit tax rates on labour and capital, consider the French national accounts for the year 2004. Gross wages of employees of 430 billion Euro, 149 billion Euro of employers' payroll contributions and gross operating surplus of 277 billion Euro add up to total value added at factor costs of 857 billion Euro. In other words, the tax rate on gross wages of employees (labour) implied by employers' social security contributions corresponds to 34.7 percent and the tax rate on gross operating surplus (capital) to 0 percent. Indeed only the implicit tax rate on capital implied by the social security system is 0. Obviously, there are other taxes that act as implicit tax on capital: corporate taxes, taxes on vehicles, buildings, etc. Would the government decide to shift the tax base for employers' social security contributions entirely to value added, it could achieve revenue neutrality by taxing gross wages of employees and gross operating surplus at the uniform rate of 21.1 percent. Clearly, the government would decide to levy only part of employers' social security contributions on value added while maintaining the current system for the other part. However, this accounting exercise illustrates the following basic point that holds independently of the specifics of the implementation of the VAC: Since in France the ratio of gross wages of employees to gross operating surplus is approximately 1.5, a 1 percentage point decrease in the implicit tax rate on gross wages of employees has to be compensated by a 1.5 percentage point increase in the implicit tax rate on gross operating surplus to be revenue neutral for the social security system.

As an illustrative benchmark for the economic analysis of the introduction of the VAC, consider for a moment a closed economy. Clearly, the reduction in the implicit tax rate on labour and the simultaneous increase in the tax rate on capital will lead to a substitution of domestic capital by domestic labour. In the short run, it can thus be expected that the introduction of the VAC leads to a positive effect on employment. Even in this closed economy setting there are three major caveats, however.

Firstly, the same substitution effect that yields short run employment gains, reduces these positive employment effects in the long run. The increase in the implicit tax rate on capital leads to a decrease in the rate of capital accumulation and in the marginal productivity of labour. In the presence of labour market rigidities this may lead to a decrease in employment in the long run.

Secondly, the reduction in the implicit tax rate on labour will increase labour demand at constant wages. This is likely to drive up wages and to reduce the net effect on labour cost faced by employers.

The third and last shortcoming is even more subtle. Within the firm that produces with the average French capital intensity, the increase in the cost of capital is exactly offset by the reduction in the cost of labour. However, the introduction of the VAC may change profits for firms producing with a capital intensity that differs significantly from the French average. Accordingly any firm producing with a very capital intensive technology will incur profit losses from the introduction of the VAC since the higher implicit tax payments on capital will outweigh its reduced payroll contributions. OECD data for the year 2000 indicate that some of the most capital intensive sectors in the French economy such as the chemical industry or transport equipment also have high values of R&D intensity. By implicitly taxing these sectors more heavily, the introduction of the VAC may therefore not only act as a brake on capital accumulation but also as a brake on innovation.

With these preliminary considerations at hand, it can be proceeded to the economic analysis of the introduction of the VAC in an open economy. Now, international trade enables substitution between domestic and foreign factors of production. Firms can choose to offshore their labour intensive stages of production to low wage countries and capital intensive stages of production to countries with a low cost of capital.

Indeed, the VAC reduces firms' incentives to outsource the labour intensive stages of production but increases firms' incentives to outsource the capital intensive stages of production. The relevant question is whether quantitatively significant changes in behaviour can be expected from these changes in incentives.

Outsourcing the labour intensive stages of production has a fixed cost (moving the factory, finding a subcontractor, etc.). Accordingly, firms that do already offshore their labour intensive stage of production are unlikely to be lured back to France by the introduction of the VAC: they have already sunk the fixed cost.

Secondly, labour intensive firms are likely to consider only countries with very low labour costs in their offshoring decision. Given labour cost ratios in 2004 ranging from approximately 4/1 between France and the Czech Republic or Hungary and approximately 10/1 between France and Mexico, even a decrease in the implicit tax rate on labour in the range implied by the above accounting exercise would only have a marginal impact on the firm's offshoring behaviour. In other words, the wage savings motive dwarves the tax savings motive for the offshoring of the labour intensive stage of production.

The extension to an open economy setting thus reinforces the insights from the simple closed economy setting: The beneficial effects on employment from the introduction of the VAC are likely to be small while the adverse effects on capital accumulation may be non-negligible. These adverse effects are likely to erode the tax base for the VAC which may constrain the government to either increase the VAC rate and thus to induce more very capital intensive firms to outsource the capital intensive stage of production or to reverse the reform and re-introduce the current system of employers' contributions to social security. The reform may thus be economically unsustainable and would furthermore have to deal with a bunch of exemptions which, in turn, creates new distortions. Among these exemptions are small enterprises that do not dispose of adequate accounting systems to determine value added, enterprises that employ mainly low skilled labour that benefit from reduced rates under the current system and holding companies that display extremely high capital-labour ratios.

The idea of reducing employers' payroll contributions to the social security system makes economic sense. In a context in which firms are becoming increasingly internationally mobile and in which other countries are reducing their taxes on capital, it is the idea of financing the reduction of employers' payroll contributions by unilaterally introducing an implicit tax on capital that appears problematic. A look at the policies of the German neighbour might be worthwhile.