Traditionally, most countries use the corporate tax as a means to tax capital income. A worrisome aspect of the corporate tax is that multinationals can avoid the tax by shifting profits to affiliates in jurisdictions with a lower corporate tax rate. This profit shifting leads to tax competition between jurisdictions, eventually resulting in a race to the bottom.
In this project, we have studied whether taxing a different base allows governments to tax capital owners in a smarter way. A payroll tax is a tax levied on firms on the basis of the wages they pay to employees. Unlike the corporate tax, the payroll tax is taxing assets that are firmly rooted within a country, namely employees and the wages they receive. The corporate tax can be avoided through profit shifting, whereas avoiding the payroll tax is much more costly.
Nevertheless, the payroll tax may come with its own potential downsides. According to economic theory, firms will try to shift (part of) the payroll tax on to their employees by reducing the wage rates. If too much is passed on, the payroll tax cannot serve to tax capital. However, recent empirical research indicates that the payroll tax mainly falls on the firm, and is thus an effective way of taxing capital.
The purpose of this project was to investigate who pays the payroll tax in Norway. Is it the firm or is it workers, or is the tax shared? In addition, we aimed to estimate the labor market consequences of payroll taxation.
The first part of our project is a methodological contribution that overcomes the standard challenge in empirical research that identifying both the demand and the supply elasticity requires at least two instruments. We have shown that exogenous variation in a single tax can be used to simultaneously identify both the demand and the supply elasticity. Moreover, we have developed a methodological toolkit that applied researchers can use to identify supply and demand elasticities.
Our method can be applied to payroll taxation, as it allows identification of the labor demand and supply elasticity using exogenous variation in the payroll tax. However, the method can also be applied to study the effects of variation in the value-added tax, the income tax, and various housing taxes. Moreover, it can serve as a bridge between reduced-form and structural estimation models. As such, this part of the project contributes to empirical tax research in public economics, and to economics in general. The paper describing the method is published in Econometrica.
We have applied our method, as well as the more standard reduced-form methods to investigate the effect of the payroll tax in Norway. Our data contains all manufacturing plants in Norway. Our results suggest that a one percent increase in the payroll tax reduces the wage rate by 0.5 percent (non-significant).
Employment responses within plants appear to be limited. Applying our new method to the data we have not found labor demand or supply elasticities that are significantly different from zero.
We have also considered the effect of payroll taxation on sales and investment. We found very large negative (but insignificant) coefficients between the payroll tax and both sales and investment.
We have looked in more detail at the labor market responses of entrepreneurs at the margin. We found some evidence that reduced payroll tax rates are associated with less self-employment. We also found that the payroll tax influences the choice of organization form between sole-proprietorship and incorporation: A one percentage point reduction in the payroll tax increases the probability of incorporation with approximately 0.3 to one percentage points.
Overall, we conclude that the payroll tax can be used to tax capital at the firm level, as half of the incidence remains with employers.
In Norway, firms pay the corporate tax and the payroll tax. In 2012, the payroll tax represented almost twice as much in tax revenue as the corporate tax. Recent empirical literature suggests that the payroll tax in large falls on the firm, whereas early research on Norwegian data indicates that the tax is shared.
This project seeks to answer two questions. First, to investigate who bears the burden of the payroll tax. Is it capital or workers, or both? The second research question concerns how firms an d households respond to the payroll tax. From this question follows two secondary objectives:
1. How do firms avoid and evade the payroll tax?
2. How do households avoid and evade the payroll tax?
Firms can avoid the payroll tax by moving production faci lities to different jurisdictions, hiring workers from the black economy, or hiring self-employed persons rather than employing workers. Members of households can become self-employed or enter the black economy.
Sweden has payroll tax of 31.4%, Norway?s peak rate is 14.1%, whereas Denmark does not levy a payroll tax. Sweden and Denmark will have a tax rate of 22%, if not lower, in the near future, whilst the Norwegian rate stands at 27%. This project asks the question of whether a policy of lowering the corporate tax rate in response to tax competition and increasing the payroll tax is a sensible policy. Put differently: can the payroll tax serve as a backstop for taxing capital?
Researchers in Norway are well equipped to answer these questions. The pa yroll tax rate in Norway is differentiated by region. Moreover, the payroll tax system has been reformed several times over the last decades. Finally, Norway has high quality employer-employee data for its entire labor force over a time frame of several decades. This allows us to estimate the impact of payroll taxation on the most important aspects of the Norwegian economy, through regression discontinuity and difference-in-difference designs.