Microsimulation & Econometric Data Analysis

The German withholding and income tax system for married couples is under attack from different angles. One potential improvement is to introduce the so-called Factor Method. To understand the likely effects in detail, we carry out extensive analyses and microsimulation forecasts based on large samples of microeconomic data. This work on behalf of several federal ministries involved here supports the ongoing legislative process.

The existing tax system for married couples is being criticized from economic as well as gender equality points of view. One specific reform option is part of the current coalition agreement and is to be implemented in the present legislative period. It should lead to greater fairness in the perceived tax burden of spouses. Specifically, it replaces the tax class combination III and V with the so-called Factor Method. This method applies tax class IV to both spouses, and it also takes into account the later splitting advantage in the case of unequal wage incomes by reducing the withholding tax of both spouses by the same factor.

The withholding tax rate in tax class V is significantly higher than in class IV. Thus, the Factor Method leads to lower average and marginal withholding tax rates for the spouse who opted for tax class V up to now – the wives in the majority of cases. The Factor Method does not change the final income tax burden of a married couple. However, the higher monthly net income of the spouse previously taxed at a higher rate may be a positive incentive to work longer hours or to take up more challenging, better-paid jobs.

To support the specialists in several federal ministries involved here in assessing the consequences of this change, we carry out extensive analyses and microsimulation forecasts using large samples of microeconomic data. Our microsimulation model based on wage and income tax statistics lets us calculate the tax effects, i.e. shifts in the wage tax burden between spouses and temporal shifts between wage and income tax and can estimate the number of taxpayers affected. In addition to the short-term fiscal effects, we can estimate the expected longer-term behavioral adjustments in the labor supply. Here we use another of our microsimulation models which is based on data from the Socio-Economic Panel, a large representative household survey that contains information on individuals’ shifts in working hours over time. The results of our work provide important insights for our lawmakers and contribute to an evidence-based tax policy.