Umeå University's logo

This paper develops a maximum likelihood (ML) bunching estimator of the elasticity of taxable income (ETI). Our structural approach provides a natural framework to simultaneously account for unobserved preference heterogeneity and optimization errors and for measuring their relative importance. We characterize the conditions under which the parameters of the model are identified and show that the ML estimator performs well in terms of bias and precision. The paper also contains an empirical application using Swedish data, showing that both the ETI and the standard deviation of the optimization friction are precisely estimated, albeit relatively small.

The elasticity of taxable income (ETI) is a central statistic for tax policy design. One purpose of the present paper is to use Monte Carlo simulation techniques to assess the bias and precision of the prevalent estimators in the literature, the IV-regression estimator and the bunching estimator. Thereby, we aim to provide arguments in favor of, or against, using these methods. Another is to suggest indirect inference estimation to improve the quality of the measurement of the ETI. While IV-regression estimators perform well in terms of bias under certain conditions, they are more variable than bunching estimators. We also find that bunching estimators can be biased downward. The estimators based on indirect inference principles are practically unbiased and more precise than the other estimators.

This paper analyzes the effect of the introduction of a flat tax schedule on capital income on the progressivity of the German personal income tax, with a particular focus on top income groups. The reform-induced change in net incomes is proposed as a measure for progressivity changes at the top. Changes in vertical and horizontal equity are analyzed and broken down by reform component. The analysis is based on a micro-level panel dataset of income tax returns between 2001 and 2006 that is particularly representative for the top of the income distribution. The panel structure is used to construct a permanent reform effect that is less prone to annual volatility in income composition. The reform is found to be regressive. It is shown that the reform included substantial loop-holes. Within the top of the distribution, the gain in net incomes is spread widely, but is not negative under plausible assumptions on reporting adjustments. 

We propose a maximum likelihood method to improve the bunching approach of estimating the elasticity of taxable income (ETI), and derive estimators for several model settings such as bunching with optimization frictions, notches, and heterogeneity in the ETI. Modelling optimization frictions explicitly, our estimators fit the data of several published studies very well. In the presence of a notch, the results can differ substantially from those obtained using the polynomial approach. If there is heterogeneity in the ETI, the elasticity among those who bunch exceeds the average elasticity in the population.

We propose a maximum likelihood (ML) based method to improve the bunching approach of measuring the elasticity of taxable income (ETI), and derive the estimator for several model settings that are prevalent in the literature, such as perfect bunching, bunching with optimization frictions, notches, and heterogeneity in the ETI. We show that the ML estimator is more precise and likely less biased than ad-hoc bunching estimators that are typically used in the literature. In the case of optimization frictions in the form of random shocks to earnings, the ML estimation requires a prior of the average size of such shocks. The results obtained in the presence of a notch can differ substantially from those obtained using ad-hoc approaches. If there is heterogeneity in the ETI, the elasticity of the individuals who bunch exceeds the average elasticity in the population.

Measuring the elasticity of taxable income (ETI) is central for tax policy design. Yet, there are few arguments which support or infirm that current methods yield measurements of the ETI that can be trusted. Our first purpose is to use simulation methods to assess the bias and precision of the prevalent methods used in the literature (IV estimation and bunching methods). Thereby, we aim at (i) explaining the huge differences in empirical results, and (ii) providing arguments in favor of or against using these methods. Our second purpose is to suggest indirect inference estimation to improve the quality of the measurement. We find that the IV regression estimators may suffer from considerable bias and be quite imprecise, whereas the bunching estimators perform better in our controlled environment. We also show that using more of the information available in the data, estimators based on indirect inference principles produce more precise estimates of the ETI than any of the most commonly used methods.

I analyze German top income mobility using micro-level panel data of personal income tax returns which are highly representative for top income taxpayers for the years 2001–06. Top income mobility is assessed in three dimensions: (i) persistence in top income fractiles and its stability over time, (ii) measures of individual mobility that are not dependent on the fractile size: the degree of mobility between equally sized groups and mobility in ranks, and (iii) mobility's impact on top income shares. Persistence in top income fractiles is comparatively high and fairly stable across the analyzed period. Top income recipients are less prone to downward mobility and see less variation in annual ranks than less rich tax units. Mobility's impact on income concentration is moderate. The top percentile's share is reduced by roughly 5 percent over six years.

A large literature has documented top income share series based on income tax statistics using the common methodology established by Piketty (2001, 2003). The widespread disappearance of capital income from the income tax base poses a major challenge to the comparability of these series both over time and between countries. In Germany, capital income was gradually excluded from the income tax base between 2001 and 2009. Using a rich data set containing all income taxpayers' les we provide a homogeneous top income share series including full capital incomes from 2001 to 2010. Missing capital income since 2009 is extrapolated using a composite measure of stock dividends and interest income tax ows. We nd that up to the top percentile the drop displayed in the German raw-data series in 2009 is largely attributable to the disappearance of capital income from the income tax base and not to the crisis. However, the very top of the income distribution is disproportionately hit by the crisis.