Data and Fiscal Institutions of the Surveyed Countries
This chapter describes the data base and provides a detailed survey of the fiscal and social institutions of the countries examined in this book. It is explained how the authors dealt with the LIS microdata and how they rearranged the original data sets t
- PDF / 4,225,021 Bytes
- 81 Pages / 439.36 x 666.15 pts Page_size
- 4 Downloads / 206 Views
Data and Fiscal Institutions of the Surveyed Countries
5.1
Data
Our empirical investigation addresses several problems: Firstly, recall that the method of comparing tax progression which we proposed determines not a complete, but only a partial ordering. This provokes the question whether it is of major relevance because it might be that the tax schedules associated with their respective income distributions are so involved that only few clear-cut dominance relations emerge. In other words: is our method of comparison of tax progression in the real world only a will-o’-the-wisp, or can it command major occurrence? Secondly, if our first question is responded in the affirmative, what is the relative performance of the six proposed measures of the comparison of tax progression? What are the economic message and content of these methods? Thirdly, what is the relative importance of sufficient and necessary conditions of comparisons of tax progression? For which fraction of cases of greater tax progression the relative concentration curves would be strictly convex or concave, and for which they would just not cross the diagonal of the unit square without being strictly convex or concave? If the respective elasticity conditions do not hold, can we safely assume that greater tax progression is unlikely or can we expect it to be rather common? Fourthly, what is the pattern of comparisons of tax progression when dominance relations do not hold? Do we mainly encounter bifurcate or more intricate progression patterns? Is there a change of the progression pattern at a unique threshold, or do we have a whole series of changes of the progression relations? Fifthly, LIS data are sample data only. Are the patterns of dominance determined from our qualitative results statistically significant with respect to the data universe? Our analyses are carried out for both household data and equivalized data. We used the respective data from the Luxembourg Income Study Database (see LIS (2011)). We mainly focus on international comparisons of tax progression, but also carry out intertemporal comparisons for selected countries. Furthermore, we
C. Seidl et al., Tax Progression in OECD Countries, DOI 10.1007/978-3-642-28317-8 5, © Springer-Verlag Berlin Heidelberg 2013
71
72
5 Data and Fiscal Institutions of the Surveyed Countries
perform a sensitivity analysis with respect to the LIS equivalence scale parameter. To address these and related questions, we need household data for both gross and net incomes. In addition to that, for intertemporal comparisons for each country we required the data from at least three recent survey periods (“waves” in LIS terminology; in particular, we used Waves III up to VI if the respective data allowed). Data sets appropriate for our approach and showing gross incomes, direct taxes, payroll taxes, and net incomes were available just for 13 countries. In this chapter, we start with a short description of handling LIS data, continue with observations on the heuristics of progression concepts for grouped
Data Loading...
