In addition to affecting individuals’ current labour supply, taxation might also influence other behaviour that affects labour output (including the choice of occupation, investment in human capital, personal effort, etc), and we know relatively little about these impacts (see, e.g. Feldstein (2012) criticism that the recent comprehensive Mirrlees Review of the tax system mostly ignored these margins).
In this project we will contribute to the literature by analysing the impacts of taxation on some of these rarely-studied margins of responses with cross-country micro data from the Luxembourg Income Study (LIS) and EU Labour Force Survey (EU LFS). Building on the framework of Jäntti et al (2013), we utilise tax changes between groups of people within a country, and tax changes across approximately 15 OECD countries for which relevant data is available, to identify the effects of taxation on some of the longer run decisions, such as the decision to acquire entrepreneurial income, or the decision to obtain higher education. This approach provides evidence on some neglected but yet potentially very important issues in tax research, with an arguably more plausible identification strategy than some of the structural econometric analyses.
To be more precise, when analysing the choice between entrepreneurial vs labour income in LIS data, we can compare groups of individuals, as in Blundell et al (1998) (where groups are formed by cohort and education), and use within-country and cross-country changes over time in tax rates on entrepreneurial income and labour income as instruments for the tax treatment of these two forms of compensation. Alternatively, one can examine how the frequency of job changes or the type of occupation (measures used to gauge career progress in Keane and Wolpin, 1997) in EU LFS across similar groups is affected by tax progressivity. Finally, the educational decisions for cohorts can be explained by time-varying cross-country differences (that are likely to be much larger than within a single country) in the take-home pay for different educational groups.