Description
To gain further insight into how tax-benefit systems affect behaviour, we next turn to the effects of tax-benefit complexity. Recent literature on behavioural public economics (Congdon et al., 2011) has shown that the complexity of tax-benefit systems affects individuals’ understanding of tax incentives. Looking at the tax system from a broader perspective, i.e. taking into account the effects of other features of the system besides tax (or benefit) rates and the tax base, is a relatively novel strand of public economics (Slemrod 2013). Complexity may have adverse effects on efficiency (e.g. providing costly incentives that do not have the desired effects on behaviour) and on the income distribution (e.g. if poor people are ill-equipped to deal with complexity – Shah et al. (2012) report potential reasons). We analyse how complexity affects the take-up of benefits and the accuracy of tax reporting, using Finnish data that provides exogenous variation (based on place of residence) in tax-benefit complexity.
The type of complexity we will analyse relates to the practices used to determine individuals’ eligibility for benefits and/or their tax liability – how many forms one needs to fill out, how many pieces of information to report, and so on. Our first application looks at how complexity affects benefit take-up (see e.g. Currie 2006), utilising individual-level register data combined with self-collected municipal-level data on the procedures for applying for childcare allowances in Finland (1995-2007). In this setting (i) we have a well-defined measure of complexity, (ii) there is exogenous variation in this measure, (iii) welfare stigma (see e.g. Moffitt 1983) is likely not a concern as the programme is utilised by individuals from a wide variety of backgrounds and (iv) we are thus able to examine heterogeneities in the effect of complexity on take-up according to individual characteristics. To our knowledge, the combination of features (i) – (iv) is unique in the literature.
The second application concerns the effects of the complexity of the income tax-filing system on taxpayers’ reporting behaviour. Reporting behaviour, on the other hand, affects individual tax liabilities (hence tax incidence) and tax revenue (hence financing of the welfare state). We utilise data from a natural experiment, where a proportion of Finnish taxpayers received a pre-filled tax return instead of having to fill out a complete multiple-page form (the normal practice). Again we combine individual-level register-data on approx. 1 million Finns (1995-2000), with municipal-level information on the rules that determined individuals’ inclusion into the experiment. Conditional on these criteria, assignment of individuals into the experiment was random (determined e.g. by place of residence with no possibilities for self-selection). This allows us to estimate the causal effect of the details of the tax-filing system on reporting behaviour (accidental misreporting and/or purposeful tax evasion – on the latter issue, see e.g. Kleven et al. 2011).
Researchers
Dr Jani-Petri Laamanen, Lecturer in Economics, University of Tampere, jani-petri.laamanen@uta.fi
Professor Kaisa Kotakorpi, Professor of Economics, University of Turku, kaisa.kotakorpi@utu.fi
Mr Tuomas Kosonen, Senior Researcher, Government Institute for Economic Research, Finland, tuomas.kosonen@vatt.fi
Professor Tomer Blumkin, Professor of Economics, Ben-Gurion University of the Negev, Israel, tomerblu@bgu.ac.il