Heterogeneity in Labour Supply Responses: Evidence from a Major Tax Reform
| DOI | http://doi.org/10.1111/obes.12168 |
| Date | 01 October 2017 |
| Published date | 01 October 2017 |
769
©2017 The Department of Economics, University of Oxford and JohnWiley & Sons Ltd.
OXFORD BULLETIN OF ECONOMICSAND STATISTICS, 79, 5 (2017) 0305–9049
doi: 10.1111/obes.12168
Heterogeneity in Labour Supply Responses: Evidence
from a Major Tax Reform*
Mauro Mastrogiacomo,†,‡ Nicole M. Bosch,§Miriam D. A. C.
Gielen,¶ and Egbert L. W. Jongen**
†VU University Amsterdam, De Nederlandsche Bank and Netspar, De Boelelaan 1105,
1081HV, Amsterdam, The Netherlands
‡VU University Amsterdam, Faculty of Economics and Business Administration, De
Boelelaan 1105, 1081 HV Amsterdam, The Netherlands (e-mail: m.mastrogiacomo@vu.nl)
§CPB Netherlands Bureau for Economic Policy Analysis, Bezuidenhoutseweg 30, 2594 AV
The Hague, The Netherlands (e-mail: n.m.bosch@cpb.nl)
¶Ministry of Social Affairs and Employment, Parnassusplein 5, 2511 VX The Hague, The
Netherlands (e-mail: mgielen@minszw.nl)
**CPB Netherlands Bureau for Economic Policy Analysis and Leiden University, CPB
Netherlands Bureau for Economic Policy Analysis, Bezuidenhoutseweg 30, 2594 AV The
Hague, The Netherlands (e-mail: e.l.w.jongen@cpb.nl.)
Abstract
We use a large and rich administrative household panel data set to estimate labour supply
responses for a large number of subgroups in the Netherlands. The identification of the
parameters benefits from a major tax reform in the data period. Weuncover large differences
in behavioural responses. In particular, we find differences in labour supply responses
between households with and without children that are much bigger than suggested by
previous studies that had to pool these household types in the estimation of preferences.
An efficient tax-benefit system should take the substantial heterogeneity in behavioural
responses into account.
I. Introduction
In the wake of the Great Recession, and faced with an ageing population, policy makers
are reconsidering their tax-benefit policies and the trade-off between equity and efficiency.
The response of labour supply to changes in financial incentives, traditionally measured
by the wage elasticity of labour supply, plays a key role in the efficiency losses from
JEL Classification numbers: C25, C52, H31, J22.
*The authors thank the editor Jonathan Temple and three anonymous referees for very helpful comments and
suggestions. Furthermore, we have benefitted from comments and suggestions by Hans Bloemen, Richard Blundell,
Henk-Wim de Boer, Chris van Klaveren,Arthur van Soest, Isolde Woittiez, participants of the IZA Labour Supply
Workshop October 2011 in Bonn and participants of the EALE Conference September 2012 in Bonn. Remaining
errors are our own.
770 Bulletin
redistribution via the tax-benefit system.1Furthermore, as policy makers target taxes and
benefits at different groups in the labour market, they also require detailed knowledge of
labour supply responses for different subgroups (e.g. singles vs. couples, households with
vs. without children, high vs. low skilled), and differentdecision margins (e.g. participation
in persons vs. hours worked per employee), to optimize the tax-benefit system for given
political preferences for redistribution.
In this paper, we use a large and rich household panel data set to estimate the labour
supply responses for a large number of subgroups within the labour market in the Nether-
lands. The size of the data set allows us to estimate preferences separately for subgroups
that had to be pooled in earlier studies. Furthermore, a major tax reform in our data period
benefits the identification of the structural parameters.2
We estimate the labour supply responses for a large number of subgroups within the
Dutch labour market using the Labour Market Panel (Arbeidsmarktpanel in Dutch) of
Statistics Netherlands (2009). The Labour Market Panel is an administrative household
panel data set covering the years 1999–2005. The 2001 tax reform in the Netherlands
caused major exogenous variation in budget constraints.We use a discrete choice model to
estimate preferences over income and leisure and the corresponding labour supply elastici-
ties. To determine household budget constraints we use the advanced tax-benefit calculator
MIMOSI (Romijn et al., 2008).3
Our main findings are as follows. First, there are large differences in the labour supply
elasticity across different demographic groups and decision margins.We find that the labour
supply elasticity of men and women is quite similar when they are single. But in couples,
men have much smaller elasticities than women. Furthermore, cross-elasticities of men’s
wages on women’s labour supply in couples with children are non-negligible. We find the
highest labour supply elasticities for single parents, in particular for single parents with
preschool age children. For all household types we find that low-skilled individuals have
much higher labour supply elasticities than their high-skilled counterparts, and that the
participation response is much larger than the hours-per-week response. We also find that
immigrants have a higher labour supply elasticity than natives.
Second, differences in the labour supply elasticity between households with and with-
out children are much larger than previously thought. Earlier studies had to pool these
household types when estimating preferences, due to the small number of observations,
and they find small differences in labour supply elasticities between households with and
without children. Indeed, in their overview paper, relying on estimates where they pool
households with and without children, Bargain, Orsini and Peichl (2014) conclude for
women in couples that ‘... elasticities are only slightly larger for women with children’
(Bargain et al., 2014, p. 738). Furthermore, for single women they conclude that ‘[S]ingle
1A related strand of literature studies the elasticity of taxable income (ETI) to measure the distortions of taxation,
see e.g. the overview in Saez, Slemrod and Giertz (2012). The ETI also captures other responses to changes in tax
rates, like changes in effort and tax evasion. However, there is an active academic debate over whether the ETI is
a sufficient statistic to measure the distortions from taxation, see e.g. Chetty (2009), Piketty, Saez and Stantcheva
(2014) and Doerrenberg, Peichl and Siegloch (2017).
2Chetty (2012) shows that large changes in budget constraints allow for more precise estimates of structural
parameters, by reducing the role of frictions in behavioural responses.
3MIMOSI is the official tax-benefit calculator of the Dutch government,developed and maintained at CPB Nether-
lands Bureau for Economic PolicyAnalysis.
©2017 The Department of Economics, University of Oxford and JohnWiley & Sons Ltd
To continue reading
Request your trial