Making the Switch From Joint to Individual Taxation in Luxembourg. Cost, Behavioural Response and Welfare Effects

TL;DR: In this paper, the effect of a move from joint to individual taxation system using 2,276 couple household living in Luxembourg was studied, where the authors estimate simultaneously labour supply and social assistance participation, exploiting a discrete choice model.

Abstract: We study the effect of a move from joint to individual taxation system using 2,276 couple household living in Luxembourg We estimate simultaneously labour supply and social assistance (RMG) participation, exploiting a discrete choice model We focus on the distributional, work (extensive and intensive margin) incentive, and the social welfare effect of introducing a mandatory individual taxation system in Luxembourg The work incentive of married women increases by 227% in intensive margin and 258% in extensive margin after the reform The incentive of married men is almost zero Equivalised disposable income, after the behavioural adjustment, decreases on average 21 per cent After adjustments to direct and indirect taxes, the net revenue-neutral result is a budget surplus for the central government of around €10 million

Summary

Introduction

• Making the switch from joint to individual taxation in Luxembourg.
• 1 Making the switch from joint to individual taxation in Luxembourg.
• Previous empirical research has evaluated the labour market effect of hypothetical shifts from joint to individual taxation.
• Section 6 summarizes the main findings of the paper.
• Taxpayers are accorded a tax class according to their family status.

2.2 Marginal and average tax rates

• Incentives to work are often measured using the average or the marginal tax rate.
• The average tax rate (ATR) is defined as the total tax paid divided by total income while the marginal tax rate (MTR) is defined as the income tax rate applied for a specific tax bracket.
• The higher the marginal tax rate a worker faces, the more will be taxed away from each additional Euro earned.
• One important characteristic of the Luxembourg taxation system is that MTRs exceed ATRs by a large margin .
• High MTRs reduce the incentive to work extra hours for those already in the labour force and may also reduce the incentive to join the 6 labour force in the first place, especially for the secondary earner in a couple in a system of joint taxation.

2.3 Tax incidence

• In this section the authors present the incidence of taxation among couple households in 2009.
• Almost 24% of households have a zero ATR in the joint taxation system compared to just 8% in the individual tax system, indicating the higher tax burden imposed on low earning households in an individual taxation system.
• This data is the Luxembourgish version of the European Union-Statistics on Income and Living Conditions (EU-SILC) and includes variables on various sources of income, benefits (means tested as well as non means tested), wages, basic individual and household characteristics such as age, gender, education, nationality, marital status, employment status, children and region.
• Households in which both spouses are unavailable for the labour market, on 3.
• The husbands in their sample have, on average, many more years of work experience than their wives (26 compared to 16).

3.2 Budget constraints

• Figure 3 shows that the married households in which the spouses have unequal incomes lose money in an individual taxation regime compared with a joint taxation regime.
• In the joint taxation system (dashed grey line), the initial disposable income is large (as the primary earner uses the secondary earner’s tax credits and tax bands) but decreases rapidly compared to gross income as the secondary earner joins the labour force and increases her hours of work.
• Disposable income increases by more in the individual regime than in the joint regime as the secondary earner has access to her own tax credits and low band rates.
• The difference that remains between the individual budget constraints for the secondary earner once both spouses are working full-time can be attributed to the primary earner.
• Using PSELL-3 data from 2010 (income reference period 2009) and Luxembourgish tax benefit rules for 2009, LuxTaxBen simulates household disposable income before (joint tax system) and after (individual tax system) the hypothetical reform.

4.1 The structural labour supply model

• To evaluate individual behavioural reactions induced by switching from joint to individual taxation the authors use a direct translog utility model9 as proposed by van Soest (1995)10.
• These variables are, in general, annually determined.
• The main form of social assistance, the RMG, is calculated at the household level in both the joint and individual systems.
• 15 account for observed and unobserved preferences for leisure and welfare participation, a set of parameters for individual and household characteristics as well as a set of random terms are also included in the model.
• The x-vector includes k observed characteristics such as age, level of education, number of children in the household and a regional dummy.

4.2 Estimation results and Model prediction

• Equation 1 is estimated using a direct translog utility model that allows both spouses to choose among 18 various combinations of market hours, household disposable income and welfare participation.
• A stronger preference for leisure is further observed in households where the female has a secondary education.
• The underlying unobserved heterogeneity, which splits the population into three subgroups (three mass points), is assumed to be based on the person’s latent preference or ability to work.
• Labour supply and welfare participation decisions of partners in couples have been estimated jointly.

4.3 Labour supply elasticities

• Different studies report elasticities of different magnitudes (and sometimes sign), depending on methodological choices, including the type of data used (tax register data or interview-based surveys), sample selection (e.g. households with or without children), the period of observation and the estimation method (see for example Heim, 2007, Kornstad and Thoresen, 2007, Fuchs et al., 1998, Bargain and Peichl, 2014).
• There is consensus that females usually have a higher wage elasticity than males.
• The authors simulate the elasticities by increasing gross wages by 10% (the elasticities in the Table 5 are expressed as the percentage change in labour supply for a 10% increase in gross wages).
• The results show that a 10% increase in the female wage rate raises the overall labour supply of married women by about 0.44% while a 10% increase in the male wage rate increases the labour supply of married men by about 0.29%.
• The result implies very low labour elasticities.

5.1 Fiscal impact

• So the authors simulate the transition, and modify taxes at the same time to keep the government budget neutral.
• The first round or morning after effect and the behavioural or second round effect.
• Table 6 demonstrates the aggregated outcomes for the central governmental budget.
• Indeed, no changes in an individual taxation system when behavioural response is not taken into account.
• The effect of increasing or decreasing household disposable income will have knock-on effects on consumption and, therefore, on VAT receipts.

5.1 Labour supply response

• Table 7 shows the labour supply response to switching from joint to individual taxation in Luxembourg.
• Married men’s working hours can be expected to marginally decrease (by 0.13%).
• At the extensive margin, an increase in the labour force participation rate of 0.03% for men is noted while an increase in the labour force participation rate of married women of 16 There was no change of the standard VAT rate between 1992 to 2015.
• The authors calculate the standard error of these percentage changes at the mean using the delta method.
• The last column of Table 7 shows that the percentage change in working hours and participation for males is statistically insignificant while that of females is statistically significant at the 1% level.

5.2 Winners and losers

• The authors used unitary household labour supply model (equation 1) to determine the behavioural reaction and ranked households by their equivalised (adjusting for the number of adults and children) disposable income.
• Equivalised disposable income, after the behavioural adjustment, decreases on average 22 around 2 per cent.
• The households who gain the most are located in the top of the income distribution while the middle and bottom deciles lose the most.
• Note that the evaluation are based on a unitary model of the household, where equivalence scales are used to accommodate different needs depending on the family size.
• An alternative model entirely based on individual income and welfare would probably register an increase of well-being in the lower part of the individual income distribution, due to the increase of labor supply in the female population.

5.4. Validation exercise

• EUROMOD which is an integrated tax-benefit calculator covering the systems of the EU-27 countries (Sutherland and Figari (2013)).the authors.
• The welfare measure that the authors have used in the paper is relevant to measure the change in the level or distribution of disposable income.
• Quantifying the Disincentive Effects of Joint Taxation on Married Women's Labor Supply.

Figures

Figure 4: Individual budget constraint for secondary earner in Luxembourg

Table 8: Comparing simulation results of individual taxation reform in the LuxTaxBen software to findings from the same exercise using EUROMOD

Table 5. Effect of 10% wage increase on labour supply of men and women in couples14

Table 1: The schedule of income tax in Luxembourg in 2009

Table 6. The effect of an individual tax reform on central governmental budget using cohabiting sample age 18-65 (in EURO)

Figure 2.2 The proportion of households (aged 18-65) at each Marginal Tax Rate

Table 7. Change in working hours and participation (Sample size 2,276)

Figure 3: Household budget constraint for Luxembourg

Figure 1. Marginal Tax Rate (MTR) and Average Tax Rate (ATR) of couple household in Luxembourg in 2009.

Table 2: Summary statistics for married households in Luxembourg

Figure 5. Percentage change of equalised disposable income by decile after behavioural response to switch to individual taxation reform

Figure 2.1. The proportion of households (aged 18-65) at each Average Tax Rate

Table 4. The percentages of actual and predicted welfare participation of couple household with 18 combinations using PSEL 2010 (income year 2009).

Table 3 The percentages of actual and predicted working hours for males and females in couples with 18 combinations using PSELL 2010 (income year 2009).

Full text

www.liser.lu
WORKING PAPERS
Making the switch from joint to
individual taxation in Luxembourg.
Cost, behavioural response and welfare
effects
Nizamul ISLAM
1
Karina DOORLEY
2, 3, 4
Lennart FLOOD
5
1
Luxembourg Institute of Socio-Economic Research (LISER), Luxembourg
2
Economic and Social Research Institute (ESRI), Ireland
3
Institute for Labor Studies (IZA), Germany
4
Trinity College Dublin (TCD), Ireland
5
University of Gothenburg, Sweden
n° 2020-05 May 2020

LISER Working Papers are intended to make research ndings available and stimulate comments and discussion.
They have been approved for circulation but are to be considered preliminary. They have not been edited and have not
been subject to any peer review.
The views expressed in this paper are those of the author(s) and do not necessarily reect views of LISER.
Errors and omissions are the sole responsibility of the author(s).

1
Making the switch from joint to individual taxation in Luxembourg.
Cost, behavioural response and welfare effects.
NIZAMUL ISLAM KARINA DOORLEY LENNART FLOOD
Abstract:
We study the effect of a move from joint to individual taxation system using 2,276 couple
household living in Luxembourg. We estimate simultaneously labour supply and social
assistance (RMG) participation, exploiting a discrete choice model. We focus on the
distributional, work (extensive and intensive margin) incentive, and the social welfare effect of
introducing a mandatory individual taxation system in Luxembourg. The work incentive of
married women increases by 2.27% in intensive margin and 2.58% in extensive margin after
the reform. The incentive of married men is almost zero. Equivalised disposable income, after
the behavioural adjustment, decreases on average 2.1 per cent. After adjustments to direct and
indirect taxes, the net revenue-neutral result is a budget surplus for the central government of
around €10 million.
Key words: Microsimulation, Labour supply, Joint taxation, Welfare
JEL classification numbers: B21, B31, D31, H24, H31

Nizamul Islam is affiliated to the Luxembourg Institute of Socio-Economic Research (LISER). E-mail:
nizamul.islam@liser.lu. The authors are grateful to Eugenio Peluso LISER Luxembourg, and other anonymous
referees for their valuable comments.

Karina Doorley is affiliated to the Economic and Social Research Institute (ESRI), Institute for Labor Studies
(IZA) and Trinity College Dublin (TCD). E-mail: karina.doorley@esri.ie

Lennart Flood is affiliated to the Department of economics, University of Gothenburg. E-mail:
Lennart.Flood@handels.gu.se.

2
1. Introduction
Since January 2018, members of a couple can elect to be taxed individually in Luxembourg.
Under individual taxation, each spouse's labour income is taxed independently from the other
spouse, meaning that two households with the same household income might face different
income tax schedules. Thus, a move from joint to individual taxation can change net household
income. The focus of this paper is the consequence of mandatory individual taxation, which is
still under discussion in Luxembourg.
Over the last few decades, the general trend in European Union countries has been a move from
joint towards individual taxation. However, joint taxation, whereby the income of a couple in
a household is fully split between them or aggregated in determining the income tax liability,
is still present in some EU countries, such as Luxembourg, France and Portugal, despite the
fact that this system imposes higher marginal tax rates for the secondary earner in the couple
(Blundell and Macurdy, 1999). As women are generally the secondary earners in a couple
household and as they typically have more elastic labour supply, theory predicts that the
optimal taxation model for labour supply should not impose higher marginal tax rates on
women. Recent work by Bick and Fuchs-Schündeln (2017), which studies the US and 17 EU
countries, finds large disincentive effects of joint taxation on the hours worked of married
women. A shift, therefore, from joint to individual taxation, which would equalize the marginal
tax rates of the primary and secondary earner, is worth considering from the perspective of
increasing female labour supply. It is a policy measure which could also decrease the
administrative burden of the tax authorities, given that modern couples are less likely to marry
and more likely to separate than the previous generation.
Previous empirical research has evaluated the labour market effect of hypothetical shifts from
joint to individual taxation. In Germany, the participation rate of married women can be
expected to increase by up to 6% (Decoster and Haan, 2014). In a comparison of cohabiting
and married couples in France, Kabatek et al. (2014) show that switching to an individual
system of taxation would increase female labour supply by 3.7% and decrease male labour
supply by 0.8%. Evaluating the opposite reform for Italy, Colonna and Marcassa (2015) study
the effect of a hypothetical shift from individual to joint taxation and find a labour supply
response of 5-6 ppt. There have also been studies which have evaluated the incentive effects of
individual taxation in a natural experiment framework. LaLumia (2008) shows that the labour

3
supply of married women decreased by 2ppt after the shift from individual to joint taxation in
the US in 1948. Kaliskova (2014) also found that switching from individual to joint taxation in
Czech Republic in 2005 lead to a decline in the employment rate of married women with
children of about 3 percentage points. Between 1969 and 1975, a 10 ppt increase in the labour
force participation of married women in Sweden has been attributed to the introduction of
individual taxation in 1971 (Selin 2014). A reform which partly individualized the Irish
taxation system increased the participation rate of married women by around 5 ppt (Doorley,
2018). Finally, using a fiscal reform from 1988, which abolished one of the last "joint" features
of the Canadian system, Crossley and Jeon (2007) find an increase in the labour supply of low
educated women of around 10 ppt. This paper builds on these findings.
We use a discrete choice labor supply model to assess the effects of a transition from joint to
individual taxation. We focus on the distributional, labour supply, and the social welfare effect
of introducing a mandatory individual taxation system in Luxembourg. We assume that
household utility depends not simply on income and leisure, but also on participation in the
Social Assistance / Guaranteed Minimum Income (RMG) scheme. We estimate simultaneously
labour supply model and the Guaranteed Minimum Income (RMG)/Social Assistance receipt
taking into account the joint decision of couple labour supply and RMG participation. This
approach offers an alternative way to examine the welfare implications of switching to
individual taxation. That means that the correct model is not a two-dimensional model, labor
time/consumption, but a three-dimensional one in which the third dimension is a zero-one
dimension about the willingness to participate RMG scheme when they are available. We find
that labour supply of married women increases by around 2.27% after the reform. There is no
effect on the labour supply of married men. Equivalised disposable income, after the
behavioural adjustment, decreases on average 2.1 per cent. After adjustments to direct and
indirect taxes, the net revenue-neutral result is a budget surplus for the central government of
around €10 million. The major share of RMG recipient-households have strong preferences for
leisure over work.
The next section describes the institutional background. Section 3 briefly summarize the EU-
SILC data that have been used for labour supply model. The estimated model is presented in
section 4, together with the model prediction and labour supply elasticity. Section 5 discuss the
simulation results of the reform. Section 6 summarizes the main findings of the paper.