Universal, targeted or both: Effects of different child support policies on labour supply and poverty - A simulation study

Bruckmeier, Kerstin; D'Andria, Diego; Wiemers, Jürgen Working Paper Universal, targeted or both: Effects of different child support policies on labour supply and poverty - A simulation study IAB-Discussion Paper, No. 6/2022 Provided in Cooperation with: Institute for Employment Research (IAB) Suggested Citation: Bruckmeier, Kerstin; D'Andria, Diego; Wiemers, Jürgen (2022) : Universal, targeted or both: Effects of different child support policies on labour supply and poverty - A simulation study, IAB-Discussion Paper, No. 6/2022, Institut für Arbeitsmarktund Berufsforschung (IAB), Nürnberg, https://doi.org/10.48720/IAB.DP.2206 This Version is available at: https://hdl.handle.net/10419/262186 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by-sa/4.0/deed.de IAB-DISCUSSION PAPER Articles on labour market issues 06|2022 Universal, targeted or both: Effects of different child support policies on labour supply and poverty - A simulation study Kerstin Bruckmeier, Diego d’Andria, Jürgen Wiemers ISSN 2195-2663 Universal, targeted or both: Effects of different child support policies on labour supply and poverty - A simulation study Kerstin Bruckmeier (IAB), Diego d’Andria (IAB), Jürgen Wiemers (IAB) Mit der Reihe „IAB-Discussion Paper“ will das Forschungsinstitut der Bundesagentur für Arbeit den Dialog mit der externen Wissenschaft intensivieren. Durch die rasche Verbreitung von Forschungsergebnissen über das Internet soll noch vor Drucklegung Kritik angeregt und Qualität gesichert werden. The “IAB-Discussion Paper” is published by the research institute of the German Federal Employment Agency in order to intensify the dialogue with the scientific community. The prompt publication of the latest research results via the internet intends to stimulate criticism and to ensure research quality at an early stage before printing. Contents 1. Introduction ................................................................................ 6 2. Institutional background and related literature ......................................... 8 3. Methodology and data ....................................................................12 4. Policy scenarios ...........................................................................15 5. Results .....................................................................................17 5.1. Labour supply responses.................................................................17 5.2. Effects on the income distribution and on relative poverty ..........................20 5.3. Revenue impact ...........................................................................24 5.4. Reform effects under a balanced budget ...............................................26 6. Conclusions................................................................................30 References .....................................................................................34 A. Appendix: Tables and Figures ............................................................38 A.1. Additional tables and figures for simulation results under partial take-up.........38 A.2. Additional tables and figures for simulation results under full take-up.............42 IAB-Discussion Paper 06|2022 3 Abstract We study a set of hypothetical reforms of child benefits in Germany, using a static tax-benefit microsimulation model augmented with endogenous labour supply and take-up choices (IAB-MSM). We distinguish between a reform of the universal non-means-tested child benefit, a reform of the mean-tested child benefit under the minimum income scheme, and a combination of both. Since the reforms are associated with different fiscal costs, we consider two different budget closures: an increase in the income tax or a consumption tax. The model simulates the impacts of the reforms on household income, poverty and labour supply. We find that improvements in the means-tested child benefit are well-targeted: They provide a high level of poverty reduction with a low fiscal impact at the cost of reduced labour supply incentives for low-income families. When unconditional benefits are increased, the effect on overall income inequality is more pronounced at the cost of reduced labour supply incentives for middleand high-income families. Finally, when combined, the two approaches show synergies, particularly in the form of improved poverty reduction. Zusammenfassung Wir untersuchen hypothetische Reformen der wichtigsten Transferleistungen für Kinder in Deutschland: eine Reform des nicht bedarfsgeprüften Kindergeldes, eine Reform der Kinderregelbedarfe im Rahmen der Grundsicherung und eine Kombination aus beiden Leistungen. Mit Hilfe eines statischen Steuer-Transfer Mikrosimulationsmodells (IAB-MSM), das auch endogene Arbeitsangebotsund Inanspruchnahmeentscheidungen abbildet, untersuchen wir die Auswirkungen der Reformen auf die Einkommensverteilung, relative Armut und das Arbeitsangebot von Eltern. Da die Reformen mit unterschiedlichen fiskalischen Kosten verbunden sind, verwenden wir zwei alternative Ansätze zum Ausgleich des Staatshaushalts: eine Erhöhung der Einkommensteuer oder eine erhöhte Konsumsteuer. Wir zeigen, dass eine Erhöhung der Kinderregelbedarfe zu einer erheblichen Verringerung der relativen Armut bei gleichzeitig relativ geringen fiskalischen Kosten führt. Allerdings ist die Erhöhung der Kinderregelbedarfe mit einem Rückgang des Arbeitsangebots von Familien mit niedrigem Einkommen verbunden. Eine alternative Erhöhung des Kindergeldes wirkt sich vergleichsweise stärker auf den Rückgang der Einkommensungleichheit (gemessen am Gini-Koeffizienten) aus, allerdings auf Kosten geringerer Arbeitsanreize für Familien mit mittlerem und hohem Einkommen. Die Gegenfinanzierung der Reformkosten - insbesondere bei einer progressiv wirkenden Einkommensteuererhöhung - verstärkt den Rückgang der relativen Armut und der Einkommensungleichheit, während gleichzeitig das Arbeitskräfteangebot erheblich reduziert wird. IAB-Discussion Paper 06|2022 4 JEL C15, D31, H53, I38 Keywords child benefits, inequality, labour supply, microsimulation IAB-Discussion Paper 06|2022 5 1. Introduction Recent reform proposals in the European Union (EU) have advanced the idea of a new “child guarantee”. In 2015 the European Parliament asked the European Commission to “introduce a Child Guarantee so that every child in poverty can have access to free healthcare, free education, free childcare, decent housing and adequate nutrition” (Frazer/Guio/Marlier, 2020). Recently, the Employment, Social Policy, Health and Consumer Affairs Council has adopted the proposal on establishing a European Child Guarantee.1 However, while the basic child guarantee emphasises the goal of guaranteeing children access to key services, policy approaches to combat child poverty and social exclusion and their underlying concepts are broad and can take different forms. Among various approaches, child-related income support payments to families represent the classic policy approach to addressing financial poverty and material deprivation among children (Dahl, 2020). Hence, income poverty plays an important role in evaluating the effectiveness of child-related benefit programmes. According to Eurostat and based on data from the EU Statistics on Income and Living Conditions (EU-SILC), 24.2 percent of children in the EU were at risk of poverty or social exclusion in 2020. The level, design and accessibility of child-related benefits have also gained importance in the political debate following the Corona crisis and the temporary adjustments made to social benefit systems in many countries (Baptista et al., 2021). Within the EU, children in Germany have a moderate risk of poverty (Bruckmayer/Picken/Janta, 2020; Frazer/Guio/Marlier, 2020), but within Germany, children have an above-average risk of poverty, mainly due to living in single-parent households (Fertig/Tamm, 2010). Several political parties in Germany denounced that the current minimum income scheme would leave many children out of adequate protection. On top of the insufficient benefit level, the current system is also seen as too complex, which is reflected in low benefit take-up rates (Bruckmeier/Wiemers, 2011b, 2018a), meaning that many households who are entitled to a means-tested child benefit do not claim it. It is therefore not surprising that the idea of a “child guarantee” has also found its way into the German political debate. In its coalition agreement, the German government elected in 2021 sets out the introduction of a basic child allowance to protect children from poverty. In this paper we examine the effects of increasing the level of two existing child-related benefits to families on the income distribution and child poverty in Germany. We distinguish between reforming universal, non-means-tested child benefits, reforming 1See Council of the EU Press release 14 June 2021, https://www.consilium.europa.eu/en/press/press-releases/2021/06/14/ access-to-key-services-for-children-in-need-council-agrees-european-child-guarantee/. IAB-Discussion Paper 06|2022 6 mean-tested child benefits included in the minimum income scheme, and a combination of both. These stylised reform scenarios mimic important characteristics of actual proposals in the current political debate in Germany. In addition to changes in income, we report the fiscal costs of the reforms and present reform effects with and without refinancing. In the case of refinancing, we distinguish between financing the costs of the reform through an income tax increase and through a consumption tax increase. Assessing the fiscal impact of potential benefit reforms may become more important given the new fiscal pressures caused by the Corona crisis in many countries. Our analysis employs the microsimulation model of the Institute for Employment Research, IAB-MSM, to simulate changes in the benefit system. IAB-MSM provides a detailed implementation of the German tax and benefit system and is based on rich survey data from the German Socio-Economic Panel (GSOEP). The GSOEP allows for a characterisation of each individual household (and of its individual members) that includes demographic information, employment status, current and past worked hours, different sources of income, housing conditions and region of residence. Since children in most cases do not earn their own income, their economic situation is determined by the household to which they belong. Moreover, eligibility for means-tested child benefit policies also depends on the household’s economic situation. This close link between the overall economic situation of a household and child benefit policies also implies that the latter may distort the behaviour of other members of the household. The way means-testing, the marginal benefit reduction rate for earned income, and interactions with other social policies are designed may affect parents’ labour supply or consumption patterns in non-trivial ways. Therefore, our approach takes into account potential employment effects of increasing child benefits triggered by a change in labour supply. The IAB-MSM simulates changes in labour supply due to a policy reform based on an econometrically estimated labour supply model. Our model also allows us to take into account and endogenise benefit non-take-up, which is especially important for assessing the effect of an increase in means-tested benefits. Our paper contributes to the literature in several ways. First and foremost, we extend the relatively small amount of research on this topic for Germany and relate it to the most recent reform proposals. In a broader perspective, our paper is also related to studies that focus on the effectiveness of the design of social benefits in reducing poverty, as we provide an in-depth analysis of the two components of a basic child income (unconditional and means-tested benefits) and the different outcomes obtained in terms of income distribution, poverty risk and labour supply. In contrast to many microsimulation studies, we explicitly include labour supply effects in our analysis and thus also contribute to the literature examining the relationship between welfare benefits for children and the labour supply of their parents. The paper is structured as follows. Section 2 presents an overview IAB-Discussion Paper 06|2022 7 of the institutional background and the previous research on child benefits. Section 3 describes the IAB-MSM model and the database. Section 4 introduces the current German policy context, which provides our baseline scenario, and details the three reform scenarios that we simulate in IAB-MSM. Section 5 discusses the main results from our simulations and provides guidance for policy-makers. Section 6 concludes. 2. Institutional background and related literature The motivations behind child support policies are both ethical and economic. From an ethical point of view, children are passive agents, which means they bear the consequences of other people’s decisions. It seems unfair, then, to leave children unprotected from mistakes they could never avoid. An economic motive lies in the concept of “equality of opportunities”: a level playing field for the young implies that allocation of talent works at its best, compared to a situation with strong path-dependencies due to the economic situation of the households of origin, which might trap young people into poverty and low-skill human capital endowments (so much that some authors proposed to interpret children, or parenting, as a public good: Folbre 1994, England/Folbre 1999). Child support policies can help providing children with a fair level and quality of economic support, education and health care, thus at least partly compensating for deficiencies in the household of origin or due to random events that might adversely affect a child’s life. Moreover, as in most OECD countries the participation of women to the workforce is lower than for men, and even lower when looking at participation rates of mothers, child care support policies may bear the effect of reducing barriers preventing mothers to participate in the labour market (see, e.g., Haan/Wrohlich 2011, Bick 2016). In relative terms, Germany fares well when compared to other European member states. According to a publication by the European Commission’s DG Employment (Bruckmayer/Picken/Janta, 2020), Germany ranks 9th among the 28 EU member states in terms of the share of children below 18 years old who are considered at risk of poverty or social exclusion. Other indicators presented in Frazer/Guio/Marlier (2020) comparing the European countries report a relatively robust performance of the German welfare system for children, with low percentages of children in deprivation or receiving inadequate levels of healthcare or education. Similar results were published in Gornick/Nell (2018) using the Luxembourg Income Study Database data for 2010, where Germany was found to have one of the lowest child poverty rates among the 24 highand medium-income surveyed countries. One concern, however, is the high share (the fourth-highest among the EU28 IAB-Discussion Paper 06|2022 8 4. Policy scenarios In recent years, various parties and organisations have put forward proposals for the introduction of a basic child allowance in Germany. Most of the proposals include the following basic features: ● A means-tested component, the level of which is higher than the current level in the basic income system (ALG2) ● An unconditional component, the level of which is equal to the maximum of income tax savings due to the child allowance granted by the income tax. The first point is obviously aimed at combating child poverty, while the second aims at eliminating disparities in child benefits and tax-allowance between highand low-income earners, which are often perceived as unfair. Other features of a basic child benefit that are addressed in many proposals are the take-up of the benefits, the treatment of parents’ earned income, and the integration of the basic child allowance into a system of family support, thus removing it from the basic income system. However, in the absence of a government proposal for the exact design, we analyse the two benefit components based on the existing system: the unconditional standard child benefit (Kindergeld) and the means-tested child benefit included in ALG2 (Kinderregelbedarf ). Our reform scenario (1) increases the Kindergeld to 315 euro for each child, regardless of the number of children in the household and their age. This amount equals the benefit obtained from the tax allowance by a household paying the maximum marginal income tax rate in 2021. Hence, reform scenario (1) effectively substitutes the tax allowance for all households with the increased Kindergeld. The unequal treatment of highand low-income households is thus abolished in this reform scenario, which is one of the core elements of current reform proposals. Our reform scenario (2) addresses political demands to increase the minimum income and simulates an increase in the (Kinderregelbedarf). Because the calculation of the benefits is based on a statistical model using the national income and consumption survey, the calculation method would have to be changed to increase the benefit. We rely on a recalculation of the benefit level for children for the year 2021 by Becker/Held (2021), which is supported by several social welfare associations in Germany. According to this calculation, the ALG2 benefit for children below 6 years would stay constant. Since the goal of improving income for poor families would not be achieved, we increase the amount for children in this age group by the amount that would be achieved with the increased child benefit, which corresponds to 315 euro. For the other age-groups we follow Becker/Held IAB-Discussion Paper 06|2022 15 (2021) and simulate an increase to 385 euro for children between 6 and 13 (up from 309 euro in 2021), and to 444 euro for children between 14 and 17 (up from 373 euro in 2021). Therefore, contrary to reform scenario (1) where Kindergeld provides a fixed amount per child, the benefit in scenario (2) is age-dependent. Under this reform scenario, the general tax allowance is also raised proportionally to the increase in the Kinderregelbedarf to reflect a larger overall value for the minimum existential income. Finally, reform scenario (3) combines the effects of both previous scenarios (1) and (2). Thus, all households receive an increased Kindergeld and, moreover, households under the ALG2 regime benefit from increased child-related grants. The combination of scenario (1) and (2) corresponds to the potential income effects of the proposed basic child allowance: Low-income households with children without further (earned) income are financially better off compared to the status quo. As earned income increases, ALG2 benefits are reduced until the household receives only the unconditional standard child benefit. Table 1 summarises the values used for the Kindergeld and Kinderregelbedarf, per child in the household, in the baseline (status quo) and the three reform scenarios. Table 1.: Summary of the baseline and simulated reform scenarios Reform scenario Baseline (1) (2) (3) Kindergeld per child Eur 219-250 (based on age) Eur 315 Eur 219-250 (based on age) Eur 341 Kinderregelbedarf per child Eur 283-373 (based on age) Eur 283-373 (based on age) Eur 315-444 (based on age) Eur 315-444 (based on age) Source: Own representation. For those households who have income above the maximum eligibility threshold set for the ALG2 benefit but still encounter difficulties in covering all expenses, Germany has alternative benefits, which offer an allowance for living costs (Wohngeld) and a supplementary child allowance (Kinderzuschlag). While our reform scenarios do not change these other policies in a direct way, it is important to stress that several interactions happen anyway. For instance, the law considers the Kindergeld benefit as part of the income calculated for the means testing of the ALG2, which means that increasing Kindergeld (as in our reform scenarios (1) and (3)) also reduces the number of eligible households to ALG2 and may push some households into the Wohngeld + Kinderzuschlag regime. The latter change happens in scenario (1) even though, contrary to scenarios (2) and (3), it does not reform ALG2 at all. Excluding children from the basic income system, as politically demanded, would increase this effect even more and parents could leave ALG2 with even less earned income than under the status quo. How this affects income and labour supply cannot be determined until the benefit-reduction rate in the basic child benefit system is known. Since the benefit reduction rate is in most cases lower if the household receives Wohngeld and Kinderzuschlag instead of ALG2 in the status quo, our simulation results give IAB-Discussion Paper 06|2022 16 an indication of the direction of the labour supply effect.8 This holds if the effective marginal tax rate, resulting from the income tax and the benefit reduction rate inherent in the Wohngeld and the (new) basic child benefit, is below the current marginal tax rate under ALG2, which is also a political goal. In our analysis, we focus mainly on the effect of increasing benefit levels and cannot comment on other possible reform elements, such as a change in the effective marginal tax rate or excluding children from the basic income system. The latter could also have a greater impact on parents’ labour supply if they are no longer subject to the ALG2 activation regime due to the their children’s neediness. 5. Results In this section, we present our simulation results. All the results in Sections 5.1 - 5.3 do not compensate for the budgetary costs of the simulated reforms. In Section 5.4, we will also present and discuss results obtained by reform scenarios that include compensatory tax measures to obtain budget parity. Furthermore, the results in all tables and figures in this section are simulated under the assumption of partial take-up, using the method described in section 3. Selected results under the assumption of full take-up are reported in Appendix A.2 to allow for comparison. 5.1. Labour supply responses Table 2 shows the simulated changes in labour supply for the three reform scenarios. The changes are further disaggregated by weekly working hours and family status. For the increase in the unconditional benefit (Kindergeld) in scenario (1), we find very different effects between men and women in couple households. For women, we observe a shift towards part-time work, i.e., hours categories below 40 hours, resulting in an overall negative effect on labour supply of 23 thousand persons working full-time.9 These results are in line with an ex-ante study of a strong increase in the regular child benefit in Germany in 1996 based on individual survey data by Hener (2016). He finds a strong negative effect on the intensive margin of mothers’ labour supply and estimates that a reallocation towards more part-time work leads to a reduction in mothers’ weekly working hours by 2.3 hours. In Haan/Wrohlich (2011), the authors employ GSOEP data and find that an increase in unconditional child benefits would bear a reduction in (working) women’s labour supply. 8With high earned income, the rate is currently 100 percent in ALG2, which means that an additional euro net wage reduces the benefit by one euro. 9Working full-time is defined as 40 work hours per week. IAB-Discussion Paper 06|2022 17 The picture is the opposite for single parents, mainly women, compared to mothers living with a partner. Scenario (1) leads to a significant labour supply response for lone parents, in that both more mothers take up employment (participation effect) and more work 30 or 40 hours per week. For male partners in couples, we simulate that the increase in Kindergeld leads to a positive effect on total labour supply, although the effect is small at only about 5 thousand full-time equivalents. We also simulate a positive income effect, which leads to a decrease of work in the 50 hours category. However, this does not lead to more part-time work. Instead, the 40 hours category increases, which could indicate the strong preference of male partners for full-time work. The overall impact of scenario (2) is smaller because the reform is limited to a much smaller group. As the non-work income of ALG2 recipients increases, we find negative effects on individual labour supply for all family types. This general pattern of labour supply response to an increase in means-tested benefits was also found in Blömer (2022). The study finds that an increase in means-tested child benefits has a large negative effect on labour supply, which he explains with strong substitution effects that make non-work financially more attractive than work. In scenario (3), which combines scenarios (1) and (2), the negative labour supply effects in scenario (2) outweigh the (overall) small positive effects in scenario (1). Thus, we determine an overall decrease in labour supply of 19 thousand persons working full-time when both benefits are increased. Looking at the results achieved with full take-up of means-tested benefits (see Table A.26), it becomes clear that taking into account endogenous take-up significantly affects some results. In scenario (1), the positive labour supply effects are more pronounced for men in couples and single parents, whereas the effects differ less for women in couple households. Since the additional child benefit in scenario (1) is counted against means-tested benefits the household would receive when not working, working more hours becomes more attractive than not working. Under reform scenario (2), on the other hand, the negative effects are amplified, as the increase in ALG2 benefits increases non-labour income for a larger number of households. This could explain the stronger negative labour supply effects that Blömer (2022) find for an increase in means-tested benefits, as they assume full take-up of these benefits. From an economic perspective, this suggests a conflict in terms of effective income support and minimising negative labour supply responses. Next, we look at the labour supply effects stratified by income deciles based on the baseline income distribution. This differentiation is particularly important with regard to the distributional and poverty effects of the reforms, which can be both amplified and reduced by labour supply reactions. Table 3 shows the participation effect in the upper part and the overall labour supply change in the lower part broken down by income deciles. IAB-Discussion Paper 06|2022 18 Table 2.: Changes in labour supply, unbalanced budget, partial take-up Couples (men) Couples (women) Single parents Total Scenario (1) Part. effect 4.8 -9.1 21.2 17.0 10 hours -3.5 5.0 -5.4 -3.9 15 hours -0.9 7.0 -0.9 5.3 20 hours -3.8 8.6 3.5 8.3 30 hours 3.5 3.1 11.3 17.9 40 hours 25.7 -29.6 12.7 8.7 50 hours -16.2 -3.2 -0.1 -19.4 Full-time equiv. 5.0 -23.1 21.2 3.1 Scenario (2) Part. effect -10.4 -10.8 -8.1 -29.4 10 hours 2.6 -2.2 1.3 1.7 15 hours 0.6 -3.9 0.1 -3.1 20 hours 3.1 -5.5 -0.4 -2.8 30 hours -0.7 -3.5 -4.6 -8.7 40 hours -19.1 2.9 -4.3 -20.5 50 hours 3.0 1.3 -0.2 4.1 Full-time equiv. -13.4 -2.8 -7.8 -24.0 Scenario (3) Part. effect -0.6 -21.6 19.4 -2.8 10 hours 0.4 5.0 -4.8 0.6 15 hours -0.4 6.2 -2.0 3.8 20 hours -3.4 9.1 3.2 8.9 30 hours 4.0 0.0 10.4 14.4 40 hours 17.5 -38.6 12.7 -8.4 50 hours -18.5 -3.4 -0.1 -22.0 Full-time equiv. -4.5 -34.7 20.0 -19.1 NOTE. — Changes in labour supply compared to the baseline. Numbers expressed in 1,000 persons. Part. effect = participation effect (negative change in the 0 hours category). Full-time equiv. = full-time equivalents (change in labour supply expressed in 1,000 persons with a 40 hour working week). Source: IAB-MSM. Scenario (1) has a positive and relevant effect on labour supply in the second decile, both on labour force participation and on the overall effect. For all other deciles we find negative labour supply responses due to an income effect. The first income decile is only mildly affected by all three reform scenarios. Consequently, the rise in labour supply in the second decile alone accounts for the overall positive effect of scenario (1). Moreover, all positive IAB-Discussion Paper 06|2022 19 labour supply reactions found for scenario (1) for men in couple households and single parents come from low-income households in the second decile. This contrasts with the findings from Haan/Wrohlich (2011), who report a negative effect of an increase in the regular child benefit on both employment and worked hours across the entire population. This discrepancy can be explained by the fact that our microsimulation model (in contrast to Haan/Wrohlich 2011) also takes into account that some households who were in the phase-out range of ALG2 entitlements (and who mostly fall into the second income decile), will be pushed out of ALG2 eligibility because the (increased) universal child benefit is completely deducted from entitlements. For households pushed out of ALG2, the increase in the Kindergeld acts like a reduction in the effective marginal tax rate. Studies on the previously mentioned EITC (Whitmore Schanzenbach/Strain, 2021), which provides a tax credit supplement to earnings, also found an increase in the labour supply of single mothers and of couples with children (especially couples with two or more children, due to the policy design being more generous for them than for single-child couples, see Kleven 2019). For scenario (2) we find that the increased benefit obtained under ALG2 decreases labour supply for the beneficiaries (who mostly fall into the first and second income deciles). Furthermore, scenario (2) shows that the increased tax allowance for higher income deciles increases their labour supply. 5.2. Effects on the income distribution and on relative poverty In Figure 1 the bar graphs show the change in equivalised disposable household income before and after accounting for endogenous take-up and labour supply responses. Income changes are presented by income deciles measured for the total population. In reform scenario (1), households from the lower end of the income distribution, except for the second decile, benefit the most, and as household income increases, the relative income gain decreases. Without accounting for behavioural changes, households in the first decile are those benefiting the most, with an increase in the average equivalent household income of around 3 percent. In absolute values (figures are not shown here to economise on space), we find the largest average income change in the third and sixth income deciles. Figure A.11 in Appendix A.1 shows the same results only for the group of households affected by the reforms, i.e., households with at least one child that is entitled to the benefits. The results are comparable to those related to the total population. For lower income groups, the increase in the child benefit often reduces received means-tested benefits, which dampens the income effect in these groups. Since the income benefit from the standard benefit increase outweighs the tax benefit from the child allowance at higher incomes, the income effect is smallest in the tenth decile. Generally the IAB-Discussion Paper 06|2022 20 Table 3.: Changes in labour supply by income decile, unbalanced budget, partial take-up Scenario (1) Scenario (2) Scenario (3) Part. effect Decile 1 -3.5 -6.3 -5.0 Decile 2 41.1 -24.8 28.5 Decile 3 -4.3 -1.2 -6.2 Decile 4 -3.9 0.2 -4.9 Decile 5 -3.4 0.5 -4.2 Decile 6 -4.8 1.4 -6.0 Decile 7 -2.5 0.5 -2.8 Decile 8 -0.9 0.2 -1.0 Decile 9 -0.7 0.1 -0.9 Decile 10 -0.2 0.0 -0.2 Total 17.0 -29.4 -2.8 Full-time equiv. Decile 1 -2.9 -6.1 -5.3 Decile 2 44.0 -26.1 30.7 Decile 3 -4.2 -1.1 -6.6 Decile 4 -4.7 0.2 -5.6 Decile 5 -4.2 2.6 -4.9 Decile 6 -7.0 3.4 -8.5 Decile 7 -6.3 1.9 -6.7 Decile 8 -7.8 0.8 -8.1 Decile 9 -3.3 0.2 -3.5 Decile 10 -0.4 0.1 -0.5 Total 3.1 -24.0 -19.1 NOTE. — Changes in labour supply (sum over all household types) by income deciles compared to the baseline. Numbers expressed in 1,000 persons. Deciles based on net equivalent income. Net equivalent income is calculated with the modified OECD scale. Part. effect = participation effect (negative change in the 0 hours category). Full-time equiv. = full-time equivalents (change in labour supply expressed in 1,000 persons with a 40 hour working week). Source: IAB-MSM. change in unconditional Kindergeld affects all income deciles, with the largest increase in disposable income in absolute terms happening for the middle income deciles. This holds at least up to the sixth decile, after which the changes start to decline (not shown). Scenario (1) aims at better positioning families with children who do not benefit maximally from the child tax allowance due to their low tax burden, thus favouring middle income groups (in absolute values) and lower income groups (in relative values). IAB-Discussion Paper 06|2022 21 Figure 1.: Relative change in equivalised household income by income deciles, partial take-up, unbalanced budget 1 2 3 4 5 6 7 8 9 10 All 0 1 2 3 4 Relative income change (in percent) a) Scenario (1) without behavioural changes with behavioural changes 1 2 3 4 5 6 7 8 9 10 All 0 1 2 3 4 Relative income change (in percent) b) Scenario (2) without behavioural changes with behavioural changes 1 2 3 4 5 6 7 8 9 10 All 0 1 2 3 4 Income deciles Relative income change (in percent) c) Scenario (3) without behavioural changes with behavioural changes NOTE. — Relative change in mean household equivalent net income by income deciles compared to the baseline. The equivalent income is calculated based on the modified OECD scale. Results are calculated assuming partial take-up. Without behavioural change = working hours choice probabilities and the conditional take-up probabilities are fixed at their levels in the baseline scenario. With behavioural change = choice probabilities change according to the predictions of the labour supply model. All = Mean change over deciles. Source: IAB-MSM. IAB-Discussion Paper 06|2022 22 The changes induced by reform scenario (2) are more polarised at the extremes of the income distribution and concentrated on the lowest two deciles, as the benefit is targeted towards the poorest households. We also find small positive income effects for higher income households. The reason for the latter lies in the increased tax allowance for children, which is not means-tested and therefore benefits also the middle and especially higher deciles. Scenario (3) is found to always produce the largest change in disposable income with highest income gains in the third and smallest in the tenth decile. Note that this finding refers to the total population. If we focus on the households with children only, the income effect turns progressive (see figure A.11), as a disproportionately large number of single people live in the first decile and hence do not benefit from the reforms. Behavioural changes, both due to changes in labour supply and take-up, have the strongest effects on the simulated income change in the first two deciles. As discussed in Section 5.1, labour supply changes are the strongest in the second decile in both reform scenarios (1) and (2) and they work toward different directions, increasing household income in scenario (1) and decreasing it in scenario (2). For middle and high income groups behavioural adjustments always reduce income gains mainly produced by reform scenario (1). Looking at the effects of the reforms on relative poverty, Table 4 reports a set of indicators measuring income inequality. These are the Gini coefficient and the poverty rates measured as percentage of households below 60 percent and below 50 percent of the median income (poverty lines), with the corresponding change compared to the baseline. To isolate the effect on the poverty rates, we fix the poverty lines to the median income observed in the baseline. Again, the results refer to the total population. Results are obtained after accounting for labour supply changes induced by the respective reform scenario. Taken together, these indicators suggest that all reforms scenarios simulated here are capable of reducing inequality, but do so in different ways. Compared with scenario (2), reform scenario (1) leads to a greater reduction in inequality, as measured by the Gini coefficient, and to a slightly greater reduction in poverty when defined relative to 60 percent of the median income. This is due to scenario (1)’s non-targeted design: All income deciles are affected by the reform, but especially middle incomes, as shown in the previous Section 5.2. Interestingly, the poverty rate in reform scenario (1) actually slightly increases when using a narrower definition of relative poverty based on 50 percent of median income. Because the poverty line remains unchanged in all reform scenarios, this cannot be explained by an increase in median income through scenario (1). The explanation for this result is found in a simulated change in the take-up of means-tested benefits after an increase in the IAB-Discussion Paper 06|2022 23 non-means-tested Kindergeld. If the Kindergeld increases, the simulated entitlements from other benefit systems, foremost the basic income system for children and their parents, ALG2, are reduced to the same extent. Therefore, some households with low entitlements to ALG2 may choose not to claim these benefits after the reform. Receiving ALG2 includes not only monetary costs but also non-monetary costs, for example due to social stigma. Our model of ALG2 take-up assumes that, for each possible choice of weekly hours of work at which the household is eligible for ALG2, the household compares the net utility of claiming ALG2 (i.e., taking into account the cost of take-up) with the utility in the case of non-take-up, and chooses the take-up option that yields the highest utility. In scenario (1), the utility of the non-take-up option increases while the net utility of the take-up option remains unchanged. In some cases, this leads to households foregoing their ALG2 entitlement, even though this results in a worse monetary position than when receiving the benefit. This explanation is supported by the fact that, when assuming full take-up of means-tested benefits in our model, we do not find an increase in the 50 percent poverty rate, whereas poverty rates are already significantly lower in the baseline (see Table A.25). This result points to a trade-off in terms of poverty reduction when an unconditional, not-means-tested benefit does not meet the minimum income and households have to apply for additional means-tested benefits. The effect of both reform stages, shown by reform scenario (3), on the 60 percent poverty rate is strong, with a decrease of 4.7 percent or almost 1 percentage point. The poverty-reducing effects of both scenarios (1) and (2) are amplified to a small extent when combined, which could also be due to the different labour supply responses to scenarios (1) and (2). It is also to note that although reforms (1) and (2) both achieve a reduction in poverty rates (as said, when looking at rates based on 60 percent of median income), scenario (1) is much more costly for the public budget. As shown in the next section, the revenue cost of reform scenario (1) is indeed about four times larger than scenario (2). Again, if we do not focus on the entire income distribution but only on households with children, the poverty-reducing effects of the reform scenarios are more pronounced. For example, reform (3) simulates a decline in the 60 percent poverty rate of families of 14.2 percent (see Table A.11 in the Appendix). 5.3. Revenue impact Table 5 reports the changes for the government budget, broken down in such a way that the first set of rows (Panel A) reports changes compared to baseline values without taking behavioural effects into account. The second set of rows (Panel B) also takes behavioural IAB-Discussion Paper 06|2022 24 Table 7.: Changes in labour supply by income decile, balanced budget (through consumption tax adjustment), partial take-up Scenario (1) Scenario (2) Scenario (3) Part. effect Decile 1 -18.5 -10.4 -28.1 Decile 2 29.7 -27.9 11.3 Decile 3 -6.5 -1.9 -9.7 Decile 4 -5.5 -0.2 -7.3 Decile 5 -5.3 0.2 -6.6 Decile 6 -6.5 1.1 -8.6 Decile 7 -3.6 0.3 -4.4 Decile 8 -1.5 0.1 -2.1 Decile 9 -1.3 -0.1 -1.9 Decile 10 -0.6 -0.1 -0.9 Total -19.6 -38.8 -58.2 Full-time equiv. Decile 1 -18.6 -10.4 -28.4 Decile 2 31.3 -29.4 11.7 Decile 3 -7.4 -2.0 -11.7 Decile 4 -7.3 -0.4 -9.5 Decile 5 -7.8 1.9 -9.5 Decile 6 -10.1 2.8 -13.0 Decile 7 -8.7 1.4 -10.8 Decile 8 -10.4 0.1 -12.0 Decile 9 -5.8 -0.3 -7.1 Decile 10 -2.1 -0.2 -3.1 Total -46.9 -36.6 -93.3 NOTE. — Changes in labour supply (sum over all household types) by income deciles compared to the baseline. Numbers expressed in 1,000 persons. Deciles based on net equivalent income. Net equivalent income is calculated with the modified OECD scale. Part. effect = participation effect (negative change in the 0 hours category). Full-time equiv. = full-time equivalents (change in labour supply expressed in 1,000 persons with a 40 hour working week). Source: IAB-MSM. pronounced. However, with significantly higher fiscal costs, a clear poverty-reducing effect can be achieved as well. Since an increase in unconditional benefits imposes very high fiscal costs, refinancing through an income tax amplifies the achieved redistributive effects. Concerning non-take-up of means-tested benefits, the results point to a trade-off in terms of poverty reduction when an unconditional, not means-tested benefit does not reach the minimum income and households have to apply for additional means-tested benefits. After increasing unconditional benefits, some households may choose not to claim means-tested IAB-Discussion Paper 06|2022 31 Table 8.: Changes in labour supply by income decile, balanced budget (through income tax adjustment), partial take-up Scenario (1) Scenario (2) Scenario (3) Part. effect Decile 1 -17.1 -9.4 -23.6 Decile 2 27.4 -28.5 9.2 Decile 3 -8.9 -2.6 -13.1 Decile 4 -8.2 -0.9 -11.4 Decile 5 -7.9 -0.6 -10.3 Decile 6 -10.7 0.3 -15.5 Decile 7 -5.5 -0.2 -7.7 Decile 8 -3.2 -0.2 -5.0 Decile 9 -3.3 -0.3 -5.1 Decile 10 -2.4 -0.4 -4.1 Total -40.0 -42.8 -86.5 Full-time equiv. Decile 1 -20.3 -9.7 -28.3 Decile 2 26.6 -30.8 5.6 Decile 3 -12.4 -3.1 -18.1 Decile 4 -12.5 -1.8 -17.2 Decile 5 -14.7 0.1 -20.1 Decile 6 -18.3 1.0 -25.2 Decile 7 -16.2 -0.0 -21.8 Decile 8 -17.3 -2.0 -22.1 Decile 9 -12.6 -2.1 -16.7 Decile 10 -10.0 -1.9 -16.1 Total -107.7 -50.1 -180.0 NOTE. — Changes in labour supply (sum over all household types) by income deciles compared to the baseline. Numbers expressed in 1,000 persons. Deciles based on net equivalent income. Net equivalent income is calculated with the modified OECD scale. Part. effect = participation effect (negative change in the 0 hours category). Full-time equiv. = full-time equivalents (change in labour supply expressed in 1,000 persons with a 40 hour working week). Source: IAB-MSM. benefits, although they would lose income. One approach to avoid this would be to simultaneously take measures to increase the use of means-tested benefits. As far as labour supply effects are concerned (which may be a major concern, to the extent that entry into the labour force implies long-term improvements in the conditions of poorer households), our findings suggest that raising the non-means-tested child benefit (Kindergeld) would improve labour force participation among low-income households, especially among single parents, while having a negative impact on labour supply for IAB-Discussion Paper 06|2022 32 middleand high-income households. The negative effects of an increase in the non-means-tested benefit on labour supply are substantially amplified by the inclusion of refinancing, especially when an increase in the income tax is considered. On the contrary, we find that reforms based on means-tested benefits tend to depress labour supply for the lowest two income deciles. The negative effects are amplified when full take-up of means-tested benefits is assumed, indicating a conflict in terms of effectiveness of income support through low non-take-up and minimising negative labour supply responses. Interestingly, when combined, the two approaches show synergies in the form of a lower negative impact on labour supply and improved poverty reduction. Non-means-tested benefits then have the same effect as a reduction in the effective marginal tax rates for recipients of means-tested benefits, as non-means-tested benefits are considered in the means test. This result suggests that a more efficient policy design might combine a reform of both, means-tested and non-means-tested benefits for children in order to achieve minimal labour supply reductions and more targeted poverty relief. Alternatively, an increase in means-tested benefits could be combined with a reduction in the transfer withdrawal rate. With respect to a non-means-tested benefit equal to the subsistence level of children (universal minimum income), our results indicate that this would be associated with very high costs, strong negative labour supply effects, but also significant poverty reduction. IAB-Discussion Paper 06|2022 33 References Arntz, Melanie; Clauss, Markus; Kraus, Margit; Schnabel, Reinhold; Spermann, Alexander; Wiemers, Jürgen (2007): Arbeitsangebotseffekte und Verteilungswirkungen der Hartz-IV-Reform. IAB Forschungsbericht 10/2007. Baptista, Isabel; Marlier, Eric; Spasova, Slavina; Peña-Casas, Ramon; Fronteddu, Boris; Ghailani, Dalila; Sabato, Sebastiano; Regazzoni, Pietro (2021): Social protection and inclusion policy responses to the COVID-19 crisis. An analysis of policies in 35 countries. European Social Policy Network (ESPN), Luxembourg: Publications Office of the European Union. Becker, Irene; Hauser, Richard (2012): Kindergrundsicherung, Kindergeld und Kinderzuschlag: Eine vergleichende Analyse aktueller Reformvorschläge. WSI-Diskussionspapier, No. 180, Hans-Böckler-Stiftung, Wirtschaftsund Sozialwissenschaftliches Institut (WSI), Düsseldorf. Becker, Irene; Held, Benjamin (2021): Regelbedarfsbemessung - eine Alternative zum gesetzlichen Verfahren. Diakonie Deutschland. Bick, Alexander (2016): The quantitative role of child care for female labor force participation and fertility. In: Journal of the European Economic Association, Vol. 14, No. 3, p. 639–668. Blos, Kerstin; Feil, Michael; Rudolph, Helmut; Walwei, Ulrich; Wiemers, Jürgen (2007): Förderung Existenz sichernder Beschäftigung im Niedriglohnbereich – Schätzung von Angebots-, Verteilungsund fiskalischen Effekten des SMWA-Vorschlags. IAB-Forschungsbericht, 07/2007, 78 S., Nürnberg. Blömer, Maximilian Joseph (2022): Wie wirkt das Teilhabegeld und was kostet es? Simulationsrechnungen für ein Kindergrundsicherungsmodell Studie mit Tabellenband. ifo Forschungsberichte / 130. Blömer, Maximilian Joseph; Litsche, Simon; Peichl, Andreas (2021): Gutachten zum Reformvorschlag Kindergrundsicherung. ifo Forschungsberichte / 124. Bonin, Holger; Reuss, Karsten; Stichnoth, Holger (2016): The Monetary Value of Family Policy Measures in Germany over the Life Cycle: Evidence from a Dynamic Microsimulation Model. In: CESifo Economic Studies, p. ifw014. Brewer, Mike; Duncan, Alan; Shephard, Andrew; Suarez, María José (2006): Did working families’ tax credit work? The impact of in-work support on labour supply in Great Britain. In: Labour economics, Vol. 13, No. 6, p. 699–720. IAB-Discussion Paper 06|2022 34 Bruckmayer, Michaela; Picken, Natalie; Janta, Barbara (2020): Developments in Child and Family Policy in the EU in 2019. Tech. Rep., European Commission. Bruckmeier, Kerstin; Wiemers, Jürgen (2018a): Benefit take-up and labor supply incentives of interdependent means-tested benefit programs for low-income households. In: Comparative Economic Studies, Vol. 60, No. 4, p. 583–604. Bruckmeier, Kerstin; Wiemers, Jürgen (2011a): A New Targeting – A New Take-Up? Non-Take-Up of Social Assistance in Germany after Social Policy Reforms. IAB Discussion Paper No. 10/2011, Institute for Employment Research, Nuremberg. Bruckmeier, Kerstin; Wiemers, Jürgen (2011b): A new targeting: a new take-up? In: Empirical Economics, p. 1–16, URL http://link.springer.com/article/10.1007%2Fs00181-011-0505-9. Bruckmeier, Kerstin; Wiemers, Jürgen (2018b): Benefit take-up and labor supply incentives of interdependent means-tested benefit programs for low-income households. In: Comparative Economic Studies, Vol. 60(4), p. 583–604. Christl, Michael; De Poli, Silvia; Varga, Janos (2022): Reducing the income tax burden for households with children. In: Fiscal Studies, p. forthcoming. Chzhen, Yekaterina (2017): Unemployment, social protection spending and child poverty in the European Union during the Great Recession. In: Journal of European Social Policy, Vol. 27, No. 2, p. 123–137. Dahl, Mary (2020): Children and their Rights and Entitlements in EU Welfare States. In: Journal of Social Policy, Vol. 49, No. 2, p. 343–360. England, Paula; Folbre, Nancy (1999): Who should pay for the kids? In: The Annals of the American Academy of Political and Social Science, Vol. 563, No. 1, p. 194–207. Eurofound (2015): Access to social benefits: Reducing non-take-up. Publications Office of the European Union, Luxembourg. Ferragina, Emanuel; Seeleib-Kaiser, Martin (2014): Determinants of a Silent (R)evolution: Understanding the Expansion of Family Policy in Rich OECD Countries. In: Social Politics, Vol. 22, No. 1, p. 1–37. Fertig, Michael; Tamm, Marcus (2010): Always poor or never poor and nothing in between? Duration of child poverty in Germany. In: German Economic Review, Vol. 11, No. 2, p. 150–168. Figari, Francesco; Paulus, Alari; Sutherland, Holly (2009): Measuring the size and impact of public cash support for children in cross-national perspective. EUROMOD Working Papers EM6/09, EUROMOD at the Institute for Social and Economic Research, URL https://EconPapers.repec.org/RePEc:ese:emodwp:em6-09. IAB-Discussion Paper 06|2022 35 Folbre, Nancy (1994): Children as public goods. In: The American Economic Review, Vol. 84, No. 2, p. 86–90. Frazer, Hugh; Guio, Anne-Catherine; Marlier, Eric (2020): Feasibility Study for a Child Guarantee: Final Report, Feasibility Study for a Child Guarantee (FSCG). Gornick, Janet C; Nell, Emily (2018): Children, poverty and public policy: A cross-national perspective. In: Handbook of family policy, Edward Elgar Publishing. Haan, Peter; Wrohlich, Katharina (2011): Can child care policy encourage employment and fertility?: Evidence from a structural model. In: Labour Economics, Vol. 18, No. 4, p. 498–512. Haisken-DeNew, John P; Frick, Joachim R (2005): Desktop Companion to the German Socio-Economic Panel (SOEP): Version 8.0. In: Deutsches Institut für Wirtschaftsforschung (DIW), Berlin. Hener, Timo (2016): Unconditional Child Benefits, Mothers Labor Supply, and Family Well-Being: Evidence from a Policy Reform. In: CESifo Economic Studies, Vol. 62, No. 5, p. 624–649. Jacobebbinghaus, Peter; Steiner, Viktor (2003): Dokumentation des Steuer-Transfer-Mikrosimulationsmodells STSM: Version 1995-1999. Tech. Rep., ZEW-Dokumentation. Kleven, Henrik (2019): The EITC and the extensive margin: A reappraisal. Tech. Rep., National Bureau of Economic Research. Levy, Horacio; Matsaganis, Manos; Sutherland, Holly (2013): Towards a European union child basic income?: Within and between country effects. In: International Journal of Microsimulation, Vol. 6, No. 1, p. 63–85. Lietzmann, Torsten; Wenzig, Claudia (2020): Materielle Unterversorgung von Kindern. Magda, Iga; Kieczewska, Aneta; Brandt, Nicola (2020): The Effects of Large Universal Child Benefits on Female Labour Supply. In: IZA Journal of Labor Policy, p. 10–17. Meyer, Bruce D; Rosenbaum, Dan T (2001): Welfare, the earned income tax credit, and the labor supply of single mothers. In: The quarterly journal of economics, Vol. 116, No. 3, p. 1063–1114. Nygård, Mikael; Lindberg, Marja; Nyqvist, Fredrica; Härtull, Camilla (2019): The Role of Cash Benefit and In-Kind Benefit Spending for Child Poverty in Times of Austerity: An Analysis of 22 European Countries 20062015. In: Social Indicators Research, Vol. 146, p. 533–552. IAB-Discussion Paper 06|2022 36 Popova, Daria (2016): Distributional impacts of cash allowances for children: A microsimulation analysis for Russia and Europe. In: Journal of European Social Policy, Vol. 26, No. 3, p. 248–267. Rainer, Helmut; Bauernschuster, Stefan; Danzer, Natalia; Fichtl, Anita; Hener, Timo; Holzner, Christian; Reinkowski, Janina (2013): Kindergeld und Kinderfreibeträge in Deutschland: Evaluierung der Auswirkungen auf familienpolitische Ziele. In: ifo Schnelldienst, Vol. 66, No. 9, p. 28–36. Salanauskaite, Lina; Verbist, Gerlinde (2013): Is the neighbour’s grass greener? Comparing family support in Lithuania and four other New Member States. In: Journal of European Social Policy, Vol. 23, No. 3, p. 315–351. Schirle, Tammy (2020): The effect of universal child benefits on labour supply. In: Canadian Journal of Economics/Revue canadienne d’économique, Vol. 48, No. 2, p. 437–463. Urban, Ivica; Pezer, Martina (2018): Microsimulation of child benefits: a review of studies. In: International Journal of Microsimulation, Vol. 11, No. 3, p. 134–169. Van Lancker, Wim; Van Mechelen, Natascha (2015): Universalism under siege? Exploring the association between targeting, child benefits and child poverty across 26 countries. In: Social Science Research, Vol. 60-75. van Soest, Arthur (1995): Structural Models of Family Labor Supply - A Discrete Choice Approach. In: Journal of Human Resources, Vol. 30, p. 63–88. Wagner, Gert G; Frick, Joachim R; Schupp, Jürgen (2007): The German socio-economic panel study (SOEP): Scope, evolution and enhancements. Tech. Rep., SOEPpapers on Multidisciplinary Panel Data Research. Wang, Julia Shu-Huah (2021): State TANF Time Limit and Work Sanction Stringencies and Long-Term Trajectories of Welfare Use, Labor Supply, and Income. In: Journal of Family and Economic Issues, Vol. 42, p. 650–696. Whitmore Schanzenbach, Diane; Strain, Michael R (2021): Employment effects of the Earned Income Tax Credit: Taking the long view. In: Tax Policy and the Economy, Vol. 35, No. 1, p. 87–129. Wiemers, Jürgen (2015): Endogenizing take-up of social assistance in a microsimulation model – A case study for Germany. In: The International Journal of Microsimulation, Vol. 8, No. 2, p. 4–27. Wiemers, Jürgen; Bruckmeier, Kerstin (2009): Forecasting Behavioural and Distributional Effects of the Bofinger-Walwei Model Using Microsimulation. In: Journal of Economics and Statistics, Vol. 229, No. 4, p. 492–511. IAB-Discussion Paper 06|2022 37 A. Appendix: Tables and Figures A.1. Additional tables and figures for simulation results under partial take-up Table A.11.: Inequality and poverty indicators, affected population only, partial take-up Reform scenarios Baseline (1) (2) (3) Gini coefficient Level (in percent) 28.30 27.51 28.12 27.21 Abs. diff (in pp) . -0.79 -0.18 -1.09 Rel. diff (in percent) . -2.79 -0.63 -3.86 Poverty rate (60%) Level (in percent) 16.00 15.03 15.29 13.73 Abs. diff (in pp) . -0.97 -0.70 -2.27 Rel. diff (in percent) . -6.04 -4.40 -14.19 Poverty rate (50%) Level (in percent) 9.29 9.53 8.43 8.16 Abs. diff (in pp) . 0.24 -0.86 -1.13 Rel. diff (in percent) . 2.63 -9.28 -12.20 NOTE. — Absolute and relative differences in Gini coefficient and poverty rates compared to the baseline. Poverty rates are relative to 60% and 50% of median net equivalent income. The net equivalent income is calculated based on the modified OECD scale. pp = percentage points. Results refer to the population living in households affected by the reform scenarios, i.e., families having at least one child entitled to Kindergeld or ALG2. Source: IAB-MSM. IAB-Discussion Paper 06|2022 38 Figure A.11.: Relative change in equivalised household income by income deciles, affected population only, partial take-up, unbalanced budget 1 2 3 4 5 6 7 8 9 10 All 0 2 4 6 8 Relative income change (in percent) a) Scenario (1) without behavioural changes with behavioural changes 1 2 3 4 5 6 7 8 9 10 All 0 2 4 6 8 Relative income change (in percent) b) Scenario (2) without behavioural changes with behavioural changes 1 2 3 4 5 6 7 8 9 10 All 0 2 4 6 8 Income deciles Relative income change (in percent) c) Scenario (3) without behavioural changes with behavioural changes NOTE. — Relative change in mean household equivalent net income by income deciles compared to the baseline. The equivalent income is calculated based on the modified OECD scale. Results are calculated assuming partial take-up. Without behavioural change = working hours choice probabilities and the conditional take-up probabilities are fixed at their levels in the baseline scenario. With behavioural change = choice probabilities change according to the predictions of the labour supply model. All = Mean change over deciles. Results refer to the population living in households affected by the reform scenarios, i.e., families having at least one child entitled to Kindergeld or ALG2. Source: IAB-MSM. IAB-Discussion Paper 06|2022 39 Table A.12.: Change in number of households (in 1,000), unbalanced budget, partial take-up Reform scenarios (1) (2) (3) (A) Changes before labour supply effects ALG2 (living costs) -56 24 -28 ALG2 (accommodation costs) -37 45 -47 Housing benefits 9 -38 -9 Supplementary child allowance 37 -42 23 (B) Changes after labour supply effects ALG2 (living costs) -171 87 -134 ALG2 (accommodation costs) -161 107 -162 Housing benefits 26 -46 4 Supplementary child allowance 55 -55 34 (C) Labour supply effect (B) - (A) ALG2 (living costs) -115 62 -106 ALG2 (accommodation costs) -124 63 -116 Housing benefits 17 -9 13 Supplementary child allowance 18 -13 11 NOTE. — Differences in households (in 1.000) receiving a positive amount of each item in the table compared to the baseline. ALG2 = unemployment benefit II (SGB II). Source: IAB-MSM. Table A.13.: Change in number of children (in 1,000), unbalanced budget, partial take-up Reform scenarios (1) (2) (3) (A) Changes before labour supply effects ALG2 (living costs) -105 53 -60 ALG2 (accommodation costs) -86 98 -50 Housing benefits 12 -80 -33 Supplementary child allowance 86 -95 51 (B) Changes after labour supply effects ALG2 (living costs) -300 192 -227 ALG2 (accommodation costs) -296 236 -233 Housing benefits 35 -96 -18 Supplementary child allowance 114 -125 58 (C) Labour supply effect (B) - (A) ALG2 (living costs) -195 139 -167 ALG2 (accommodation costs) -209 138 -182 Housing benefits 23 -16 15 Supplementary child allowance 28 -30 8 NOTE. — Differences in children (in 1.000) who live in households which receive a positive amount of each item in the table compared to the baseline. ALG2 = unemployment benefit II (SGB II). Source: IAB-MSM. IAB-Discussion Paper 06|2022 40 Table A.211.: Inequality and poverty indicators, affected population only, full take-up Reform scenarios Baseline (1) (2) (3) Gini coefficient Level (in percent) 26.95 26.05 26.83 25.75 Abs. diff (in pp) . -0.89 -0.11 -1.19 Rel. diff (in percent) . -3.32 -0.43 -4.43 Poverty rate (60%) Level (in percent) 15.01 12.35 14.06 11.18 Abs. diff (in pp) . -2.66 -0.95 -3.83 Rel. diff (in percent) . -17.71 -6.32 -25.55 Poverty rate (50%) Level (in percent) 5.79 4.01 4.40 2.89 Abs. diff (in pp) . -1.78 -1.39 -2.90 Rel. diff (in percent) . -30.68 -23.96 -50.02 NOTE. — Absolute and relative differences in Gini coefficient and poverty rates compared to the baseline. Poverty rates are relative to 60% and 50% of median net equivalent income. The net equivalent income is calculated based on the modified OECD scale. pp = percentage points. Results refer to the population living in households affected by the reform scenarios, i.e., families having at least one child entitled to Kindergeld or ALG2. Source: IAB-MSM. IAB-Discussion Paper 06|2022 47 List of Figures Figure 1: Relative change in equivalised household income by income deciles, partial take-up, unbalanced budget ................................................22 Figure A.11: Relative change in equivalised household income by income deciles, affected population only, partial take-up, unbalanced budget..................39 List of Tables Table 1: Summary of the baseline and simulated reform scenarios......................16 Table A.11: Inequality and poverty indicators, affected population only, partial take-up.38 Table A.12: Change in number of households (in 1,000), unbalanced budget, partial Table A.13: Change in number of children (in 1,000), unbalanced budget, partial take-up40 Table A.14: Fiscal effect (in million Euro) under different balanced budget Table A.27: Changes in labour supply by income decile, unbalanced budget, full take-Up44 Table A.29: Change in number of households (in 1,000), unbalanced budget, full take-up46 Table 2: Changes in labour supply, unbalanced budget, partial take-up ................19 Table 3: Changes in labour supply by income decile, unbalanced budget, partial take-up ................................................................................21 Table 4: Inequality and poverty indicators, partial take-up ...............................25 Table 5: Fiscal effect (in million Euro), unbalanced budget, partial take-up ............27 Table 6: Inequality and poverty indicators under different budget regimes, partial take-up ................................................................................29 Table 7: Changes in labour supply by income decile, balanced budget (through consumption tax adjustment), partial take-up....................................31 Table 8: Changes in labour supply by income decile, balanced budget (through income tax adjustment), partial take-up ..........................................32 take-up ................................................................................40 assumptions, partial take-up .......................................................41 Table A.25: Inequality and poverty indicators, full take-up ...................................42 Table A.26: Changes in labour supply, unbalanced budget, full take-Up ....................43 Table A.28: Fiscal effect (in million Euro), unbalanced budget, full take-up ................45 Table A.210: Change in number of children (in 1,000), unbalanced budget, full take-up ...46 Table A.211: Inequality and poverty indicators, affected population only, full take-up.....47 IAB-Discussion Paper 06|2022 48 Imprint IAB-Discussion Paper 06|2022 Publication Date 25.04.2022 Publisher Institute for Employment Research of the Federal Employment Agency Regensburger StraSSe 104 90478 Nürnberg Germany Rights of use This publication is published under the following Creative Commons licence: Attribution - ShareAlike 4.0 International (CC BY-SA 4.0) https://creativecommons.org/licenses/by-sa/4.0/deed.de Download https://doku.iab.de/discussionpapers/2022/dp0622.pdf All publications in the series “IAB-Discusssion Paper” can be downloaded from https://www.iab.de/en/publikationen/discussionpaper.aspx Website www.iab.de/en ISSN 2195-2663 DOI 10.48720/IAB.DP.2206 Corresponding author Kerstin Bruckmeier +49 (911) 179 4432 E-Mail [email protected] Diego d’Andria +49 (911) 179 7909 E-Mail [email protected] Jürgen Wiemers +49 (911) 179 8671 E-Mail Juergen.[email protected]