Vous consultezProgressivity of Childcare Tax Policies in Belgium

AuteursMaria-Isabel Farfan-Portet[*] [*] Health Sociology and Economics, School of Public Health,...
suitedu même auteur

1 - Introduction

Until 1988, supply-side subsidies for childcare were provided at a community level and were the only ones available in Belgium. They were distributed among regulated institutions to cover 70% to 80% of the running costs. In 1988, the federal government introduced the tax deduction of childcare expenses. This reform was an important change in the orientations of childcare since it corresponds to the introduction of demand-side policies. Families using formal childcare, whether subsidised or not, could deduct 80% of their care expenses subject to a limit of €8.70 per day and families not claiming the tax deduction and having dependents under 3 years old were automatically granted a tax credit of 272EUR (Loi portant sur la réforme de l’impôt des revenus et modification des taxes assimilées au timbre, 16-12-1988). Tax credits in Belgium are set as a portion of income that is not taxed, and the reduction on the tax payments is computed by using the tax rates on the total amount of the tax credit.

2 In 2000 and 2001, the federal government increased the limit on maximum qualified expenses and the amount of the tax credit. The daily limit was raised to 11.2EUR per day in 2000 and the percentage of deductible childcare expenses changed from 80% to 100% in 2001. The tax credit was increased to 297EUR in 1995 and to 322EUR in 2000. The tax deduction and the tax credit are non-refundable and so only families with positive tax liabilities can benefit from this subsidy

3 The tax relief of childcare expenses has been criticized as regressive given that take-up rates are much higher among middle-and upper-income families (Robins, 1990). Furthermore, given that the tax benefit depends on the marginal tax rate, families with higher income obtain greater fiscal benefit from the deduction of childcare expenses. Nevertheless, Gentry and Hagy (1995) point out that the progressivity of the tax relief does not only depend on take-up rates but also on features of the tax relief that determine the distribution of benefits. Indeed, the tax relief can be progressive if deductible care expenses are: (1) limited; (2) negatively related to the family’s income; (3) and refundable (Gentry and Hagy 1995).

4 The literature based on the US experience, shows that tax relief of childcare expenses is progressive at least for some parts of the income distribution. Nevertheless, Gentry and Hagy (1995) mention that the progressivity of the tax relief might vary if measured among different groups of households. Dunbar and Nordhauser (1991) found that the childcare tax credit increases the overall progressivity of the tax system when considering all families with and without dependents (Dumbar and Nordhauser, 2002). Yet, Altshuler and Schwartz (1996) found that for families with dependents the tax deduction is regressive in the first part of the income distribution, while among families claiming the deduction, it is progressive for the whole income distribution (Altschuler R. and Schartz A., 1996). Gentry and Hagy’s results are similar to those of Altshuler and Schwartz (1996). Indeed, among a sample of families with dependents, the tax relief is regressive through the lowest quintile of the income distribution while being progressive throughout the remainder of the income distribution (Gentry and Hagy 1995).

5 The purpose of this work is to analyze the evolution of the progessivity of the tax deduction and the tax credit for childcare in Belgium from 1994 to 2001. We evaluate the progressivity of tax relief not only among the families who can be targeted by these policy instruments (all families with dependents), but also among the families who claim the deduction of childcare expenses. This allows us to examine both the effective use and the distribution of benefits from the tax deduction. Based on Gentry’s and Hagy’s work (1995) we might expect that the tax deduction of childcare expenses is not progressive in Belgium. Indeed, deductible childcare expenses are neither negatively related to the family’s income nor refundable. Furthermore, the limit on maximum qualified expenses was increase in 2000, benefiting families paying more than 8.7EUR a day, thereby excluding low-income families who pay less than this amount.

6 The paper is organized in five sections. Section 2 describes the data used in our work. Section 3 presents the methodology. The results are presented in Section 4. Section 5 concludes the paper.

2 - Data

7 The data used is a sample of the overall tax claims of the households having dependents aged 0 to 3 years olds. The database (IPCAL) is provided by the National Institute of Statistics (INS) for the 1994-2001 fiscal years, corresponding to the earned incomes of 1993-2000. A few observations had to be deleted from the file because of invalid information. The numbers of tax declarations before and after validation are provided in the following table.

Number of observations

Table 1 - 

8 The fiscal database contains information about yearly net and gross income, family structure, yearly deductible childcare expenses, and yearly tax payments. It must be noted that no information is available neither on real expenditures, the effective utilization of day care nor the parents’ working hours. Nevertheless, available information on the yearly deductible childcare expenses allows us to identify the families using formal childcare. Indeed, only families paying for care services from registered childcare providers can deduct their care expenses.

9 It is important to note that even if the tax deduction of childcare expenses and the tax credit were set at the same time, they usually are set to attain different policy objectives. The tax deduction is set to encourage the participation in the labour market, while the tax credit is a general subsidy generally targeting poorer families. In the case of Belgium, the tax deduction was clearly set to reduce the costs of formal childcare in order to allow mothers to work, while the tax credit was the outcome of a political compromise. Indeed, the tax credit was set to reduce the tension among political groups to the implementation of the tax deduction. Both measures (tax deduction and tax credit) should help families to reduce the cost of raising young dependents, independently of the type of care used.

2.1 - Evolution of tax relief for childcare

10 The percentage of households with young dependents using the deduction of childcare expenses has increased from 27,3% in 1994 to 41,8% in 2001, which has increased the government’s revenue loss on childcare subsidies from 38,6 million euro in 1994 to 61,2 million euro in 2001. This loss represents 1,7% of total tax payments from families with young dependents in 1994 and 2,6% in 2001. The government’s loss from granting the tax credit is smaller than from the tax deduction of childcare expenses. In 1994 it accounted for 27,6 million euros while in 2001 it was only of 17,1 million euros. The reduction on the revenue loss from the tax credit is due to the decline in the number of families that receive the credit. Total revenue loss for the tax policies for childcare accounted for 66,2 million euros in 1994 and for 78,3 million euros in 2001. In the U.S. revenue loss from the Child Care Tax Credit was of 2.8 billion dollars in 1.999 (Blau 2000).

Percentage of households with young dependents claiming the tax deduction for childcare

Figure 1 - 

11 Although there has been a steady increase of the percentage of households claiming the tax deduction between 1994 and 1999, the increase between 1999 and 2000 mostly doubled in comparison with the annual rate of 1994-99: respectively 10.8% and 5.7%. This result is associated to the 2000’s reform that increased the deductible limit from 8.7EUR to 11.2EUR per day. Indeed by further reducing the cost of childcare, families who were using either parental care or informal care, may choose to pay for formal childcare. This result points out that tax relief can also be use to encourage the use formal care. It is beyond the scope of this paper to analyze how tax relief is likely to influence families’ choice of childcare, nevertheless it is important to mention that there is a wide debate related to the efficiency of using tax relief as means to improve the type and quality of care purchased by parents (Blau 2000; Blau and Hagy, 1998; Organization for Economic Co-operation and Development, 2001).

2.2 - Employment and the tax deduction

12 Figure 2 shows the percentage of households claiming the tax deduction with respect to the employment status of the parents (taxpayers). Since 1994, the percentage of working parents claiming the tax deduction has increased. For two-parent households the percentage of families deducting their childcare expenses when both parents are working increased from 36,5% in 1994 to 53,6% in 2001. For one-parent households, the percentage of families with young dependents deducting their childcare expenses increased from 26,9% in 1994 to 39,2% in 2001. Our results show that lone parents who are employed are less likely to claim the tax deduction than two parent households (when both parents work). Several hypotheses can explain this result. Firstly, lone parents might be more prone to use informal care from family and friends. Indeed, family and friends can provide care arrangements that can be less rigid than formal childcare (for instance caring longer hours for children) enabling them to better cope with their employment and caring responsibilities. Secondly, lone parents might find it difficult to cover the costs associated with formal childcare, which might lead them to use more informal care.Percentage of households claiming the tax deduction by employment status of the parents

Figure 2 - 

Previous works have shown that reducing the cost of childcare might change household’s decisions on the use of childcare and labour supply, in particular for low-income mothers (Baum Ch., 2002; Blau and Robins, 1988; Connelly, 1992; Gustafsson and Stafford, 1992; Michalopoulos, Robins and others, 1992; Ribar, 1992). As a consequence, we might expect that the increasing percentage of families claiming the deduction for childcare expenses is associated with a higher participation of parents in the labour market. Although it is beyond the purpose of this paper to analyze changes in labor supply and use of childcare, it is important to mention that these changes might affect the distributional impact of tax relief for childcare. For instance, the tax relief can induce low-income parents depending on the welfare system to work. This can affect the income distribution because the families’ income might increase. Furthermore, because job market participation is significantly affected by the cost of childcare (Kimmel, 1998), the tax credit might increase families’ income by enabling married mothers to work, thus leading to higher take-up rates among middle and high-income families.

13 Our results also show that a small percentage of unemployed lone parents and of two parent households where at least one parent is not working claim the deduction of childcare expenses. This shows that parent’s participation in the labour market is not the only reason why families use formal childcare. Duncan and Giles (1996) mention that families considered that the benefits from using formal childcare go beyond enabling mothers to work (Duncan A. and Giles Ch., 1996).

3 - Methodology

14 The progressivity of the tax relief is calculated by comparing the performance of the tax system on distribution of income with and without the tax policies for childcare. We use Aronson and Lambert (1994) decomposition of the Gini Coefficient to evaluate how the tax relief affects the vertical equity, horizontal equity and reranking of the tax system for all families with dependents but also only for families claiming the tax deduction (Aronson J. and Lambert P., 1994). Indeed, when using tax relief for childcare, the distribution of benefits might change not only among households belonging to different income groups (vertical equity), but also among households within the same income group (horizontal equity).

15 The Aronson and Lambert methodology breaks down the total redistributive effect (RE) of taxation into three components: vertical equity involving the reduction of income gap between unequal individuals (V), horizontal equity loss for the unequal treatment of equals (H), and the redistribution loss from reranking households differently in the pre-tax and post-tax income distribution (R). The overall redistributive effect RE is computed as the difference between the pre-tax Gini coefficient Gx and the post-tax Gini coefficient Gx – t. The decomposition of the total redistributive effect of taxes and transfers into a vertical equity effect, and a loss of redistribution due to either horizontal inequity or reranking is the following:

16

17 where V captures the gain that would have existed if equals had been treated equally by the taxation system. In this case we compute what would have happened if all individuals belonging to the same income group paid the same amount of taxes. V is calculated as the difference between the pre-tax Gini coefficient and the post-tax Gini coefficient (Gb) obtained by assuming that each individual in the same income group pays the group’s average tax (i.e. perfect horizontal equity). The Atkinson-Plotnick (Atkinson A., 1980; Plotnick Robert, 1981) index of reranking R captures the effect of individuals being re-ranked from pre-tax to post-tax distribution and is computed as the difference between the post-tax Gini coefficient and the concentration index for post-tax income Cx – t. Indeed let Gx and Gx – t be the Gini coefficients for the pre-tax and post-tax income distributions respectively. Then the reduction of income inequality brought about by the tax can be decomposed as where the second term captures the redistribution loss of re-ranking. The argument is simple. If the tax causes reranking of income groups, then post-tax income is not an increasing function of pre-tax income and the Gini coefficient calculated over post-tax income must be corrected for the fact that the Lorenz curve must be adjusted accordingly since the n% poorest individuals in the distribution are no longer the same. More precisely, the concentration index is computed for the post-tax income using the ranking from the pre-tax income within each income group. Finally H captures the total loss of treating individuals with equal pre-tax income unequally. H is calculated as the weighted sum of the intra-group post-tax income Gini coefficient Gj. The weight of each group j is calculated as the square of the number of individuals in this group Nj divided by the square of the total population, multiplied by the average post-tax income of this group μj and divided by the average post-tax income μ. H takes only positive values. When H equals zero there is no differential treatment within groups (Gj=0 for all j).

18 To compute the net effect of the tax based policies for families with dependents under three years old we must compare the redistributive effect with and without the tax policies for childcare. From equation 1, we obtain:

19

20 where the superscript 0 indicates the net redistribution without tax deduction and tax credit; and the superscript 1 indicates the net redistribution when tax deduction and tax credit are included. Detailed derivation of equation 2 can be found in the appendix. If the tax relief for childcare improves the after tax income distribution then ΔV is positive. On the contrary, ΔV is negative if benefits from the tax relief are positively related to the families’ income. If ΔH is positive it implies that the childcare subsidies are horizontally inequitable. Similarly if ΔR is positive the childcare subsidies increase the reranking from the pretax to the post-tax distribution of income.

21 As mentioned, progressivity of tax relief is evaluated for all families with dependents and for families claiming the deduction of childcare expenses. We choose to do so for two reasons. Firstly, when considering all families with young dependents, we evaluate progressity of tax relief among the target population. This is particularly important in Belgium’s context, given that tax policies for childcare were set to grant some benefit to all families independently of their use of formal care. Secondly, because the main objective of introducing demand-side policies was to reduce the cost of formal childcare, progressivity of the tax relief needs to be evaluated among the families who pay for formal childcare.

4 - Results

4.1 - Take-up rate and households’ income

22 To understand how the tax deduction operates at different levels of income, we calculated the percentage of households that have deductible childcare expenses in each decile. This allows us to analyze if the take-up rate depends on the families’ income. Firstly, we notice in Figure 3 that there has been in increase in the percentage of households claiming the tax deduction in all income deciles from 1994 to 2001. Secondly, the percentage of families in each income decile that claim the tax deduction increases exponentially with income. Thirdly, the growth in the percentage of households using the tax deduction from families belonging to the income deciles 2 to 7 was bigger than 50%.

Percentage of households with young dependents claiming the tax deduction by income decile

Figure 3 - 

23 However, we found that among the families deducting, those belonging to income deciles two to six had an increased share in the use of the childcare tax deduction. This implies that middle-income families increased their use of the tax deduction, even though 59% of the total number of households using the deduction of childcare expenses still belonged to the upper part of the income distribution (income deciles seven to ten). Several authors have argued that higher take-up rate among middle and high-income families is not only related to the cost of formal care, but also to preferences on the type of childcare used. Indeed, a recent OCDE report showed that low-income families tend to rely more on informal care arrangements independently of the costs of formal childcare (Organization for Economic Co-operation and Development 2001).

24 The average fiscal benefit was also computed by income decile for years 1994 and 2001 (Table 2). There is a remarkable difference between the average fiscal benefit for all families with young dependents when compared with families claiming the tax deduction for formal childcare. This is due to the fact that the tax credit that can be claimed by families not using formal care is relatively small in comparison to the tax deduction. The fiscal benefit appears to increase with income. Nevertheless, focusing on the families claiming the tax deduction suggests a somewhat more even distribution, even if the average fiscal benefit still increases with income. We also found that the average fiscal benefit received by families increased for all income deciles from 1994 to 2001. Furthermore, families with deductible childcare expenses belonging to income deciles 4 to 6 increased their average fiscal benefit by more than 25%.

25 To have a better idea of the impact on the tax payments of the tax reliefs we computed the effective subsidy rate as the ratio between the fiscal benefit and the income. Altschuler and Swartz use this measure to analyze how progressive is the Child Care Tax Credit in the US (Altschuler R. and Schartz A. 1996). If this ratio increases (decreases) with income the tax relief is regressive (progressive). The results in Table 2 show that for all families with young dependents the effective subsidy rate credit is almost constant between income decile 4 to 9 and regressive for families in the income decile 1. Comparing 1994 and 2001, the relation between the fiscal benefit and the income increased for all income deciles. For the families with deductible childcare expenses the results in Table 2 (bottom) show that tax deduction is far more progressive given that the ratio decreases with income throughout income distribution. Furthermore when computing the ratio between the average fiscal benefit and the average tax payments we found that for income deciles one to three the tax deduction reduces their average tax payments by more than 10%. When comparing our results to those of the literature, we found a less equitable distribution of benefits from tax relief for families with dependents in Belgium than in the U.S. Indeed, Altshuler and Schwartz (1996) and Gentry and Hagy (1995) found that for families with dependents the tax relief was at least progressive in the second part of the income distribution (Altschuler R. and Schartz A. 1996; Gentry and Hagy 1995).

4.2 - The redistributive effect

26 Hereafter, we examined the performance of the tax system without including the tax policies for childcare. Detailed results can be found in Table A 1 and Table A 2 in the appendix. We found that in Belgium income tax was progressive between 1994 and 2001 for both families with young dependents and for families using formal care. Furthermore, from 1996 to 2001 the redistributive coefficient RE0 increased constantly, implying that income inequalities have been further reduced through the tax system. During this period, the redistributive coefficient RE0 was on average of 6,3% for families with young dependents and of 5,3% for families claiming the tax deduction. It is interesting to note that gains in vertical equity outpaced the loss of redistribution due to either horizontal inequity (H0) or reranking (R0).

4.2.1 - Impact of childcare tax policies from 1994 to 1999

27 Table 3 shows the changes on the distribution of income due to the tax policies for childcare for families with young dependents. Detailed results, on Gini indices and Concentration indices can be found in Table A 1 in the appendix. We found that post-tax Gini index including the tax policies for childcare is bigger than without the use of the tax deduction and the tax credit , as a consequence, tax deduction and tax credit together make the income distribution more unequal. This result might be related to the fact that take-up rates are higher among middle-income and high-income families, who receive higher benefits from the tax deduction. Indeed, the benefits from using the tax deduction are higher than the benefits from receiving the tax credit for childcare. It is important to mention that, because the deduction of childcare expenses and the tax credit are relatively small with respect to total tax payments, the changes in the income distribution are small. As a consequence, the tax policies for childcare do not affect the overall progressivity of the tax system. Gentry and Hagy (1995) also found that in the U.S., the tax relief is also too small to significantly influence the income distribution in the U.S.

28 Comparing each term of the decomposition of , we find that the regressive effect from the deduction of childcare expenses and tax credit comes from an increase in the vertical inequality. As a matter of fact, ΔV is smaller than zero, showing that tax deduction of childcare expenses and the tax credit worsens vertical equity. On the other hand, tax deduction and tax credit improve horizontal equity and the reranking component. When using the tax deduction of childcare expenses and the tax credit, families in the same income group are more equally treated by the tax system. Our results are not in accordance to those Altschuler and Schartz (1996) who found that the tax credit increased the overall progressivity of the income tax for taxpayers with dependents. Indeed, Altschuler and Schartz (1996) found that when including the tax credit, the Suits index increased of approximately 0,8%. Furthermore, our results are in line with those of Robins (1990) who point out that regressivity of tax relief of childcare expenses might be linked to higher take-up rates among middle and high-income families.

Estimation of the fiscal benefit by income decile

Table 2 - 

Decomposition of the redistributive effect of childcare tax policies for all households with dependents under three years old (%)

Table 3 - 

29 Table 4 shows the results on the equity of the fiscal benefit for families claiming the tax deduction for childcare expenses. Opposite to the results for all families with dependents, the net redistribution effect RE1 – RE0 is positive for families using formal childcare. Furthermore, between 1995 and 1999, this equalizing effect increased constantly. In 1995 RE0 was smaller than RE1 by 0.09% and in 1999 this difference was of 0.148%.

Decomposition of the redistributive effect of childcare tax policies for households using formal childcare (%)

Table 4 - 

30 We expected that tax relief for families using formal care would be regressive given that deductible childcare expenses are neither negatively related to the family’s income nor refundable. Nevertheless, progressivity of the tax relief might be due to the daily limit on the deductible care expenses. If the deductible limit binds the deduction for higher income families, it will enhance the progressivity of the tax deduction. Figure 4 shows the relation between deductible childcare expenses (before and after the reforms) and the daily fees paid by parents to subsidized care providers. It is important to mention that fees in the subsidized sector are set according to the family’s income. In 1999, the daily limit restricted the deduction for families with a post-tax income of more than 1.984EUR. As a consequence, the deduction made by high-income families was constrained by the daily limit. Yet, the progressivity of the tax relief also depends on the average amount paid by the users of formal childcare. In order to verify this hypothesis, we compared average fees and the maximum amount of childcare expenses that can be deducted. We found that in the French Community the average daily fee paid for a subsidized childcare place in 1999 was around 9,76EUR. In the Flemish Community the average daily fee amounted to 11,46EUR. Given that in 1999 the maximum deductible amount was set at 8,7EUR per day, we can say that in the subsidized sector at least half of the families’ deduction was restricted by the daily limit. In the private sector, fees are normally set above the maximum fee in the subsidized sector. This value is over 18EUR a day in both Communities. Because of this, we may say that families using private childcare facilities will be constrained to deduct the daily limit. Therefore, more than half of all the families using formal childcare are limited to claim the maximum deductible amount.

Daily fees in the subsidized sector and deductible expenses limits

Figure 4 - 

31 Previous studies showed that tax relief for childcare expenses is progressive for families claiming the tax deduction. Altschuler and Schartz (1996) found that when including the tax credit, the Suits index increased of approximately 4% for the families claiming the credit. As pointed out by Gentry and Hagy (1995), the progressivity of the tax relief does not only depend on take-up rates but also on features of the tax relief that determine the distribution of benefits. In Belgium’s case, the tax deduction of childcare expenses is progressive for families using formal care thanks to the daily deductible limit, while in the U.S, progressivity of the tax relief depends not only on the limits on deductible care expenses but also on the limits on the amount deducted with respect to the family’s income (Gentry and Hagy 1995).

4.2.2 - Impact of the tax policies for childcare in 2000 and 2001

32 In 2000 and 2001, the tax policies for childcare deteriorate the distribution of income for all families with young dependents. Indeed, the reforms were set to increase the amount of deductible care expenses while the tax credit was maintained almost unchanged.

33 For families claiming the tax deduction, the equalizing effect changed from 0.148% in 1999 to 0.145% in 2000. The reduction of 0.003% implies that the deduction of childcare expenses became less equitable in 2000 than it was in 1999. We may argue that this loss is created by the fact that the families benefiting from the limit change from 8.7EUR to 11.2EUR correspond to middle-income and high-income families. Indeed, in the subsidized sector, the increase in the deductible limit benefited families with a post-tax income of more than 1.984EUR (Figure 4, dotted area). When analysing the decomposition of RE1 – RE0 for 2000 we notice an improvement on the vertical equity (Δ > 0) as well as on the horizontal equity (– ΔH > 0) and reranking component (– ΔR > 0) due to the tax deduction of childcare expenses.

34 Opposite to the results in 2000, we observed an equity gain in 2001 that corresponds to the abolition of the 80% limit on the amount of childcare expenses that could be deducted. This benefited families with a post-tax income of less than 2.559EUR (Figure 4, grey area). Furthermore, given that the daily limit is not modified, the deductible amount for households who were already paying more that 11.2EUR was unchanged.

35 In order to have a better idea of the impact of the limit reforms on the distribution of the fiscal benefit for families with deductible childcare expenses we plotted the Lorenz curve for years 1999, 2000 and 2001. As we notice from Figure 5, the Lorenz curve for year 2000 lies below the 1999 curve indicating that, increasing the daily limit made the distribution of childcare fiscal benefit still more unequal. Nevertheless, setting the maximum percentage of deductible childcare expenses to 100% did partially offset the negative effect of increasing the daily limit.

Unequal distribution of the fiscal benefit

Figure 5 - 

Legend: Lorenz curve for the fiscal benefit for families with deductible childcare expenses.

5 - Conclusion

36 This paper examined the progressivity of the tax policies for childcare in Belgium for families with young dependents and for the users of formal childcare. Families using formal childcare can deduct a portion of their care expenses while those using informal or parental care arrangements benefit from a tax credit. Using an original sample of taxpayers with young dependent in Belgium, evidence is provided on the unequal distribution of childcare subsidies among income groups. When we take into account all families with young dependents, the fiscal subsidy programs turn out to have a regressive effect. On the contrary, among those families using formal childcare, it is shown that the tax deduction with a limit has a progressive effect on the distribution of income. Our results differ from those of Altshuler and Schwartz (1996) and Gentry and Hagy (1995) who find that the tax relief for childcare is progressive among all the families with young dependents. This can be due to the fact that in Belgium the amount deducted is not restricted by the family’s income, while in the U.S. families with income higher than 10.000 dollars see the credit rate decrease by one percent for each additional 2000 dollars of adjusted gross income.

37 We also found that the percentage of working parents (both from lone or two parent households) claiming the tax deduction increased from 1994 to 2001. Because the tax deduction for childcare is an effective mechanism to reduce the costs associated with the use of formal childcare, household’s participation in the labour market might have increased during this period. Although it is beyond the purpose of this paper to analyze changes in labor supply and use of childcare, it is important to mention that those changes might affect our results on the progressivity of the tax relief. Indeed, if tax relief for childcare induces low-income mothers to work, the distribution of income might be improved and our estimate on the redistribution effect might be underestimated. In addition to this, among all families with young dependents, regressivity of the tax policies for childcare might be in part due to the fact the tax relief enabled married mothers to work. Indeed, families claiming the deduction might have higher income because the mother was able to participate in the labour market.

38 Our results showed that the tax deduction could not always target the policy objectives of being progressive while reducing the cost of formal care. Indeed, increasing the daily deductible limit reduced the cost of formal childcare only for middle and high-income families, and increased inequality of income distribution. Furthermore, policy makers must consider that the tax deduction can reduce the cost of formal childcare only if access to childcare services is not restricted. Yet, in Belgium as in other developed countries, the increasing demand for childcare has not been matched by the supply of formal childcare. Further research is needed to understand how, in this context of rationing, tax relief of childcare expenses reduces the cost of formal childcare among families from different income groups. Indeed, if high-income families are more likely to overcome the rationing of childcare places, the tax relief can enhance the inequalities in the access to formal childcare.

Annexe

Appendix B

Decomposition of total redistributive effects of taxes and transfers for all households with dependents

Table A 1 - 

Year Total redistributive effect Vertical effect V 0 Horizontal effect H 0 Reranking effect R 0 Gini Index on gross Income Gini Index on the net income(No tax relief) Gini Index on the net income(With tax relief) Gini IndexWithout TaxDeduction Gini IndexWith Tax Deduction Concentration Index(No tax relief) Concentration Index (With tax relief) 1994 6,14 6,75 0,56 0,05 28,12 21,98 22,00 21,37 21,39 21,93 21,94 1995 5,72 6,26 0,47 0,07 27,04 21,32 21,35 20,78 20,81 21,25 21,28 1996 6,22 6,86 0,58 0,06 29,06 22,84 22,87 22,20 22,23 22,78 22,80 1997 6,35 7,05 0,63 0,07 29,97 23,62 23,64 22,92 22,95 23,55 23,57 1998 6,35 7,02 0,61 0,06 30,37 24,02 24,05 23,35 23,37 23,96 23,98 1999 6,39 7,08 0,63 0,06 30,70 24,31 24,33 23,62 23,64 24,25 24,27 2000 6,60 7,32 0,66 0,06 31,53 24,93 24,96 24,21 24,25 24,87 24,90 2001 6,87 7,64 0,61 0,16 31,50 24,63 24,66 23,86 23,90 24,47 24,51

Decomposition of total redistributive effects of taxes and transfers for households with deductible childcare expenses

Table A 2 - 

Year Total redistributive effect Vertical effect V 0 Horizontal effect H 0 Reranking effect R 0 Gini Index on gross Income Gini Index on the net income(No tax relief) Gini Index on the net income(With tax relief) Gini IndexWithout TaxDeduction Gini IndexWith Tax Deduction Concentration Index(No tax relief) Concentration Index (With tax relief) 1994 5,12 5,67 0,49 0,06 23,95 18,83 18,73 18,28 18,19 18,77 18,67 1995 4,57 5,06 0,42 0,07 22,09 17,52 17,43 17,03 16,95 17,45 17,36 1996 5,21 5,79 0,51 0,07 24,72 19,51 19,39 18,93 18,83 19,44 19,33 1997 5,43 6,05 0,56 0,06 25,60 20,17 20,04 19,55 19,43 20,11 19,98 1998 5,31 5,91 0,54 0,06 25,65 20,34 20,21 19,74 19,61 20,28 20,14 1999 5,36 5,99 0,56 0,07 26,15 20,79 20,64 20,16 20,03 20,72 20,58 2000 5,6 6,27 0,61 0,06 27,28 21,68 21,54 21,01 20,88 21,62 21,47 2001 5,8 6,65 0,59 0,26 27,65 21,85 21,65 21,00 20,81 21,59 20,39

Bibliographie

References

Altschuler R. and Schartz A. (1996). On the progressivity of the child care tax credit: Snapshot versus time-exposure incidence. National Tax Journal, 45(1), 55-71.

Aronson, J. and Lambert, P. (1994). Decomposing the gini coefficient to reveal the vertical, horizontal and reranking effects of income taxation. National Tax Journal, 47(2), 273-294.

Atkinson, A. (1980). Horizontal Equity and Distribution of the Tax Burden. In H. Aaron and M. Boskin (Eds). The Economics of Taxation. Washington, D.C.: The Brookings Institute.

Baum C. (2002). A dynamic analysis of the effect of child care cost on the work decisions of low income mothers with infants. Demography, 39(1), 139-164.

Blau, D. (2000). Child care subsidy programs, NBER Working Paper Series, WP7806. Cambridge, MA: National Bureau of Economic Research.

Blau, D. and Hagy, A.P. (1998). The Demand for Quality in Child Care. Journal of Political Economy, 106(1), 104-146.

Blau, D. and Robins, P.K. (1988). Child Care Costs and Family Labor Supply. Review of Economics and Statistics, 70(3), 374-381.

Connelly, R. (1992). The Effect of Child Care Costs on Married Women’s Labor Force Participation. Review of Economics and Statistics, 74, 83-90.

Dumbar, A. and Nordhauser, S. (2002). Is the Child Care Credit Progressive. National Tax Journal, 44(4), 519-528.

Duncan A. and Giles C. (1996). Should we subsidise Pre-school Childcare, and so, How? Fiscal Studies, 17(3), 39-61.

Gentry, W. and Hagy, A. (1995). The distributional effects of the tax treatment of child care expenses, NBER Working Paper Series, WP 5088. Cambridge, MA: National Bureau of Economic Research.

Gustafsson, S. and Stafford, F (1992). Child Care Subsidies and Labor Supply in Sweden. Journal of Human Resources, 27(1), 204-230.

Kimmel, J. (1998). Child care costs as a barrier to employment for single and married mothers. Review of Economics and Statistics, 80(2), 287-299.

Michalopoulos, C., Robins, Philip K., and Garfinkel, I. (1992). A Structural Model of Labor Supply and Child Care Demand. Journal of Human Resources, 27, 166-203.

Organization for Economic Co-operation and Development(2001). Starting Strong. Paris: OECD.

Plotnick R. (1981). A Measure of Horizontal Inequity. Review of Economics and Statistics, 63, 283-288.

Ribar, D.C. (1992). Child Care and the Labor Supply of Married Women: Reduced Form Evidence. Journal of Human Resources, 27, 134-165.

Robins, P.K. (1990). Federal Financing of Child Care: Alternative Approaches and Economic Implications. Population Research and Policy Review, 9, 65-90.

Notes

[ *] Health Sociology and Economics, School of Public Health, Université Catholique de Louvain, Brussels, Belgium.
Mailing address: Farfan-Portet Maria-Isabel, School of Public Health, Université Catholique de Louvain, Clos Chapelle aux champs 30.41, 1200 Brussels Belgium. Phone: +32 2 764 3098, Fax: +32 2 764 3031
e-mail: maria.farfan@uclouvain.be
Funding: Ministère de l’Enfance of the Belgian French Community.

[ **] CORE, Université Catholique de Louvain, Louvain-la-Neuve, Belgium.

Résumé

En Belgique, les familles ayant des jeunes enfants reçoivent différentes réduction d’impôts selon le type de garde utilisé. En effet, le gouvernement Fédéral octroi d’une part, une déduction fiscale des frais de garde pour les familles utilisant des milieux d’accueils agréés (garde formelle) et d’autre part, une exemption d’impôts pour les familles utilisant un autre type de garde (garde informelle ou parentale). Ce travail montre que les politiques fiscales en matière de garde sont progressives parmi les utilisateurs de la garde formelle, mais régressives pour l’ensemble des familles ayant des jeunes enfant.
Classification JEL : H71, H24.

Mots-clés

petite enfance, équité, politiques fiscales

PLAN DE L'ARTICLE

• 1 - Introduction
• 2 - Data
  • 2.1 - Evolution of tax relief for childcare
  • 2.2 - Employment and the tax deduction
• 3 - Methodology
• 4 - Results
  • 4.1 - Take-up rate and households’ income
  • 4.2 - The redistributive effect
• 5 - Conclusion
• Annexe

POUR CITER CET ARTICLE

Maria-Isabel Farfan-Portet et al. « Progressivity of Childcare Tax Policies in Belgium », Recherches économiques de Louvain 2/2008 (Vol. 74), p. 143-165.
URL : www.cairn.info/revue-recherches-economiques-de-louvain-2008-2-page-143.htm.
DOI : 10.3917/rel.742.0143.