Project no: 028412 AIM-AP Accurate Income Measurement for the Assessment of Public Policies Specific Targeted Research or Innovation Project Citizens and Governance in a Knowledge-based Society Deliverable 1.4c Home production and fringe benefits in Belgium Due date of deliverable: January 2008 Actual submission date: March 2008 Start date of project: 1 February 2006 Duration: 3 years Lead partner: UA Revision [draft]
Project: Accurate Income Measurement for the Assessment of Public Policies (AIM-AP) Part I: Non-cash incomes Work Package: Other Non-Cash Incomes THE DISTRIBUTIONAL IMPACT OF OTHER NON-CASH INCOMES IN BELGIUM Gerlinde Verbist (*) & Stijn Lefebure March 2008 (*) corresponding author Centre for Social Policy Herman Deleeck University of Antwerp St.-Jacobstraat 2 B-2000 Antwerpen Tel.: + 32 (0)3 275.55.53 Fax.: + 32 (0)3 275.57.90.
Contents 1. Introduction 2 2. The calculation of non-cash income: home production and employee benefits 2 2.1 Home Production 3 2.2 Employer-provided fringe benefits 3 3. Results 4 3.1 Income advantages from company cars 5 3.2 Effect on income inequality and poverty 5 3.3 Breakdowns for characteristics of the household 6 4. Summary 7 5. References 7 6. Tables 9 1
1. Introduction Distributional analyses mainly focus on inequality of cash incomes (see e.g. Atkinson et al. 1995; Marical et al. 2006). However, as the balance between cash transfers and social benefits in kind may vary between countries, in kind services should in principle be included in the analysis to give a more accurate picture. In previous reports we have already discussed the distributional effects of the most important noncash incomes, namely imputed rent as private income in-kind, and education and health care as publicly provided benefits (see Goedemé &Verbist, 2006; Verbist & Lefebure 2007a & 2007b). In this report we discuss other private incomes in-kind, and the way these can be incorporated for Belgium in the income concept. We distinguish fringe benefits on the one hand and own production of goods and services for direct consumption or barter with others on the other hand. These two income types are expected to be located at different sides of the income distribution. Fringe benefits in the form of employerprovided perquisites are usually most prominent among employees who are already higher up the wage ladder. Home production is thought to occur more among farmers or small shop owners, who tend to be more present in the lower end of the cash income distribution. In the following section we present a brief overview of these other private non-cash incomes in Belgium, and to what extent we can account for them in our distributional analysis. We then calculate the distributive effects and present the results in section 3. Section 4 summarizes the main findings. 2. The calculation of non-cash income: home production and employee benefits We briefly discuss here non-cash income from home production and employerprovided fringe benefits in Belgium. Our distributional analyses is performed on the Belgian EU-SILC of the survey year 2004 (with income data referring to 2003). We have used the Belgian dataset, which apart from the variables provided to EUROSTAT also contains some extra information. The SILC-2004 of Belgium contains 5,275 households and 12,971 individuals. For the distribution analysis households with a negative or zero household income were excluded, which leaves us with 5,248 households and 12,930 individuals for our analyses. 2
2.1 Home Production Previous estimates indicate the overall impact of home production on income distribution in Belgium to be fairly limited. Survey based estimates indicate goods produced for own consumption make out no more than.09% of total household income in 2005, a figure that has not changed between 1978 and 2005 (Algemene Directie Statistiek en Economische Informatie 2005). A broader definition of home production which also includes household work, which is based on time use data, indicate home production does affect the relative income position of specific social groups (Van Dongen e.a. 1995, Van den Bosch 1999). Van den Bosch (1999) shows home production lowers inequality between single and dual earner families. After inclusion of home production the equivalised income gap between single and dual earner families is halved. Also, poverty in households with children is higher when non-cash income from home production is included. Obviously, estimates of the impact depend largely on the valuation of one hour of household work. As the 2004 wave of EU-SILC does not provide information on the monetary value of home production, or on time spent on this activity, we are not able to calculate the distributional effect of including the value of home production in the income concept. Estimates on home production will however be included in later waves of the data, so in the near future, this type of analysis should be possible. 2.2 Employer-provided fringe benefits Employer provided fringe benefits grasp a broad range of advantages including company cars, meal vouchers, company products, company homes, service regarding sports and leisure, child care compensation and membership fees to professional organisations. Employer provided fringe benefits make up 1.87% of total labour cost in Belgium (Algemene Directie Statistiek en Economische Informatie 2004). This share is larger in the service sector (2.07%) than in the industry (1.87%). An 3
important share of these benefits in Belgium are rather near-cash than non-cash income. A widespread system of meal vouchers is in place which can be used to pay in supermarkets or restaurants. In 2004, 24.4% of all employed adults received these vouchers with an average yearly value of 841. Household income and wages in the EU-SILC already includes this near-cash benefit. Since the utility of the meal voucher virtually equals the utility of it s monetary value, this near-cash income will not be separated back from the monetary income and be considered as part of the monetary wage. The 2004 Belgian EU-SILC data does not provide much information on employer provided non-cash benefits. At this point, we are restricted to information on the monetary value of the private use of company cars. Other non-cash income components are scheduled to be included in later waves of EU-SILC. The variable PY020N should encompass Company cars and associated costs (e.g. free fuel, car insurance, taxes and duties as applicable) provided for either private use or both private and work use. [ ] The value of goods and services provided free shall be calculated according to the market value of these goods and services. The value of the goods and services provided at reduced price shall be calculated as the difference between market value and the amount paid by the employee. (Eurostat 2006). EU- SILC reports 7% of employed adults have access to a company car, with an average income advantage per beneficiary of around 1400. 3. Results We now present the impact of including fringe benefits costs in the income concept on income inequality. The unit of analysis is the individual in the context of his household. income is household disposable income on a yearly basis excluding fringe benefits. The income advantage of company cars for the household is compared to the baseline in both absolute and relative terms. Both disposable income and the income advantage from fringe benefits are equivalised in order to take account of family size and composition. The equivalence scale used is the modified OECDscale, which attributes a value 1 to the first adult, 0.5 to each other adult and 0.3 to each child. Results are presented in the tables at the end of this paper. These results confirm a similar exercise done by Frick et al., 2007. 4
3.1 Income advantages from company cars Average income advantages from fringe benefits per quintile are presented in Table C.2. The effect of including company cars in the income concept is very small. For all households, disposable income increases with 0.4%, and only 8.2% of individuals live in a household that benefits from a company car. However, there is a clear distributional pattern: both the size of the benefit (in absolute and relative terms), as well as the number of beneficiaries increases as we go up the income ladder. In the lowest quintile, disposable income hardly increases due to fringe benefits and the mean per capita benefit is only 2 Euro, whereas in the highest quintile this amounts to 0.6%, resp. 117 Euro. For population share this is also very clear with 0.7% beneficiaries in the lowest quintile against 21.3% in the top of the distribution. A similar pattern appears if we focus only on employee income and the effect of including company cars in the wage concept (see Table C.2bis, all amounts are nonequivalised). 3.2 Effect on income inequality and poverty The effect of including fringe benefits in the income concept on income inequality and poverty is measured by calculating a series of commonly used inequality and poverty measures for both baseline income and for plus fringe benefits. The inequality measures used are the Gini index and the Atkinson index for inequality aversion parameters 0.5 and 1.5. The poverty measures are those from the FGT family with parameters 0 (head count), 1 (normalized poverty gap) and 2 (average squared normalized poverty gap)(see Foster, Greer and Thorbecke, 1984). As can be expected from our analysis on the basis of quintile distributions, inclusion of fringe benefits in the income concept pushes inequality and poverty slightly up (see Table D). The Gini coefficient decreases with 0.4%. The inequality indicators that is more sensitive to changes at the top of the distribution, namely the Atkinson 0.5, reports a somewhat higher increase in inequality (0.7%). The effect on poverty is also very small with an increase varying from 0.2% (FGT1) to 0.7% (FGT0). 5
3.3 Breakdowns for characteristics of the household Table E.1 presents average income before and after inclusion of fringe benefits broken down for household characteristics, as well as the decomposable MLD, which allows distinguishing within and between-group inequality. The household characteristics considered are: household type; socio-economic group of the reference person; educational level of the reference person and age of the household member. The highest increase in income is experienced by couples with children (+0.6%), white collar workers, individuals with a high education level and younger than 65. Inequality increases also the most within these groups. As these groups are also the ones who already have the highest income before inclusion of company cars, it is not surprising that between-group inequality goes up due to fringe benefits. This is also illustrated in the Table below which presents within and between groups inequality for employees with and without fringe benefits. As fringe benefits mainly go to individuals who already have a much higher wage before inclusion of these benefits (156% of overall average wages), inequality between these two groups increases considerably with 23%. Table 1: Between- and within group inequality for employees with and without company cars, Belgium 2003. FB Recipient % Within groups inequality % Between groups inequality All no yes Pop. share in % 92.9% 7.1% 100.0% Mean income 95.7% 155.8% 100.0% Including FB 95.2% 162.7% 100.0% % change in mean income 0.0% 5.0% 0.6% Mean Log Deviation (MLD) 0.2033 0.1072 0.1964 0.0090 0.2054 Including FB 0.2033 0.0989 0.1958 0.0111 0.2069 % change in inequality 0.0% -7.7% -0.3% 23.0% 0.7% % contribution to aggregate inequality 91.9% 3.7% 95.6% 4.4% 100.0% Including FB 91.2% 3.4% 94.6% 5.4% 100.0% Source: Own calculations on EU-SILC 2004. 6
Decomposing the FGT-poverty measures for these characteristics yields a different pattern for FGT0 and FGT2 (see Table E.2). FGT0 increases mostly for monoparental households, self-employed households, highly educated and young individuals. This is probably due to individuals who are just around the poverty line, and are pushed below this line because of the small increase in the poverty line because of including fringe benefits. However, FGT2 mainly increases for singles, pensioners and older individuals; this is mainly due that due to inclusion of company cars, the gap becomes wider for those who do not have access to this advantage. 4. Summary Summarising, we can say that the distributional effects of incorporating other noncash incomes in the income concept are very limited. This is partially due to data limitations, as no information is available in the 2004 SILC dataset on the monetary value of home production or on most employer-provided fringe benefits. We could only account for the effect of including the monetary value of private use of company cars, which resulted in a small upward shift of inequality, as this benefit is more concentrated in the top of the wage distribution. Consequently, inequality between employees with and without these fringe benefits increased considerably. 5. References Algemene Directie Statistiek en Economische Informatie (2004). Resultaten van de Enquête naar de Arbeidskosten 2004. Brussel: Federale Overheidsdienst Economie, K.M.O., Middenstand en Energie. Algemene Directie Statistiek en Economische Informatie (2005). Huishoudbudgetenquete. Brussel: Federale Overheidsdienst Economie, K.M.O., Middenstand en Energie. Atkinson, A. B., Rainwater, L., & Smeeding, T. M. (1995). Income distribution in OECD countries: Evidence from the Luxembourg Income Study (LIS). Social Policy Studies No. 18. Paris: Organization for Economic Cooperation and Development. Foster J., Greer, J. & Thorbecke, E. (1984). A class of decomposable poverty measures in Econometrica, vol. 52, pp. 761-766. Frick J., Goebel J., Grabka M. (2007), Assessing the Distributional Impact of "Imputed Rent" and "Non-cash Employee Income" in Microdata: Case Studies Based on EU-SILC (2004) and SOEP (2002., DIW, Berlin, SOEP-papers 2/2007. Goedemé T. & Verbist G. (2006), The distributional impact of public education in Belgium, Country report in the framework of the European Research Project Accurate Income Measurement for the Assessment of Public Policies (AIM-AP), Antwerp, Centre for Social Policy. 7
Marical, F., M. Mira d'ercole, M. Vaalavuo and G. Verbist (2006), Publicly-provided Services and the Distribution of Resources, OECD Social, Employment and Migration Working Papers No. 45, OECD, Paris, 59p Rainwater, L. (1979). Mothers' Contribution To the Family Money Economy in Europe and the United States. Journal of Family History, 4: 198-211 Van den Bosch, K. (1999). Identifying the poor. Using subjective and consensual measures. Dissertation: Antwerp University. Van Dongen, W. Malfait, D. & K. Pauwels (1995). De dagelijkse puzzel "gezin en arbeid" : feiten, wensen en problemen inzake de combinatie van beroeps- en gezinsarbeid in Vlaanderen. Brussel : Ministerie van de Vlaamse Gemeenschap. Verbist G. & Lefebure S. (2007a), The distributional impact of imputed rent in Belgium, Country report in the framework of the European Research Project Accurate Income Measurement for the Assessment of Public Policies (AIM-AP), Antwerp, Centre for Social Policy. Verbist G. & Lefebure S. (2007b), The distributional impact of health care in Belgium, Country report in the framework of the European Research Project Accurate Income Measurement for the Assessment of Public Policies (AIM-AP), Antwerp, Centre for Social Policy. 8
6. Tables Table C.2: Income advantages from company cars, quintile distribution, Belgium 2003. Disposable income (without fringe benefits) (EUR) mean Disposable income plus fringe benefits (EUR) mean % Increase in disposable income Income Share Disposable income Income Share plus fringe benefits Population share with fringe benefits Mean per capita benefit Equivalized benefit (EUR) mean Income advantage from fringe benefits (beneficiaries only) Quintile 1 (bottom) 7334 7337 0.0% 8.9 8.8 0.7% 2 3 455 2 11733 11748 0.1% 14.2 14.2 3.1% 9 16 508 3 15233 15263 0.2% 18.4 18.4 4.4% 18 29 654 4 19333 19416 0.4% 23.4 23.4 11.5% 51 83 723 5 (top) 29060 29239 0.6% 35.1 35.2 21.3% 117 179 839 All 16537 16599 0.4% 100.0 100.0 8.2% 39 62 755 Table C.2bis: Income advantages from company cars, quintile net-employee income distribution, Belgium 2003 (non-equivalised amounts, only individuals with PY010N>0). (EUR) mean: net employee Employee income plus fringe benefits % Increase in income Income Share Income Share plus fringe benefits Population share with fringe Fringe benefit Income advantage from fringe benefits (beneficiaries Quintile income (EUR) mean benefits (EUR) mean only) 1 (bottom) 5922 5930 0.1% 6.5 6.5 1.0% 8 862 2 13604 13635 0.2% 15.0 15.0 2.5% 31 1236 3 17348 17414 0.4% 19.2 19.1 4.7% 66 1409 4 21190 21295 0.5% 23.4 23.4 7.9% 105 1330 5 (top) 32487 32785 0.9% 35.9 36.0 19.6% 298 1518 All 18107 18209 0.6% 100.0 100.0 7.1% 102 1425 9
Table D: Inequality and poverty indices, Belgium 2003. Inequality indices Value of the Index: equivalised household disposable income Income minus fringe benefits % change Gini 0.2644 0.2655 0.4 Atkinson 0.5 0.0599 0.0603 0.7 Atkinson 1.5 0.2397 0.2409 0.5 MLD 0.1327 0.1336 0.7 Half SCV 0.1437 0.1447 0.8 DR: 90/10 3.34 3.36 0.8 DR: 90/50 1.71 1.72 0.6 DR: 50/10 1.95 1.95 0.1 FGT0 0.1530 0.1540 0.7 FGT1 0.0421 0.0422 0.2 FGT2 0.0191 0.0192 0.3 Inequality indices Value of the Index: net employee income Income plus fringe benefits % change Gini 0.2886 0.2903 0.6 Atkinson 0.5 0.0833 0.0839 0.8 Atkinson 1.5 0.3249 0.3265 0.5 MLD 0.2054 0.2069 0.7 Half SCV 0.1768 0.1784 1.0 DR: 90/10 4.41 4.49 1.9 DR: 90/50 1.63 1.65 1.1 DR: 50/10 2.70 2.73 0.8 Source: own calculations on SILC-Belgium 2004. 10
Table E.1: Inequality decomposition by household characteristics, Belgium 2003. Characteristic of household or household head A B C D E F G H I Mean Income position % increase Mean Log Deviation (MLD) % contribution to aggregate inequality Pop. share in % Including FB Including FB in mean equiv. Income Including FB % change in inequa lity Including FB Household type Older single persons or couples (at least one 65+) 15.5 13574 13577 82 82 0.0 0.1052 0.1052 0.1 12.3 12.2 Younger single persons or couples (none 65+) 22.7 18799 18862 114 114 0.3 0.1679 0.1689 0.6 28.7 28.6 Couple with children up to 18 (no other HH members) 37.8 17057 17155 103 103 0.6 0.1121 0.1133 1.1 31.9 32.1 Mono-parental household 5.8 11539 11568 70 70 0.3 0.1083 0.1092 0.9 4.7 4.7 Other household types 18.3 16749 16795 101 101 0.3 0.1172 0.1177 0.4 16.1 16.1 % Within groups inequality./../../../../../. 0.1244 0.1252 0.7 93.7 93.7 % Between groups inequality./../../../../../. 0.0083 0.0084 1.6 6.3 6.3 Socioeconomic group of HH head Blue collar worker 19.0 15675 15698 95 95 0.1 0.0766 0.0766 0.1 11.0 10.9 White collar worker 32.7 21062 21222 127 128 0.8 0.0781 0.0788 0.9 19.2 19.3 Self-employed 10.3 16819 16847 102 101 0.2 0.1904 0.1906 0.1 14.8 14.7 Unemployed 8.1 10603 10609 64 64 0.0 0.1096 0.1097 0.1 6.7 6.6 Pensioner 23.8 14314 14321 87 86 0.0 0.1148 0.1149 0.1 20.6 20.4 Other 6.2 11536 11542 70 70 0.0 0.2035 0.2038 0.1 9.5 9.5 % Within groups inequality./../../../../../. 0.1084 0.1087 0.3 81.7 81.4 % Between groups inequality./../../../../../. 0.0235 0.0241 2.7 17.7 18.1 Educational level of HH head Tertiary education 31.2 20798 20937 126 126 0.7 0.1130 0.1140 0.9 26.6 26.6 Upper secondary education 34.7 15999 16039 97 97 0.2 0.1074 0.1078 0.4 28.1 28.0 Lower secondary education 15.1 14563 14586 88 88 0.2 0.1303 0.1306 0.2 14.9 14.8 Primary education or less 18.9 12206 12216 74 74 0.1 0.1127 0.1128 0.1 16.1 16.0 % Within groups inequality./../../../../../. 0.1147 0.1153 0.5 86.5 86.3 % Between groups inequality./../../../../../. 0.0179 0.0184 2.5 13.5 13.7 11
Age of HH member Below 25 29.5 15596 15669 94 94 0.5 0.1261 0.1273 0.9 28.0 28.1 25-64 53.9 17918 17992 108 108 0.4 0.1345 0.1354 0.7 54.7 54.7 Over 64 16.6 13724 13729 83 83 0.0 0.1082 0.1083 0.1 13.6 13.5 % Within groups inequality./../../../../../. 0.1277 0.1285 0.7 96.2 96.2 % Between groups inequality./../../../../../. 0.0050 0.0051 1.8 3.8 3.8 ALL 100.0 100 100 0.4 0.1327 0.1336 0.7 100.0 100.0 Source: own calculations on SILC-Belgium 2004. 12
Table E.2: Poverty decomposition by household characteristics, Belgium 2003. Characteristic of household or household head A B C D E F G H I J K FGT0 % contribution to aggregate poverty (FGT0) FGT1 % contribution to aggregate poverty (FGT1) Pop. share in % Plus FB % change in poverty (FGT0) Plus FB Plus FB % change in poverty (FGT1) Household type Older single persons or couples (at least one 65+) 15.5 20.0 20.0 0.5 20.3 20.2 4.4 4.4 0.4 16.3 16.4 Younger single persons or couples (none 65+) 22.7 12.9 12.9 0.0 19.1 19.0 4.0 4.0 0.4 21.5 21.6 Couple with children up to 18 (no other HH 37.8 members) 12.3 12.4 1.1 30.4 30.5 3.6 3.6 0.3 31.9 31.9 Mono-parental household 5.8 33.4 33.9 1.7 12.6 12.7 8.0 8.0 0.1 10.9 10.9 Other household types 18.3 14.9 14.9 0.0 17.7 17.6 4.5 4.4 0.0 19.3 19.3 Socioeconomic group of HH head Blue collar worker 19.0 9.9 9.9 0.0 12.3 12.2 2.3 2.3-0.3 10.2 10.1 White collar worker 32.7 2.5 2.5 0.1 5.3 5.2 0.7 0.7-0.6 5.2 5.2 Self-employed 10.3 19.0 19.3 1.5 12.8 12.9 6.9 6.9 0.3 16.8 16.8 Unemployed 8.1 42.5 42.9 0.9 22.4 22.5 11.3 11.4 0.4 21.7 21.8 Pensioner 23.8 18.8 18.8 0.3 29.1 29.0 4.7 4.8 0.4 26.8 26.8 Other 6.2 42.0 42.0 0.0 17.0 16.9 12.1 12.2 0.3 17.9 17.9 Educational level of HH head Tertiary education 31.2 6.1 6.3 3.1 12.4 12.7 2.0 2.0 0.0 14.7 14.7 Upper secondary education 34.7 13.8 13.9 0.7 31.2 31.3 3.7 3.7 0.3 30.5 30.5 Lower secondary education 15.1 20.0 20.0 0.0 19.8 19.7 5.9 6.0 0.3 21.3 21.3 Primary education or less 18.9 28.6 28.7 0.3 35.4 35.2 7.1 7.2 0.3 32.0 32.1 Age of HH member Below 25 29.5 17.9 18.1 1.1 34.5 34.6 5.1 5.1 0.2 35.7 35.7 25-64 53.9 12.4 12.5 0.4 43.8 43.6 3.6 3.6 0.3 46.1 46.2 Over 64 16.6 20.1 20.2 0.5 21.8 21.7 4.6 4.6 0.5 18.1 18.1 ALL 100.0 15.3 15.4 0.7 100.0 100.0 4.2 4.2 0.2 99.9 100.0 Source: own calculations on SILC-Belgium 2004. Plus FB 13
Table E.2: Poverty decomposition by household characteristics, Belgium 2003 (continued). Characteristic of household or household head L M N O P FGT2 Plus FB % change % contribution to in poverty aggregate poverty (FGT2) (FGT2) Plus FB Household type Single persons / couples (65+) 1.7 1.7 0.6 13.8 13.8 Single persons or couples (none 65+) 2.1 2.1 0.1 24.4 24.4 Couple with children up to 18 1.7 1.7 0.1 33.4 33.3 Mono-parental household 3.4 3.4 0.1 10.3 10.3 Other household types 1.9 1.9-0.1 18.2 18.2 Socioeconomic group of HH head Blue collar worker 1.0 1.0-0.2 9.8 9.8 White collar worker 0.3 0.3 0.0 5.0 4.9 Self-employed 3.8 3.8 0.3 20.3 20.3 Unemployed 4.5 4.5 0.3 18.9 18.9 Pensioner 2.0 2.0 0.5 24.3 24.4 Other 6.1 6.1 0.3 19.8 19.8 Educational level of HH head Tertiary education 1.1 1.1 0.0 17.3 17.3 Upper secondary education 1.6 1.7 0.6 29.8 29.9 Lower secondary education 2.8 2.8 0.4 22.0 22.0 Primary education or less 2.9 3.0 0.3 29.1 29.1 Age of HH member Below 25 2.4 2.4 0.0 36.5 36.4 25-64 1.7 1.7 0.2 48.2 48.2 Over 64 1.8 1.8 0.6 15.4 15.4 ALL 1.9 1.9 0.3 100.1 100.0 Source: own calculations on SILC-Belgium 2004. 14