THE WORKING POOR IN FLANDERS, A FORGOTTEN GROUP?
Author(s): Ive Marx, Gerlinde Verbist, Pieter Vandenbroucke, Kristel Bogaerts and Josefine Vanhille
1.
Abstract
Compared with the rest of Europe, Flanders has relatively few working poor, only 2.5%. According to the standard which is commonly applied in the European Union a person is poor when s/he lives in a household with an equivalent family income lower than 60% of the country’s median income. Yet nearly one quarter of all Flemish of working age who live in financial poverty are employed. This poverty among employed Flemings is equated with relatively low-paid employment. This study contradicts this perception. Low earnings generally only become a problem if these earnings have to cover the needs of several dependents (children and/or adults). The phenomenon of the "working poor" is therefore closely related to an inability to generate sufficient earnings at family/household level, in an era in which the average living standard is strongly determined by dual earner households. A key element for adequate policy conceptualisation is the insight that the issue of the “working poor” cannot be seen as a special phenomenon that is not at all related to the general poverty issue therefore requiring a separate approach. An effective policy for the working poor is an integral and inseparable part of an effective poverty policy targeting children and people of working age in general. The policy option that is usually associated with the fight against poverty among the employed is the increase of the minimum wage. This perception is also contradicted in this study: increases, even substantial increases, of the minimum wage can only have marginal direct effects on poverty among the employed (or on poverty in general). An increase in the minimum wage would in fact represent an increase in the earnings of households who are already in the middle of the income distribution. That is why tax relief measures for low wages also have a limited impact on reducing poverty. An adequate policy should consist of a multipronged approach (1) because the "working poor" make up a diverse group and therefore require a diversified approach and (2) because each policy parameter - whether wage setting, taxation, special levies, benefits – in general only lends itself to incrementalist adjustments.
Key words: Poverty; working poor; social policy; minimum wage
2.
Study objectives
The objectives of the study were: 1. to identify and draw up a profile of the working poor in Flanders and in Europe; 2. to assess policy options to combat poverty among the employed, while analysing the pros and cons of the following options: an increase of the minimum wage; special levies and tax policy; benefits policy; encouraging dual earner households.
Clarification of key concepts This study examined poverty as financial poverty. We used the standard method of measurement applied in the European Union, i.e., someone is poor when s/he lives in a households with an equivalent household income that is less than 60% of the median income. The equivalent household income takes into account the number of family members and economies of scale in consumption. The equivalence factors used are 1 for the first adult; an equivalence factor 0.5 is added for each additional adult in the family and 0.3 for each child (this equivalence scale is called the modified OECD scale). A worker is defined as a person with earnings from paid labour. A self-employed person is someone who earns an income as a result of self-employed labour. A dependent does not have any earnings from either of these two sources. The standard of living is total disposable household income adjusted in function of the household size. The job discount is a personal income tax deduction granted by the Flemish Government between 2007 and 2010 to all workers in the Flemish Region.
1
3.
Methods and data
We use data from the European Union Survey on Income and Living Conditions (EU-SILC) of 2006, with incomes relating to 2005. The EU-SILC is a survey among a representative sample of families in Belgium and the other Member States of the European Union regarding earnings and living conditions. The survey is set up as a rotating panel. This means that for each new survey three quarters of the reference population are questioned again and one quarter of the reference population is made up of new families. The sample contains a total of 5,860 families and 14,329 individuals (N=7,764 for Flanders; N=4,771 for Wallonia; N=1,794 for Brussels). The analyses were carried among the working age population (between the ages of 16 and 64 years). The simulation technique was also used for the analyses. On the one hand the survey used the STASIM model, which calculates benefits and taxes for different types of households. On the other hand, the empirical microsimulation model MISIM was used, which simulates current and alternative tax benefit policies on the representative data of the EU-SILC. A dynamic element was also incorporated in this project in MISIM by considering the effect of a growth of employment on poverty.
4.
Findings
Compared with the rest of Europe, Flanders has relatively few working poor, only 2.5%. It is estimated that approx. 2.5% of all Flemish workers is financially poor according to the relative 60% limit (as calculated for Belgium – if a Flemish relative poverty standard is applied, this is closer to 5 percent). This is even lower for full-time workers, while it is a little higher for part-time workers. The incidence of poverty is significantly higher among self-employed workers but it is more difficult to carry out a reliable income measurement in this group than among the group of employees. Elements such as a relatively favourable social-demographic structure (relatively few single-parent families), a relatively compressed wage structure, support for the breadwinner’s role (e.g. the marriage quotient) and dual earners (e.g. the relatively strong development of childcare), universal child benefit at a relatively high level (and linked with this: relatively few working youth) contribute to keeping the proportion of working poor relatively low. This relatively positive picture as regards poverty among workers, however, needs to be strongly qualified. The relatively low poverty rate among workers is - to a large extent - a consequence of the fact that the lowest levels of income distribution are mainly populated by a comparatively high proportion of inactive people, whose poverty rate is relatively high due to the often inadequate replacement allowances. Yet nearly one quarter of all Flemings of working age who live in financial poverty are employed. This poverty among workers is strongly associated, equated even with relatively low-paid employment. This perception is contradicted in this study: Among the seven percent of Flemings who have a relatively low income from employment - traditionally the international definition for this is a gross wage for full-time employment that is lower than 67% of the median gross wage - only a minority (about 6 percent) is financially poor. This is because the vast majority of people with low earnings live in households where there are several incomes/earnings. Low earnings generally only become a problem if these earnings have to cover the needs of several dependents (children and/or adults). A full-time job at the minimum wage is sufficient for a standard of living above the poverty line if there are no dependent family members. If there are dependents - a dependent partner and/or children - then a problem arises if the family has no sources of income or insufficient other sources of income. This is the case if the family is unable to acquire a second source of income, as in the case of single parents, or if the partner does not wish to work or is incapable of working. There may be very different reasons for the latter aspect: difficulties in combining work and care, incapacity for work, unemployment due to lack of skills or local employment shortages. The phenomenon of the "working poor" is therefore closely related to an inability to generate sufficient earnings at household level, in an era in which the average living standard is essentially determined by the dual earner income. A key element for adequate policy conceptualisation is the insight that the issue of the “working poor” cannot be seen as a special phenomenon that is not at all related to the general poverty issue therefore requiring a separate approach. We use a traditional breadwinner family where the husband works and the wife does not work to illustrate this point. The male breadwinner receives a wage that is approximately the minimum wage. The family has two dependent children and therefore lives in financial poverty. If the poverty of this family is viewed from the angle of the breadwinner, then this person is defined as a “working poor person”. If this same family is observed from the viewpoint of the non-working partner then the conclusion of poverty is also reached, but this poverty is conceptualised as poverty resulting from unemployment. And if the poverty within this same family is observed from the viewpoint of the children, then it is defined as child poverty. The poverty phenomenon in all three cases is in effect the same. The consequence is that an effective policy for the working poor is an integral and inseparable part of an effective poverty policy aimed at children and people of working age in general. A further consequence is that choices
2
must be made here that are not normatively neutral. If the problem of poverty in this family is conceived as a problem of an inadequate breadwinner income then the policy consequences are clearly different than when it is conceived as a problem following from the fact that the partner has no income, or from the fact that there is no adequate compensation for the costs associated with the children. The policy option that is usually associated with the fight against poverty among the employed is the increase of the minimum wage. The analysis that was carried out in the frame of this study indicates in a robust manner that even substantial increases of the minimum wage can only have marginal direct effects on poverty among workers (or on poverty in general). One reason for this is that only a very small share of the working population is effectively paid a minimum wage. A second reason is that most people within this limited group do not find themselves at the bottom of the distribution in terms of their stand of living, the total disposable household income adjusted for family size. An increase in the minimum wage would in fact represent an increase in the earnings of households who are already in the middle of the income distribution. This should not be taken to mean that an increase of the minimum wage would not provide financial relief for single people or single-parent households, which have to survive on a minimum wage. But the fact remains that the spill-over to the people who are not poor would be quite substantial. This is relevant, however, if there is job loss associated with an increase of the minimum wage. That is why tax relief measures on low wages also have a limited impact on reducing poverty. The work bonus gives rise to a substantial increase in earnings for people with a low income. But as we have already indicated, most of these people are not at the bottom of the income distribution but rather in the middle. The vast majority of beneficiaries of the work bonus do not live in financial poverty, or are not even anywhere near the poverty line. The impact of the work bonus on poverty is thus limited. It is impossible to extend a measure such as the work bonus beyond the level of the minimum wages, unless the idea is to evolve towards an effective allowance, in line with the Anglo-Saxon model of ‘negative taxes’. The impact on poverty of such a measure would be very limited however. Moreover, the insurance logic of the social security system, and more in particular the equivalence logic of this system, restricts such forms of income-related selectivity. Tax measures in principle offer a better perspective for improving the income position of families and non-individuals with low earnings. The theoretical yield in terms of poverty reduction could potentially be much higher because resources can effectively be aimed at the bottom of the income distribution and not at the bottom section of the wage distribution, as these are two different factors, as we have already demonstrated. But there are several problems related to tax measures: (1) the problem of operating a sufficiently accurate distinction between 'real' and 'artificial' low-tax incomes, with all the problems of legitimacy that this entails; (2) the limited ability of taxation to respond to financial needs at the time that they effectively arise. Lump-sum tax measures such as the job discount at individual level, like the work bonus, have a very limited impact on poverty; on the contrary, they even give rise to a significant Matthew effect. Another policy option is better support for the work-family combination. In an era in which the average standard of living, and consequently the relative poverty line, is essentially defined by the standard of living of a dual-income family, single income families are at a considerable risk of poverty. A significant part of poverty among the working population affects 'traditional' breadwinner families. The combination of work and family continues to be a bottleneck in many cases. In recent years several initiatives were taken or set up to help eliminate bottlenecks: repeated rounds of expansion in subsidized childcare, renewed attention to priority policy, pilot projects focussing on flexible childcare, the use of service vouchers for childcare, the incentive to incorporate self-employed childminders in subsidized and hence income-related childcare facilities. To what extent these initiatives (will) actually provide sufficient solace remains to be evaluated. The position of single-parent families in this context is particularly precarious. Relatively few single-parent families succeed in realizing their full earning potential, especially when they have young children. A legitimate debate can be held about the desirability and feasibility of encouraging (full-time) employment among single parents. But helping them to realize their earning potential is clearly a significant policy option if one wants to combat financial poverty in this group. An effective implemented priority policy in terms of childcare could have an impact in this group. There is a significant overlap between child poverty and poverty among families who mainly rely on one income to make ends meet. The poverty risk of children growing up in a family without a working adult is very high (about 70%). But only a small minority of Flemish children (approximately 5%) live in such a household. A working adult considerably reduces the risk; nevertheless 60% of Flemish children who are confronted with financial poverty live in a household with at least one working adult. Nearly 1 in 5 lives in a family with two working adults. This points to the importance of child-related allowances. Internationally speaking, Belgium has a relatively generous system of childrelated allowances (child benefits and tax relief measures). Simulations show, however, that further increases of these allowances may have a significant impact on poverty, especially in larger households.
3
5.
Conclusions and policy implications
The phenomenon of the "working poor" is strongly related to an inability to generate sufficient earnings at household level. A key element for adequate policy conceptualisation is the insight that the issue of the “working poor” cannot be seen as a special phenomenon that is not at all related to the general poverty issue. An adequate policy should consist of a multi-pronged approach (1) because the "working poor" make up a diverse group and therefore require a diversified approach questions and (2) because each policy parameter - whether wage setting, taxation, special levies, benefits – in general only lends itself to incrementalist adjustments. In addition, there are important interdependencies between policy areas. To briefly illustrate this point: measures that selectively focus on low wages would eliminate a lot of pressure to negotiate higher minimum wages, potentially resulting in a shift of the cost. As an isolated measure, an increase of the minimum wage can only have a minor impact on poverty. The minimum wage nevertheless continues to be an essential ingredient of the policy mix. We use a simulation to examine whether the mere stimulation of employment growth is a policy route that has reached its limits if the objective was to reduce poverty. Using a tentative dynamic simulation model we explored the potential consequences for poverty of an employment growth to the level of 70% in Flanders. The model shows that employment growth has complex effects on poverty and on income distribution. On the one hand it is true that people who make the transition from inactivity to a job in general see significant improvement in their personal income position. On the other hand, this results in a slight increase of the relative poverty standard, meaning that others – in general those groups with the weakest employment opportunities - are once again pushed back a little bit further. The simulation indicates that the new earnings mainly go to households which already had earnings - meaning not to families at the bottom of the income distribution -, although 1 in 4 simulated earnings ends up benefiting families suffering as a result of unemployment. Although the model needs to be refined a little more to enable more detailed conclusions it is safe to say that the effect of an increase in employment, without appropriate flanking policy measures, does not have a clear positive consequence for poverty. Finally, a number of limitations of this study need to be mentioned. A first element is that this study only reviewed the situation of people who are regularly employed. Other studies have indicated that the really low-paid segments of the labour market are outside of the regular circuit. A second limitation is that not much is known yet about mobility patterns due to data restrictions. To what extent are the "working poor" upwardly mobile, or inclined to become inactive or unemployed again? A third limitation is that the profile presented in this study is rather schematic, largely due to data restrictions. The SILC sample for Flanders only allows for a limited degree of detail. These restrictions may be considered as an incentive for a more in-depth analysis. People who "quietly" survive with their earnings and who still have problems making ends meet, finding affordable housing, giving their children the necessary opportunities in life, etc., in many respects are still hidden from the 'institutional’ field of vision. In that sense they risk remaining an invisible group. And they probably deserve better.
Full reference of study report(s) and or paper(s) and other key publications of the study summarised here Marx, I., Verbist, G., Vandenbroucke, P., Bogaerts, K., & Vanhille, J. (2009). De werkende armen in Vlaanderen, een vergeten groep?. Onderzoeksrapport, Antwerpen: University of Antwerp, Centre for Social Policy. Marx, I., Verbist, G., Vandenbroucke, P., Bogaerts, K., & Vanhille, J. (2009). De werkende armen in Vlaanderen, een vergeten groep?. Over.werk: tijdschrift van het Steunpunt WSE, 19 (2), pp.133-136. Marx, I., Vanhille, J., & Verbist, G. (2010). Werkend en toch arm in de actieve welvaartsstaat. In J. Vranken, S. De Blust, D. Dierckx, A. Van Haarlem (Reds.), Armoede en Sociale Uitsluiting. Jaarboek 2010 (pp. 225-242). Leuven: Acco. Marx, I., Verbist, G., Vanhille, J., Bogaerts, K., & Vandenbroucke, P. (2011a). Het voorkomen van armoede onder werkenden in België: de beleidsopties overwogen. Belgisch Tijdschrift voor Sociale Zekerheid, 53(2), (forthcoming). Marx, I., Verbist, G., Vanhille, J., Bogaerts, K., & Vandenbroucke, P. (2011b). Wie zijn de werkende armen in België?. Belgisch Tijdschrift voor Sociale Zekerheid, 53(1), (forthcoming). Marx, I., Vanhille, J., & Verbist, G. (2012). Combating in-work poverty in continental Europe; the potential impact of higher minimum wages and targeted tax relief. Journal of Social Policy, (forthcoming in January 2012).
4