The national focus of ‘make work pay’ debates in relation to social protection and social inclusion – an evaluation from a gender perspective

Author:European Union Publications Office

1.1. The introduction or reform of tax credits to increase financial work incentives for the low-paid. 1.2. General tax reductions for the employed through reform of the personal taxation system. 1.3. Reform to unemployment benefit/assistance systems to ‘make work pay’ for the unemployed. Recent comprehensive reforms: the examples of Denmark, Germany, France and the Netherlands. Reforms with a narrower or more incremental focus on the unemployed: the examples of Austria, Portugal and Ireland. Reforms targeting employment (re-)integration of sick and disabled persons: the example of Norway. Reforms and proposals which focus on extending social protection coverage for the unemployed and inactive: the examples of Norway, Italy, Bulgaria, Romania and Lichtenstein. 1.4. Other measures to encourage the unemployed to take part-time or low-waged jobs. 1.5. ‘Making work pay’ in relation to pension reform and active ageing. 1.6. Work-family reconciliation as the main focus on ‘making work pay’.


Page 49

The first point to note is that the policy focus on ‘making work pay’ varies between the 20 countries (EU-15 plus the 5 non-EU countries, see Box 4 above for explanation) which we have reviewed, as summarised in Table 1.1. In those countries with limited social protection coverage for the working age population the emphasis of policy debates and reform is largely centred on extending social protection, as is the case in Greece and Italy. A similar situation of limited social protection coverage also applies to many of the new Member States, as indicated in their 2004 NAP/Inclusion reports (European Commission 2005) and in Bulgaria and Romania in their preparation for joining the EU. In contrast, in many of the countries with more extensive social protection systems there are examples of recent tax/benefit reforms designed to improve the financial work incentives of the unemployed and inactive and the low-paid employed. The examples presented in the national reports and discussed here are:

• Tax credits targeted at the low-paid have been introduced in Belgium, France and the UK.

• General tax reductions have been made in Luxembourg, Austria and Iceland (and more recently in Italy2).

• The system of unemployment support (either unemployment insurance and/or social assistance) has been reformed in a number of countries to tighten the eligibility criteria and/or reduce benefits, and to introduce more stringent job search conditions (Denmark, Germany, France, the Netherlands, Austria), while in Ireland this process has happened largely through the budget failure to uprate benefits which has undermined some recent positive measures to support integration into employment and education. In Portugal unemployment coverage has been extended, but accompanied by more stringent job search requirements.

• Disability pensions have been reformed to tighten eligibility and promote employment re-integration in Norway as a key part of social protection reform to enhance labour supply incentives.

In most of these countries there exists an explicit policy objective of ‘making work pay’ or removing unemployment and poverty traps for the unemployed and low-paid. However, this was not always the case – this was not a major consideration in the tax reforms in Luxembourg, the tax and unemployment benefit reforms in Austria were introduced without a public policy debate about ‘making work pay, and in Portugal the rationale presented for reforming social protection for the unemployed was to combat fraud and make the system more equitable.

However, in the other countries in this study there has been little or no emphasis on tax/benefit reform:

Page 50

• In Greece measures to help the unemployed into employment focus on wage subsidies for job creation and measures to make part-time work more attractive.

• In Finland proposals were made to subsidise low-paid jobs but these have been shelved.

• In Italy the proposal for benefit reform to extend social protection coverage has been postponed again. The lack of coverage means that there are no widespread problems of the ‘unemployment’ or ‘poverty’ trap, for only one in every five job-seekers receives some form of benefit and the vast majority rely on their families for economic aid.

• In Bulgaria and Romania there is no sustained debate about ‘making work pay’; instead in the preparation for EU membership the focus is on preparing the legal and institutional processes and developing economic and social policy in line with EU guidelines and requirements. Here the main issue is about extending social protection and introducing labour market programmes to enhance employability.

• In Lichtenstein a ‘making work pay’ debate is absent, and a more general debate about socio-economic reform is only just beginning in relation to rising social expenditure.

• In Spain a key focus in relation to ‘making work pay’ is on work-family reconciliation measures for mothers.

• In Sweden, while tax/benefit reform in relation to ‘making work pay’ is a perennial topic, which has been stimulated by the EES, no reforms have been made. Here the most relevant recent reform is the extension of childcare subsidies, which studies predict will further enhance mothers’ labour supply, particularly those in lower-income households.

The issue of pension reform in relation to work incentives and ‘making work pay’ was not included as an explicit element of the work programme, but a number of the national reports highlight reforms to pension systems which are largely designed to tighten eligibility conditions and to create financial incentives in favour of ‘active ageing’ and later retirement (France, Portugal, Austria, Italy, Norway), although Luxembourg and Lichtenstein provide examples of pension reforms which improve the situation of women with interrupted work histories (see Section 1.5).

The second overall point to make is that there is a general lack of gender mainstreaming or impact assessment in the reforms which have been implemented or discussed. This is consistent with the conclusions from previous analyses – that a developed gender perspective is lacking in most of the NAP/Inclusion and NAP/Employment reports for most of the Member States (Rubery et al., 2003, 2004; Fagan and Hebson, 2004). The absence of a gender perspective in this area is explicitly noted by the national expert in 13 of the 20 reports prepared on this topic (Austria, Denmark, France, Finland, Germany, Greece, Iceland, Ireland, Italy, Luxembourg, Portugal, Spain and Sweden), and none of the other 7 reach the conclusion that gender mainstreaming has been adequately applied to the policy area evaluated.

The absence of gender mainstreaming may indicate a lack of political commitment to promoting gender equity in some cases, or at the very least a failure to develop and implement suitable gender mainstreaming procedures. For example, the Italian expert (Villa, 2004) suggests that the gender dimension might be one of the implicit reasons for the lack of concern about ‘making work pay’ in Italy, in that the issue mainly concerns women’s situation and so it is not considered important. This is because the limited social protection coverage in Italy is primarily an issue for women in the highly dichotomised labour market – for it is women who are over-represented among the unemployed and inactive and those without social protection, while a higher proportion of men are in secure and stable jobs with relatively high forms of social protection.

In other cases, gender mainstreaming is developing but is still uneven. Thus, some of the reports (Ireland, the Netherlands, Norway and the UK) note that gender is discussed only in relation to certain groups where it is recognised either explicitly or implicitly that women predominate, such as among lone parents. In the case of Norway, the national expert notes that a positive interpretation of the lack of gender mainstreaming may reflect a certain degree of complacency that gender equity has been largely achieved based on the observation that there are few gender differences in some indicators such as qualification levels and employment rates. However, this ‘silence on gender’ is a problem forPage 51 there are still pronounced gender differences in many other areas, such as working hours and earnings (Ellingsæter, 2004). With regards to the uneven development of gender mainstreaming in the Netherlands, the Dutch expert notes that the situation of lone parents is discussed with tacit awareness that most of this group are women, but without a consideration of the gendered processes which give rise to this situation (Plantenga and Remery, 2004). At the heart of this is the way that the gender division of care responsibilities means that women are more likely than men to be raising children on their own, while at the same time being less able to secure well-paid employment due to gender wage gaps.

The tax credit reform in the UK is an example of where the policy was informed by an awareness of many of the gender impact implications, and where this gender perspective has been sharpened through dialogue between various social actors and the government. However, some of the weak elements in the policy design that were identified in this dialogue about gender impact have been retained by the government (Fagan et al., 2004), which illustrates how issues may be exposed through gender impact assessment but they may not be resolved due, for example, to their conflict with other policy priorities or to a lack of political will.

Table 1.1. Summary of the national ‘making work pay’ debates about social protection discussed in this report

The national ‘making work pay’ policy debate or reform discussed in the national experts’ reports for 20 countries

BE: Introduction of the earned income tax credit (CIBRAP), 2001 (Box 1.1)

Reforms to the earned income tax credit (CIBRAP) and proposals to reform the Income Guarantee Allowance (Box 1.2)

DK: Social assistances for the unemployed reduced to enhance financial work incentives under the ‘More people into work’ reform (Box 1.8)

DE: The ‘Hartz IV’ reform of unemployment benefit insurance and unemployment assistance (Box 1.9)

Expansion of ‘mini-jobs’ excluded from social protection coverage (Box 1.25)

EL: The focus of ‘making work pay’ policy reform is on wage subsidies and the creation of part-time jobs (Box 1.24)

ES: Work/family reconciliation measures are at the heart of the ‘making work pay’ debate (Box 1.28)

FR: The employment tax credit (the PPE) and transitional benefit retention for the unemployed entering employment (Intéressement) (Box 1.3)

Unemployment benefits reformed and new employer subsidies for recruiting the unemployed to minimum wage jobs (Box 1.10)

IE: Longstanding issues identified in the National Anti-Poverty Strategy to remove unemployment and poverty traps so as to ‘make work pay’ (Box 1.15)

The household-based emphasis in the tax/benefit system in Ireland (Box 1.16)

IT: Extension of social protection for the unemployed is postponed again (Box 1.22)

LU: Personal tax reform to reduce the already low tax burden, but few supply-side incentives for married women to enter employment (Box 1.5)

NL: ‘Making work pay’ through tax/benefit reform (Box 1.11)

The new Dutch ‘Work and Social Assistance’ Act (Wet Werk en Bijstand, WWB) came into force on 1 January 2004 (Box 1.12)

AT: The 2004-05 tax reform reduces tax payments for those with low earnings, with some reductions targeted at low-income families (Box 1.6)

Reforms to the tax/benefit system are designed to enhance work incentives (Box 1.13)

Page 52

Table 1.1. Summary of the national ‘making work pay’ debates about social protection discussed in this report (cont.)

The national ‘making work pay’ policy debate or reform discussed in the national experts’ reports for 20 countries

PT: Benefit reform for the unemployed and proposals to extend household-based means-testing (Box 1.14)

FI: Controversial proposals to subsidise low-wage jobs and expand the ‘working poor’ are deferred (Box 1.26)

SE: Sweden – high employment rates for both women and men suggest that social and economic policies have succeeded in ‘making work pay’ (Box 1.29)

UK: In-work benefits for employed parents on low income were extended in 2003 with the new Child Tax and Working Tax Credits (Box 1.4)

BG: There is no sustained debate about ‘making work pay’ through tax/benefit reform in the preparation for EU membership (Box 1.20)

IS: Tax reductions are the main element of recent tax/benefit reforms (Box 1.7)

LI: The buoyant economy means there is little pressure to stimulate debate for tax/benefit reform to ‘make work pay’, however employment rates for women are low (Box 1.23)

NO: Reforms to unemployment benefits and disability pensions to promote employment re-integration (Box 1.17) and ‘A more inclusive working life’ (the IA-agreement between the Norwegian government and the social partners in 2001) (Box 1.18)

A new initiative to integrate immigrant ethnic minorities into working life (Box 1.19)

RO: A number of reforms have been implemented in relation to the EES guideline 8 measures to ‘make work pay’ and the National Anti-Poverty and Social Inclusion Plan (Box 1.21)

Note: The national experts for the 10 new Member States were not asked to contribute to this part of the work programme, see introduction for explanation.

The third general point is that several of the experts note that where evaluation studies of policy impact have been undertaken there are often flaws in the design of the evaluation. These flaws are frequently compounded by the absence of gender mainstreaming, for when a gender perspective is taken this may inspire alternative policy resolutions. Thus, it is common for assessments to focus on the short-term, immediate financial incentives of tax/benefit reform in isolation from considerations of whether there are enough jobs available, working conditions (working hours and job quality) and childcare and other social infrastructure such as transport. Furthermore, longer-term considerations are typically neglected both in terms of the micro-household level of employment and income progression over the working life, and the macro-level of social integration, labour demand and the types of low-wage jobs being subsidised by social protection reforms (e.g. national reports for Denmark, Germany, France, Finland, the UK and Sweden).

In particular the Swedish report (Löfström, 2004) argues that gender mainstreaming produces a different angle on the perennial debate about tax levels and the cost of social protection systems. In Sweden, the ‘making work pay’ debate is a familiar one, with the political ‘left’ wanting to preserve social protection and associated training and labour market re-entry measures, while the political ‘right’ argue that benefit levels and taxes should be reduced so as to increase labour supply and employment. In recent years the emphasis of the debate has shifted towards tax/benefit reform, which in part reflects the political focus at EU level. Yet there has been little gender mainstreaming of this debate. Put simply, what would be the gender impact of lowering tax rates or reducing benefits? One impact would be upon women’s employment conditions, for 50% of Swedish women’s employment is in the tax-financed public sector, and many women – irrespective of which sector they are employed in – rely on publicPage 53 sector care services and care-related benefits. The second impact to consider is that while tax/benefit reform may improve financial work incentives, the actual effect on labour supply is harder to predict. Löftstrom (2004) argues that in Sweden during the 1990s the income replacement rates of benefits varied but without any susbtantial effects on labour supply. This might be because the economic recession neutralised any effect, but perhaps a more influential effect is that labour supply is determined more by wages, job opportunities and the system whereby most benefit eligibility rests upon building an employment work history, rather than by replacement rates. The third consideration is that taxes can stimulate labour supply if the funds are used to provide a social infrastructure that makes it easier for those with care responsibilities to take employment. Here falling birth rates across Europe may be one indication of the difficultires of work-family reconciliation, and in itself may be reason enough to review tax/benefit reform from a broader gender perspective than narrower debates about ‘making work pay’.

Having made these broad points we now turn to review the examples of recent policy reform in the different national arenas. The comparative analysis is contained in the main body of the text, while details of the national policy reform are contrained in boxes at appropriate points in the text, as indicated in Table 1.1 above. The boxes may be treated as background details on particular national policies, and some readers may wish to skip over the boxes and focus on the messages in the main text.

The discussion is organised into six sub-sections. The first Section focuses on tax credits, the second on general tax reforms, the third on reforms to unemployment benefit/assistance, the fourth on other measures to encourage the unemployed to take part-time or low-waged jobs, the fifth on ‘making work pay’ in relation to pension reforms and the sixth on two countries – Spain and Sweden – where work-family reconciliation measures are a central focus of ‘making work pay’ debates.

1.1. The introduction or reform of tax credits to increase financial work incentives for the low-paid

Tax credits have been recently introduced or reformed to promote work incentives for low-paid workers in Belgium, France and the UK in order to reduce the ‘unemployment trap’ and ‘poverty trap’. These traps are created when the interaction of the tax and benefit systems with the wage structure means that there is little or no net financial gain from taking employment or for employed persons with low income to increase their earnings (either by working longer hours or moving to a slightly better paid job).

In Belgium an individual earned income tax credit (CIBRAP) was introduced in 2001 as part of a wider tax reform to reduce the tax burden on earnings in general and the low-paid in particular, and as part of a wider drive to individualise the income tax system (see Box 1.1). A specific objective of the reform was to encourage more young people and women to enter employment. A refundable child tax credit was also introduced which increases in value with the number of children, where the refundable nature means that low-earning parents can receive a portion of the unused credit as a payment. An additional, non-refundable lone parent tax credit was also introduced.

Box 1.1. Belgium – Introduction of the earned income tax credit (CIBRAP) 2001

As part of a wider tax reform, an individualised earned income tax credit (CIBRAP) was introduced in 2001 as one of the measures to achieve the goal of reducing the tax burden on earnings and as part of a wider drive to individualise the income tax system. The credit is a fully individual right, independent of household type, covering most of the employed population (it is available to all persons working at least 13 hours per week with net earned income above 3 750 euro). One specific objective is to draw young people and women into employment and the other is to reduce the poverty risk for low-paid workers.

• The main drawback is that the CIBRAP is reimbursed 2 years after the income was earned, and as such cannot be considered to provide a real financial incentive to find employment.

The CIBRAP was introduced along with an alignment of the tax exempt income quota for single persons and married couples and the introduction of a refundable child tax credit which increases progressively with the number of children (1 180 euro for one child rising to 11 040 euro for four children, and + 4 420 euro for each subsequent child), and an additional non-refundable lone parent tax credit (1 180 euro). The refundable nature of the child tax credit means that low-earning parents who do not pay enough tax to consume the full credit get reimbursed the excess up to a threshold (340 euro per child).

Page 54

Box 1.1. Belgium – Introduction of the earned income tax credit (CIBRAP) 2001 (cont.)

The drawbacks of the new tax system from a gender equality perspective are:

• The system of personal income taxation is still not completely individualised and is based on the head of household. If the income of one spouse is below 30% of the household’s total income, then a share of the highest earner’s income is transferred to the lowest earning spouse in order to let the latter’s earnings rise to 30% of total household income (with a ceiling of [Euro.186]8 030) and is taxed at the lower income bracket. Husbands benefit from a reduced tax bill if their wives have limited earnings as a result of this ‘quotient conjugal’, and this contributes to an ‘inactivity trap’ for the second earner.

• The child tax credit is applied to reduce the tax bill for the spouse with the highest earnings (which in most cases is the male), rather than divided equally between the parents.

• The level of child tax credit increases progressively with family size and as such favours large families1, even though the cost of the second or third child is not higher than the first, and the opposite may even apply when the impact of motherhood on women’s employment and earnings is considered.

• The lone parent tax credit is not refundable for low-earning lone parents, unlike the provision for the child tax credit.

• In addition, parents are entitled to a childcare tax allowance paid until the child’s fourth birthday and up to a maximum threshold (2 464 euro/year or 11.2 euro/day) if they use a recognised form of formal childcare. This allowance is made to parents with either earnings or replacement income. It is divided between parents according to their contribution to household’s earnings, which usually means women receive a smaller share regardless of their contribution to the payment of the childcare costs.

Studies of the Belgian income tax system have shown that most of the tax relief is set against men’s earnings – they account for nearly all of the beneficiaries of the ‘quotient conjugal’ and the majority of the recipients of the childcare tax allowance and child tax credit.


1 The system of family allowances also increases progressively with family size (rising from 72.61 euro for one child to 200.59 euro per 3rd and each subsequent child) as well as age-related supplements.

Source: Meulders and O’Dorchai (2004)

From a gender quality perspective, several problems remain or have been created by this reform (Meulders and O’Dorchai, 2004). Firstly, the personal tax system is still not fully individualized and remains organised in relation to a ‘household head’ for there is an element of tax splitting through the ‘conjugal quotient’ and the child tax credit is set against the income of the highest earner. Both elements reduce the tax bill for the main earner and reinforce the ‘male breadwinner’ family arrangement. Secondly, the childcare tax allowance is organised differently to that of the child tax credit; for this element of tax relief is apportioned between parents according to their relative contribution to household earnings. However, the end result is similar, for men typically gain a large share of the allowance in light of their higher earnings and regardless of the actual contribution they make to the childcare costs borne when both parents are employed. Studies have shown that most of each of these elements of tax relief (conjugal quotient, child tax credit, childcare allowance) is set against men’s earnings. Thirdly, the child tax credit increases progressively in order to support larger families, yet the cost of the second and subsequent child is not higher than the first, and actually from the vantage point of the impact on the women’s employment pattern and lifetime earnings the main impact to consider is that of the transition to motherhood with the arrival of the first child. Fourthly, in addition to the problems identified by the national expert we can add that the additional tax credit for lone parents has a positive gender equity impact in that this channels relief to a low-income group which is predominantly women, however, the scale of this impact is limited because the tax credit is not refundable for low-paid lone parents, in contrast to the arrangement for the child tax credit for all parents.

Page 55

A key weakness of the CIBRAP was that it was reimbursed 2 years in arrears, and so did not provide a direct financial incentive for employment. It is being phased out and replaced by an ‘employment bonus’ which is an in-work benefit designed to ensure that even low earnings are substantially higher than remaining on benefits (see Box 1.2). There are also proposals on the table to reform the Income Guarantee Allowance to improve the incentives for those on low-income to search for longer part-time or full-time hours rather than taking ‘mini’ part-time jobs, but this has yet to be agreed by the social partners. The problems with this proposal is that it introduces tighter eligibility and lower benefit rates, and its efficacy at promoting the expansion of substantial rather than ‘mini’ jobs for the unemployed will be undermined by the lack of labour demand. An additional concern from a gender perspective is that the proposal is for a household benefit which focuses on improving the income and work incentives of the main earner, and excludes ‘dependant spouses’ (although there is equal treatment for cohabiting persons, but at an inferior rate to the main earner for married couples). Overall, the combination of both the ‘employment bonus’ and the proposed IGA reform has a limited impact on both disposable income and poverty in the context of job shortages. Labour supply models do predict that female participation rates will rise, but there is a lack of jobs for them to enter and so the impact of the reforms on promoting women’s employment will be limited overall.

An employment tax credit was also introduced for low-income employees in France in 2001 (see Box 1.3). It improves financial work incentives through reducing the income-related taper on benefit withdrawal, as does the recently introduced transitional intéressement payment for the unemployed who have recently found employment.

Box 1.2. Belgium – Reforms to the earned income tax credit (CIBRAP) and proposals to reform the Income Guarantee Allowance

The earned income tax credit (CIBRAP) will be replaced with the so-called ‘employment bonus’ (bonus crédit d’emploi), implemented gradually during the period 2004-07. This will replace the existing system of social security contribution reductions for low-earners and the CIBRAP. The aim of this reform is to promote labour market participation by ensuring that even low earnings are substantially higher than any replacement income. Once fully implemented the plan is that every person with gross earnings below 1 956 euro/month will be entitled to reduced social security contributions, with no contributions paid if gross earnings are below 1 095 euro/month.

There are also plans to reform the Income Guarantee Allowance to improve the incentives for those on low-income to increase their working hours. The IGA allows the full-time unemployed to retain a portion of benefit if they take on a part-time job up to the equivalent of 4/5 of full-time hours. They remain classified as unemployed and are required to meet job search criteria. Currently the payment made is calculated according to earnings, not the number of hours worked. The proposed reform will make the IGA proportional to the number of hours worked in order to provide an incentive to increase working hours, and will only be available to those working at least 30% of full-time hours.

Those with working hours that are less than 30% of a full-time job will not be able to retain this additional in-work payment, instead the benefit will be reduced by the full sum of the wage received, however, the unemployed may now refuse such short hour part-time jobs without jeopardising their ‘job-seeking’ status. The rationale is to leave ‘mini job’s’ for people voluntarily opting for part-time work while providing strong incentives for the unemployed to search for longer part-time hours or full-time work.

The main problems with this proposed reform of the IGA are:

• Many current IGA beneficiaries will not be eligible for the ‘in-work’ benefit given that one third currently are employed in short part-time jobs working less than 40% of full-time hours.

• The benefit received by those who remain entitled will be less generous than under the old formula.

• The impact of this labour-supply measure is heavily undermined by job shortages.

From a gender perspective, the main problem is that the IGA is not individualised, it is based on the household situation and organised in a way that reinforces the traditional gender roles within marriage.

• The reform focuses on improving the income and work incentives for the main earner. There is no incentive for the second earner to work longer hours, as ‘dependant spouses’ are excluded. In cohabiting couples each person can benefit, but at a lower rate of income supplement than that paid to the ‘breadwinner’ for married couples.

Source: Meulders and O’Dorchai (2004)

Page 56

Box 1.3. France – The employment tax credit (the PPE) and transitional benefit retention for the unemployed entering employment (Intéressement)

The PPE (employment tax credit) was introduced in 2001 for low-income employees. It improves financial work incentives through reducing the income-related taper on benefit withdrawal. Accompanying this, the withdrawal taper for housing benefit has also been reformed and local rates have been reduced for low-income households.

The PPE is awarded to low-paid employees (who earn between 0.3 and 1.4 times the full-time SMIC) to encourage a return to employment or to maintain employment for those who face an economic inactivity trap because of the low level of pay obtained from employment relative to minimum social benefits. The amount of the credit rises with earnings from 4.6% (for earnings between 3 372-11 239 euro/year) to 11.5% up to an individual ceiling (15 735 euro/year) or a lower total ceiling for couples (23 968 euro/year). PPE is not paid for very short part-time jobs (below 0.3 SMIC). When it was introduced it was paid proportionally for other part-timers (e.g. half the credit payment for those working half-time), but the calculation was made more advantageous in 2003.

Also of relevance is Intéressement which was introduced to reduce the ‘unemployment trap’ by allowing the low-income who enter employment to retain a small part of their benefit alongside pay for one year.

Source: Silvera (2004)

In evaluations of the PPE the main criticisms that are made are that the amounts received are small, even for the lowest paid, and the range of households covered too wide (about one quarter of households paying tax). Studies which measure the effect of PPE and intéressement on couples show that there is some incentive for dual-earner arrangements, but the incentive is reduced as the earnings level of the second earner rises and exceeds minimum pay levels, thus creating a disincentive to search for better paid jobs or to work longer hours. Furthermore, the 2003 reforms have reinforced the ‘part-time trap’ because the formula for calculating the credit creates more incentive to take a half-time job than a full-time one (Silvera, 2004). Thus, as for the Belgian reform discussed above, the result is that these elements of the French tax/benefit system encourage a main earner/second earner arrangement rather than a more equal dual-earner one, which is reinforced by the general thrust of the French tax splitting system which provides fiscal relief for one-earner families.

The French expert notes that there has not been a detailed gender impact assessment of this element of tax/benefit reform, but that women predominate among the target group of low-paid workers (Silvera, 2004). This is because the majority (more than 3/4) of the low-paid are women, and overall 27% of women are low-paid compared to 3.8% of men, as detailed in the statistical appendices of the French NAP/Employment 2003.

The UK’s tax credit system has been developed and extended as a key element of the government’s social inclusion and employment policy (Box 1.4). It has also gained some influence at EC level as one form of reform to consider in ‘make work pay’ debates, for the UK’s tax and benefit reform has been evaluated under the European Commission’s Peer Review Programme of the European Employment Strategy, the purpose of which is to identify and exchange good practices in employment policies3. This and other evaluations of the UK tax credit reforms have identified a number of positive and negative impacts from a gender and social inclusion perspective (Fagan et al., 2004), which merit some attention here for the general points they raise are pertinent for gender impact assessments of the development or reform of tax credits or other employment-conditional benefits in other Member States as part of the drive to ‘make work pay’.

Page 57

Box 1.4. The UK – In-work benefits for employed parents on low income were extended in 2003 with the new Child Tax and Working Tax Credits

The introduction and extension of tax credits for low-paid workers in the UK are designed to enhance financial work incentives for the low-paid. The main target groups identified by the 2003 NAP/Inclusion report are:

• ‘Workless’ couples of working-age – the UK has a high proportion of such households where neither adult is employed.

• Lone Parent households – the UK has one of the highest rates of lone parent households in the EU and a low employment rate for these households.

• Children living in poverty – the UK government is committed to halving the child poverty rate and raising the employment rate and income of low-income working families (couples and lone parents) is identified as the ‘best route out of poverty’.

The extension of tax credits have been developed in the context of:

• The introduction and extension of a National Minimum Wage. In combination with the tax credits this provides a guaranteed minimum income for all working families, although the level of the NMW remains low despite recent increases.

• ‘New Deal’ active labour market measures which increase the job search and training requirements for the non-employed.

• A ‘National Childcare Strategy’ to expand childcare provision, but where most of the places created have been part-time nursery places in school and where pre-school and out-of-school provision is still costly and scarce, notwithstanding some targeted provision in economically deprived areas.

The Child Tax Credit (CTC) was introduced in April 2003 (replacing an earlier scheme called the Children’s Tax Credit which was introduced in April 2001). It provides income-related support for in recognition of the costs of parenthood made through the income tax system for low and middle-income families (household income upper limit = 58 000 pound in 2004). 90% of families with children are eligible for some portion of CTC. In 2004 it is worth up to a maximum of 38 pound/week for one child, 65.70 pound for two children and 93.30 pound for three children. A higher payment is made when a child is under one year old (the baby element), and for families with children with a disability. CTC is paid direct into the bank account of the main carer. The main carer is usually the mother, who also receives Child Benefit, which is the other main form of financial transfer targeted at parents (Child Benefit is a universal cash benefit paid for each dependent child up to 16 years of age, regardless of family income. It has virtually a 100% take-up, and the mother is targeted as the main carer to claim this benefit following the birth of the child).

The Working Tax Credit was also introduced in April 2003. It provides a targeted top-up to wages for those in low-paid jobs via the personal tax system. It replaced the Working Families Tax Credit (introduced in 1999, which in turn had replaced previous in-work credits for low-income families) and extended tax credits to couples without children. The policy objective is to ensure that employment pays more than welfare. It is available to (a) all persons provided they are at least 25 years old and employed for at least 30 hours per week (b) persons with dependent children and/or a disability if they work at least 16 hours per week. The amount a person is eligible for depends on their circumstances (e.g. higher amounts will be available to workers with a disability). The main element of the Working Tax Credit – the targeted top-up to wages – will generally be paid through the pay packet for employees, and direct for the self-employed.

It includes a specific childcare element for those with children in receipt of this credit. To apply for the childcare element, lone parents must work 16 hours or more. Couples can apply if both work 16 hours or more; or one works 16 hours or more per week and the other receives a disability/invalidity benefit. This childcare element is paid direct to the main carer, alongside the Child Tax Credit. It covers up to 70% of eligible childcare costs (registered childcare services) up to a maximum of 135 pound per week for one child (equal to a credit of up to 94.50 pound per week) and 200 pound per week for two or more children (equal to a credit of up to 140 pound per week). The childcare element of the Working Tax Credit is available to families with incomes over the threshold (currently at 5 060 pound), but will be reduced at the rate of 37p for every pound of gross income over the threshold. In this way, the childcare element of the Working Tax Credit payable decreases proportionately as a family’s income increases.

Source: Fagan, Hebson and Rubery (2004)

Page 58

One positive effect is that the reform has enhanced work incentives by raising the net income of low-income families where someone is employed. Secondly, the reform has also achieved higher take-up rates than those achieved for earlier versions of tax credit (Family Credit and Family Income Supplement), which may be partly attributed to the transfer being made through the tax rather than the benefit system, which has helped to reduce the social stigma attached to claiming this form of support. Thirdly, both the Child Tax Credit (CTC) and the Childcare Tax Credit (CCTC) is now paid direct to the main carer and this is a positive transfer of resources towards women, since they constitute the majority of ‘main carers’4. Fourthly, the additional money will reduce the financial pressures on mothers to work, particularly in the first year following child birth, which was presented by the Chancellor as a favourable outcome which gives mothers more choice over when they resume employment, although the actual effects on behaviour are as yet unknown (Brewer and Clark, 2002). The fifth positive effect is that the new tax credits improve incentives for lone parents to enter the labour market. However, the regulations with respect to couple households are less favourable, as discussed below. Finally, the 16 hours threshold for eligibility for WTC and CCTC makes ‘longer’ part-time jobs more attractive than shorter hour part-time jobs. This may encourage the expansion of longer rather than short part-time hour jobs for carers (usually mothers) of young children, which may improve their labour market integration over their lifetime.

Alongside these positive effects, from a gender mainstreaming perspective there are several problems which remain or have been created by the reform. The first set of problems relate to the mechanisms for assistance for childcare costs. The increased help with childcare costs for low-income families remains limited in relation to childcare costs. Childcare costs can only be claimed once in employment, yet the costs of putting childcare arrangements in place before securing employment may be a barrier to job-seeking (although some provision is available for those job-seekers eligible under New Deal measures). Childcare costs can only be claimed by couples if both parents are working for 16 or more hours a week; this means that if in a workless household the woman finds work first, she must rely on her partner or other family members to undertake childcare. Some men may resist this ‘role reversal’, and a more effective way of supporting women’s integration in the labour market would be to make the childcare tax credit available as soon as one member of a couple finds employment, or even prior to employment.

The second set of problems is that resources are targeted at households rather than at individuals, through increased joint assessment and joint ownership of credits, rather than individualisation. Bennett (2002) argues that the reforms reflect a growing emphasis on the couple/household as a policy focus (as a means of targeting resources) which is in tension with the shift towards ‘individualisation’ in other parts of the tax and social protection system5. The emphasis on joint assessment explicitly introduces the notion of joint ownership of credits, which is a new development in UK policy. Bennett (2002) argues that this is preferable to one person owning a benefit on behalf of another adult, but it creates problems. It reinforces a widely held ideal that relationships are stable and involve a consensual pooling of resources, both of which are contradicted by empirical evidence. It undermines the principle of individual taxation because both partners are jointly liable for information on tax returns and overpayments, which create problems in families where relationships are breaking down or separation has recently occurred.

Page 59

The third problem is that analysis conducted by the Institute for Fiscal Studies showed that the household focus of the reform has increased work incentives for primary earners but reduced work incentives for some second earners6. Partly in response to this evidence and lobbying by feminist groups the government introduced an earnings disregard (the first 2 500 pound) to improve the financial incentives for second earners. The magnitude of the negative impact on the behaviour of second earners will be influenced by the effects that the Childcare Tax Credit, for which the empirical evidence is as yet uncertain. In turn, this will depend on how the supply and cost of childcare expands under the National Childcare Strategy in coming years, but the current Childcare Tax Credits arrangements fall quite a long way short of actual childcare costs.

The fourth set of problems is that the system reinforces the notion of a single ‘main earner’ as the normal household model. One way that this happens is that the credit is only available to couples if one person works over 30 hours (couples with combined hours reaching this threshold are not eligible), another is that eligibility for the credit has been extended to include sole earner couples without children. A third important way is that the government has stuck to its aim that the WTC credit will normally be paid via the pay packet of the main earner, while the child-related credits (CTC and CCTC) are paid to the main carer. The government made the concession that families can elect to have the WTC payment made to the carer in response to a long-running campaign by anti-poverty and feminist campaigners since the 1970s that income transfers should be directed through the main carer on the basis that this is the most effective mechanism for the money to reach children and women. However, the government persists in connecting the transfer to the wage/tax system paid default to the main earner (mainly men) as a way of reducing the stigma for claimants as well as making the distribution more politically acceptable as it is presented as a form of tax relief rather than benefit expenditure.

Finally, the fifth problem is the articulation of in-work credits for the low-paid with the wage structure. Rubery (2003a) argues that the government has failed to consider the implications of extending the tax credit system as the danger is that it reinforces a structure of wages where employers operating at the bottom of the labour market rely on a supply of workers who are subsidised directly by the state (tax credits) or via the wage of the ‘main earner’. This neglects the principle of equal pay for equal work. Furthermore, anti-poverty groups argue that the increased resources distributed through the tax credit system are still limited given the prevalence of low income in families.

1.2. General tax reductions for the employed through reform of the personal taxation system

The recent reforms to personal taxation in Luxembourg, Austria and Iceland are examples of fiscal measures which have focussed on reducing the tax rates faced by the employed and where the gender impact has not been addressed by the government in the design or justification of the reform.

In Luxembourg the Personal Income Tax (PIT) reform in 2001 has reduced the tax wedge on all taxpayers. In terms of redistribution the tax reduction was proportionately higher for the higher earning households, but in relation to low-income households tax rates remain low and there have been additional measures to increase support for low-income families through the benefit system (Box 1.5). However, key gender issues were overlooked in this reform (Plasman and Sissoko, 2004). There are few supply-side incentives for married women’s labour market participation in the tax/benefit system, and women’s employment rates are below the EU-25 average. This latest reform did not modify the principle of joint taxation although the ‘conjugal advantage’ of tax splitting has been decreased slightly as a by-product of the reduction in the tax rates and increase in tax disregard. The NAP/Employment does estimate that the reform would increase the participation of under-represented categories and low-income households (including single parents), but there is no explicit discussion of gender.

Page 60

Box 1.5. Luxembourg – Personal tax reform to reduce the already low tax burden, but few supply-side incentives for married women to enter employment

Recent debates on ‘making work pay’ in Luxembourg have mainly concerned the 2001-02 tax reform to reduce tax rates.

The Personal Income Tax (PIT) reform introduced in 2001 was stimulated by a buoyant economy, recent tax reforms carried out in neighbouring countries (Belgium, Germany, France and the Netherlands) and in the context of labour supply shortages. This reduced the tax burden on all taxpayers, but the size of the reduction in percentage terms is higher for higher earning households. The minimum tax rate in Luxembourg is the lowest in Europe, and the policy of ensuring little/no tax for low-income households is to continue. The combination of tax reductions and increases in family benefits is presented as the best way to improve the situation of families with children. The 2001-02 reform led to a substantial further reduction in taxes on families, in the context of an increase in the level and coverage of the minimum guaranteed income (RMG) in recent years. In addition universal family allowance payments have increased, financed by a reduction in tax credits for dependent children.

There are few supply-side incentives for married women’s labour market participation in Luxembourg due to a combination of high wages for men and the structure of the tax/benefit system. The system of joint taxation and tax splitting discourages the labour market participation of the second earner. In the social security system there are survivors’ benefits and pension arrangements linked to the husband’s professional activity which favour a non-employed wife more than one with her own employment record.

Source: Plasman and Sissoko (2004)

The recent Austrian tax reform targets additional relief at low-income families, with additional relief for single-earner households (Box 1.6). One positive impact on women’s situation is that single parent households receive this additional relief as a single-earner. However, the negative effect of this reform is that it increases the way that the system favours ‘single-earner’ rather than ‘dual-earner’ arrangements for couples through the increases in the additional earnings threshold and child supplements for single-earner households. This promotes a specific family model: the single-earner family with a (female) partner who is not employed or works part-time. Furthermore, in practice it is mostly men, and particularly male breadwinners who have benefited most from these recent tax reforms, with 470 000 male breadwinners compared to 65 000 single parents in receipt of these additional forms of tax relief (Mairhuber, 2004).

Box 1.6. Austria – The 2004-05 tax reform reduces tax payments for those with low earnings, with some reductions targeted at low-income families

• The tax exemption threshold has been raised (up to 14 500 euro/year).

• The additional earnings threshold for single-earner (including single parent) households has been raised to 6 000 euro/year.

• Single-earner households have also benefited from a newly introduced targeted child supplement (Kinderzuschlag – 130 euro for the first child, 175 euro for the second child and 220 euro for each additional child).

• Single-earner couples and single parent households already receive an additional tax credit (AlleinerzieherInnen-/AlleinverdienerInnenabsetzbetrag – 364 euro) which converts into a payment to the low-paid when it is not consumed by the tax bill. The single-earner in couples only receives this if the ‘second earner’ earns below an ‘additional-earnings’ threshold. Most of the recipients of this additional tax credit are male single-earners in couple households.

• A tax credit is paid to all households for each child (Kinderabsetzbetrag – 51 euro/month per child). This is paid out with the family allowance to the main carer (and is paid regardless of the actual tax owed by the individual carer).

• Reforms to the family allowance system in recent years have been targeted at low-income families. This benefit is paid for children (up to 18 years or until completion of education/training up to age 26). It is a univer-Page 61sal benefit (i.e. not income-related) paid by the tax office to the main carer, usually the mother. Since 1999 the level of allowance depends on the number and age of the children, with an additional supplement for low-income families with three or more children (36.40 euro/month for the third and subsequent child).

Source: Mairhuber (2004)

In contrast to the situation in Luxembourg and Austria, in Iceland the gender gap in employment rates is much lower due to the high employment rates of women. Yet the gender dimension to recent tax reductions was not debated in this context either (Box 1.7). On one hand, any positive impact on labour supply behaviour from the recent tax reductions is likely to be concentrated on women, notwithstanding that their employment rate and working hours are already high by international standards. However, on the other hand, in Iceland women’s average annual earnings are much lower than men’s (19 300 compared to 32 000), and so they will benefit less from the reduction in the higher tax rate. Instead women (and low-paid men) would benefit more from an increase in the tax exempted income. Furthermore, the tax cut will have to be financed by reducing welfare benefits and/or welfare services, and these cuts will be borne disproportionately by women, for they are the main receivers of benefits and 57% of women’s employment is in the public sector compared to 22% of men’s employment. For example, primary school teachers (the majority are women) were on strike because the local government employers are resisting wage rises on the basis that the local tax income received is too low to cover the costs of social services (Mósesdóttir, 2004). The strike lasted for seven weeks and the dispute was only solved when the government threatened to use legislation to solve the conflict.

Box 1.7. Iceland – Tax reductions are the main element of recent tax/benefit reforms

The ‘making work pay’ debate relates primarily to reducing the tax burden on the employed, rather than on reducing benefit entitlements. The political debate is polarised between the present government (conservative-liberal and central alliance) which stresses reduction in personal income tax to promote general welfare while the opposition places more emphasis upon improving benefits and reducing taxes for those on low-income, for a growing number of people are living in poverty.

The tax burden on Icelandic employees (as a percentage of gross wage earnings) is low in comparison with other EU countries. Taxation is individualised, although couples can opt for joint taxation and pool their personal allowances. There are high personal tax allowances (which amounted to about one third of regular pay for employees in 2003) and only two tax rates, the basic (currently 25.75% state + 12.83% local tax) plus a high rate (+4%), which is only paid by 5.8% of tax payers. Moreover, employees in Iceland do not pay social security contributions on their earnings as in other countries. When the effects of cash benefits such as child benefits are also taken into account, the tax burden (measured as income tax plus employee contributions less cash benefits as % of gross wage) becomes even more favourable for those with children, especially low-income lone parents with two children.

The government proposed reductions in the state and local tax rate in 2003 by around 4 percentage points, but this has been delayed until 2005, and instead it rose slightly from 38.55 to 38.85%. In the meantime, the higher rate was reduced (from 7% to 4% in 2004) which reduced the tax burden for those with the highest income. The likely gender impact of this reform is that:

• Lower tax rates are unlikely to have an impact on men’s labour supply because activity rates and working hours are already high.

• Women’s activity rates are also high by international standards (78.2% in 2002 with average working hours of 33.7) but the tax reductions will enhance financial incentives for women and low earners to increase their labour supply and working hours.

Page 62

Box 1.7. Iceland – Tax reductions are the main element of recent tax/benefit reforms (cont.)

Social expenditure in Iceland in terms of purchasing parity is the lowest among the Nordic countries, and the rates are only lower in Greece, Ireland, Portugal and Spain in the EU-15. There have been some reforms and extension to social protection in recent years.

• Since 2000, the government has raised the level of universal child benefit and reduced the means-testing of this benefit by raising the household income threshold. The Ministry of Finance estimates that this would increase the disposable income of lone parents in the lowest income group by 8% and of low income couples by about 4%.

• In 2005 unemployment benefit rates are due to rise as part of a wider agreement between the social partners.

• The government plans to simplify the social security system and to evaluate its interaction with the tax and pension system; including reducing the means-testing of benefits for disabled persons in order to encourage their participation in employment.

• The main concern of the opposition party is the low level of means-tested social assistance and that social assistance and earnings below the negotiated minimum are taxed. (Denmark is the only other Nordic coun try to tax social assistance, and there the benefit levels are much higher).

Source: Mósesdóttir (2004)

1.3. Reform to unemployment benefit/assistance systems to ‘make work pay’ for the unemployed

There are examples from seven Member States of recent measures to reform unemployment insurance benefit or assistance schemes designed to increase the work incentives for the unemployed through a combination of reduced benefit support and/or tighter job search requirements as a condition for benefit receipt: Denmark, Germany, France, the Netherlands, Austria, Portugal and Ireland. In Norway, the focus of the recent reform is on employment re-integration for claimants of disability pensions, although there are also proposals to improve financial support for those unemployed who are not covered by unemployment insurance and are actively seeking work, and new measures to support the integration of recent immigrants into employment. Finally we also review the situation in Italy, Romania and Bulgaria, where the primary focus of social protection debates for the unemployed is on extending coverage rather than on raising work incentives.

Recent comprehensive reforms: the examples of Denmark, Germany, France and the Netherlands

In Denmark a series of reforms have been introduced since 2003 under the ‘Flere I arbejde’ (‘More people into work’) agreement designed to enhance work incentives for the unemployed (Box 1.8). The focus of the solution is upon reducing financial support for those on social assistance, in the context of the high compensation rate within the unemployment benefit insurance system. Social assistance payments for the unemployed without insurance benefits have been reduced, with an additional reduction in the rate paid for married couples. One positive element is that the individual earnings disregard has been increased for married couples, thus improving the incentives for one partner to maintain or increase their employment. The reform has caused widely-publicised financial difficulties, in particular the loss of assistance with housing costs, and prompted public debate. The minister of employment has reiterated that the solution is for individuals to enter employment (Emerek, 2004).

Page 63

Box 1.8. Denmark – Social assistances for the unemployed reduced to enhance financial work incentives under the ‘More people into work’ reform

Following a government study on the obstacles to labour market entry a series of reforms were advanced in the agreement ‘Flere I arbejde’ (‘More people into work’), which have been gradually introduced since July 2003 and mostly relate to the question of financial incentives for the unemployed in receipt of social assistance. It has the political support of most of the parties in Parliament.

The Danish unemployment benefit has a relatively high compensation rate for unemployment, especially for low-income groups: 90% of former pay up to an absolute maximum (3 205 DKR a week before tax in 2004). This means that compensation is high for low-paid groups (mostly women) and low for high-paid groups (mostly men), and when combined with income-dependent public services such as childcare and housing benefit only a relatively small difference in the disposable income exists between the unemployed and the low-paid employed: 25 percent of persons on social welfare will have little economic gain of getting a low income job (less than 500 DKR a month). Immigrants from non-EU countries form a disproportionately high percentage of this group, and married couples where both are unemployed face a particular trap for there is little incentive for one of them to take employment unless the job is well-paid or their partner also finds employment.

The main initiatives in relation to the most socially vulnerable were:

• A reduction in the level of social assistance after 6 months of support as a proportion of maximum unemployment insurance benefits (from January 2004). The ceiling is 80/60% of the maximum rate of unemployment benefit for married persons and co-habitants with/without dependents and 100/80% for a single person with/without dependents.

• A higher employment supplement for married persons on social assistance (from 2003). Married persons on social assistance who take a job may keep a may keep a larger proportion of the wage before a deduction is made in to the income of their spouse.

• All political parties behind the agreement – with the exception of the Social Democratic party – have agreed to use the money saved from this reform to finance a tax reduction on earnings.

Source: Emerek (2004)

A gender perspective has been lacking from the design and debate of the Danish reform (Emerek 2004). The potential problems for lone parents appear to have been considered, although not discussed explicitly, for the calculations in the document show that it is married couples rather than lone parents who are the main group hit by the ‘inactivity trap’. The particular issues facing immigrant women have been neglected, for while the work incentives for married immigrant women may increase they may be unable to secure employment given their already high unemployment rate. The failure to consider this is in the context of recently introduced low rates of social benefits for new immigrants designed to make Denmark a less attractive destination, particularly for the unskilled. The government has set aside money for a research programme with a focus on evaluation of these and related incentive measures, and Emerek (2004) argues that it is important that both a gender and ethnic impact dimension is built into the evaluation.

In terms of policy efficacy the Danish reform is an example of how attempts to ‘make work pay’ by reducing benefits rather than by other measures may lead to greater poverty and may be more expensive in the long run. Emerek (2004) proposes that an alternative strategy of using the public money saved by the reform to raise the earnings of low-paid workers – for example in the female-dominated public care-sector – rather than for tax cuts might have created a more positive ‘virtuous circle’ of raising financial work incentives for the unemployed. This alternative strategy would bring with it a positive gender impact given that women account for the majority of the low-paid.

In Germany there has been a series of reforms since 2002 which have introduced stricter job search requirements for the unemployed, and the final step in this programme is the ‘Hartz IV’ reform to be implement by January 2005. This new package of measures reduces unemployment benefit and social assistance, tightens job search requirements further, and expands training and job creation programmes for the unemployed. It introduces new explicit definitions that non-employed parents with young children aged 3 years are to bePage 64 considered as available for employment, as well as those with young children if childcare is available, which goes hand-in-hand with new obligations on public employment agencies to expand childcare services (Box 1.9).

Some recipients of social assistance who were previously defined as ‘inactive’ will now be defined as unemployed with the introduction of the ‘ALG II’ definition of eligibility conditions, and the positive effect for this group is that this means they receive higher levels of financial support and better access to active labour market programmes in exchange for tighter obligations to take available employment. There is a regional dimension to the impact of this reform: most of the social assistance recipients who are the potential ‘winners’ from this reform through being reclassified from ‘inactive’ to ‘unemployed’ live in the west of Germany, while many of the current unemployment assistance recipients who will lose financially under this reform live in the east (Maier, 2004).

There are some specific gender impact issues identified in the German national report (Maier, 2004). The mothers who quality for ‘ALG II’ stand to benefit most from the reform, which includes some who were previously defined as ‘inactive’ rather than ‘job-seekers’. This group will benefit from a combination of better access to childcare, training and other active labour market measures, as well as higher levels of social assistance for those previously defined as ‘inactive’. However, a negative change is the tighter aggregation of married and cohabiting partners’ resources for benefit assessment.

This is detrimental for women who were previously in receipt of unemployment assistance based on their previous employment record but who will no longer be eligible for ‘ALG II’ if their partner is employed, and linked to this will not be eligible for the targeted job search, training and childcare help offered by the employment agency. This applies to a substantial proportion of women in East Germany. The reform should also be assessed in the related context of a shift in policy for women ‘returners’. Targeted rights to vocational training and job creation programmes and subsidies for women returners were revoked in 2004, as were similarly targeted measures in labour market policies for other groups of women (older, those with a disability or another special need). Thus women who are ineligible for the ‘ALG II’ and associated employment programmes have no recourse to other targeted ‘returner’ programmes. Maier (2004) also notes that the ‘ALG’ reforms have made it more difficult for migrants to access unemployment assistance and related employment programmes.

Box 1.9. Germany – The ‘Hartz IV’ reform of unemployment benefit insurance and unemployment assistance

A series of reforms since 2002 have introduced a number of measures to reduce unemployment through stricter rules for job availability and job searches in order to increase the pressure on the unemployed to accept available employment or become self-employed. The final step in this programme of reforms is the ‘Hartz IV’ reform to be implemented by January 2005, which includes:

• A reduction in the duration of unemployment insurance benefits.

• A change in the eligibility conditions for unemployment assistance through the introduction of ‘ALG II’ and a reduction in the level of unemployment assistance received (reduction in the maximum level of financial support, reduced protection of pension insurance benefit and tighter means-testing – including taking account of children’s income as well as that of a spouse/partner).

• To receive ALG rather than social assistance individuals have to be available to work at least 3 hours per day, and to participate in job search and training measures through the public employment agency. All recipients are obliged to take paid employment if this is available, and are allowed to keep a slightly better, but still modest part of their benefit to supplement low earnings.

• The introduction of ALG redefines some people in receipt of social assistance as unemployed instead of ‘inactive’ if they meet the job availability criteria. For this group there are improved levels of financial support, inclusion in insurance for pensions, health and care, and better access to active labour market measures (social assistant recipients do not have individual pension and health insurance).

• With regard to parents, the definition of who is available for employment has become more explicit and tied to childcare availability under this new regulation. Parents are not obliged to take jobs if there is a child aged below three years in the household unless adequate childcare is available. Those with older children are obliged to take employment on the basis that each child aged 3 years or older already has a right to childcare in the local community.

Page 65

Box 1.9. Germany – The ‘Hartz IV’ reform of unemployment benefit insurance and unemployment assistance (cont.)

• Employment agencies have a new responsibility to work with local childcare services to expand childcare facilities for the unemployed which fits with the available job opportunities, in a context where local communities are receiving additional support from the federal government to expand childcare services.

• Employment Agencies are being restructured to strengthen the job search services, training and job creation programmes for the unemployed.

• Some of the job creation programmes are very low-paid (so-called ‘1 euro an hour jobs’) schemes in personal services where the job is limited to 30 hours per week for up to 6 months. These jobs will not be covered by social security insurance and incumbents are not entitled to unemployment insurance at the end of the job; they return to unemployment assistance if they do not obtain employment.

Source: Maier (2004)

Similar to the direction of the German reform, the reforms in France (Box 1.10) have also reduced financial support for the unemployed and tightened job search requirements. This accompanies the expansion of financial support for low-paid employees through the new PPE tax credit scheme discussed above (see Box 1.3) so together these measures have enhanced the financial incentives for the unemployed to take low-waged work. The new French ‘return to employment’ allowance (ARE) and the means-tested forms of assistance available to the unemployed (ASS and RMI) can be combined with earnings from short part-time or occasional employment. Earlier studies of the social assistance system have shown that the poverty trap hits more women than men. This gendered impact results from a combination of women’s higher unemployment rates, lower pay rates, predominance among single parents who receive additional benefits and the disincentives for one member of an unemployed couple to take low-paid employment (Silvera, 2004). However, this gender dimension was not explicitly considered by UNIDEC in the development of this reform.

The French benefit reform is accompanied by job creation measures for the long-term unemployed in receipt of means-tested assistance. Whereas some of the job creation programmes being advanced in Germany are low-paid, part-time and of limited duration (see Box 1.9), the new ‘social cohesion plan’ in France proposes to create full-time, longer-term jobs at minimum pay levels which include a substantial training element, and thus offers the prospects of better quality job creation for the unemployed compared to the German ‘1 euro an hour jobs’. However, the gender impact of both measures will depend upon the demographic profile (gender, household situation, etc.) of those who enter these schemes, and the measures that are put in place to ensure equal treatment in these placements.

Box 1.10. France – Unemployment benefits reformed and new employer subsidies for recruiting the unemployed to minimum wage jobs

In 2001 an agreement (PARE/PAPND) was made by UNIDEC1 which reduced the duration of unemployment benefit (the rate remains the same) and introduced the requirement to sign up to a personalised job search plan with the national public employment services (ANPE).

• The duration of this new ‘return to employment’ allowance (ARE) depends on the contribution record, with longer entitlements for the older long-term unemployed, and it is an individual entitlement based on work history, not family situation.

• It can be combined with earnings from employment provided the job is occasional and limited (for less than 136 hours/month and up to a maximum of 70% of previous pay).

• The long-term unemployed who do not receive ARE may be eligible for a means-tested ‘solidarity allowance’ (ASS), and the duration of this was also reduced – to a maximum of 2 years plus a discretionary 91 days extension for those aged under 55 years. There is some aggregation of couples’ resources in the assessment. In contrast to ARE, this payment does take into account family-related breaks into calculating eligibility based on previous work history contributions (5 years employment in the previous 10, reduced for up to 3 years per child if there was a break for raising children).

Page 66

Box 1.10. France – Unemployment benefits reformed and new employer subsidies for recruiting the unemployed to minimum wage jobs (cont.)

At the end of this ASS period one can receive RMI or RMA. RMI is for the economically inactive who do not receive unemployment benefits. It is linked to one’s family situation (presence of a partner and children), and the resources of partners are aggregated for the assessment.

• Both ASS and RMI can be combined with limited earnings from a return to part-time or occasional employment (up to 750 hours/year).

• RMA was introduced in 2003 to encourage a return to work. It is a contract which provides an employer with a subsidy if they recruit a long-term recipient of RMI (2+years) for up to 12 months. The employer receives the RMI plus additional financial support, the employee receives the equivalent of the SMIC for 20 hours (626 euro).

• The new social cohesion plan aims to create a million ‘economic activity contracts’ over 4 years for those currently claiming ASS or RMI benefits. The contract will be for 35 hours per week, including 26 hours minimum paid employment (i.e. 3/4 of the monthly SMIC) and 9 hours training, for two years, renewable once. Local employers will be the managers of this scheme. The employers will receive the equivalent of the benefit plus some additional state aid for this job creation.


1 UNIDEC is the body which administers the French unemployment benefit system and is managed by the social partners

Source: Silvera (2004)

The theme of reduced benefit entitlements and increased job search requirements in order to improve the work incentives of the unemployed is also found in the example of the Netherlands (see Box 1.11 and Box 1.12). Here the provisions of the new ‘Work and Social Assistance’ Act (Box 1.12) is similar to the German ‘Hartz IV’ reform (see Box 1.9 above) in terms of the reduced benefit entitlements and increased job search requirements that have been introduced in conjunction with a shift in emphasis to define parents as ‘job-seekers’ with tighter job search requirement and a restructuring of responsibilities for the local public employment services. In the Netherlands this new act on Work and Social Assistance affects – among others – the situation of lone parents (with at least one child aged below 18 years), who accounted for 26% of all social assistance recipients in 2003, and nearly all are lone mothers (96.3%). All lone parents in receipt of social assistance are now obliged to fulfil the new, tighter job search requirements unless an individual exemption is granted, whereas previously they were exempt from this if they were caring for a young child under 5 years old. Under the old system 35% of lone parents were automatically exempt from job-seeker requirements because they had a child under 5 years old, and some lone parents with older children to care for were also exempted. While lone parents used to have a lower exit rate from social assistance compared to the outflow of single persons or couples this discrepancy has diminished because the exit rate for lone parents has remained stable while that of single persons and couples has fallen (Plantenga and Remery, 2004).

Page 67

Box 1.11. The Netherlands – ‘Making work pay’ through tax/benefit reform

In the Netherlands, the welfare state is under attack from criticisms that it is too expensive, too complex, too generous and too passive, even though unemployment remains low (but rising) by international standards. Inactivity rates are also falling, although a comparatively high proportion of the inactive are on disability benefits (9%). Around one quarter of the inactive hold a small job of less than 12 hours/week (the rate is similar for both sexes), which are not counted in the Dutch definition of the labour force.

The Dutch government has identified several traps: the poverty trap (people on social assistance), the ‘flow’ trap (people on special labour market programmes who lose benefits if they accept a job), the unemployment trap (unemployed) and the (women) returner trap (high costs of childcare, which is calculated on the basis of household income, and which means the additional net income may be small in the context of majority of women taking part-time employment).

A series of policy measures have been introduced to increase the financial attractiveness of paid work while at the same time making the social security system less attractive by reducing benefit rates and duration and introducing stricter eligibility assessments:

• In 2001 a tax credit for employed persons was introduced to make paid work more attractive in line with EU recommendations, which has been raised annually.

• The duration of non means-tested unemployment benefits has been reduced and tighter eligibility conditions based on actual work experience have been introduced. The care credit for determining work experience is also likely to change – currently periods of care for children up to the age of 12 are partly taken into account when determining work experiences, this is likely to be lowered to 4 years.

• The social assistance system has been reformed to encourage people to re-enter employment (see separate Box 1.12 for further details).

• The disability scheme (WAO) is to be reformed quite drastically after years of fierce debate. Assessments will be more stringent and limited. There will be more emphasis on adapting the work environment for those who have a limited work capacity rather than on income compensation.

• New proposals to stimulate labour market participation of older workers are to abolish tax facilities for early retirement schemes.

Source: Plantenga and Remery (2004)

There has been no explicit GM or GIA of the reform to the Dutch social assistance system, but there is some attention to the impact on lone parents, with a focus on their particular work-family reconciliation needs and aspects such as motivation (Plantenga and Remery, 2004). Most of these needs are meant to be covered by general policy measures for parental leave and care facilities, in conjunction with some other general measures such as government subsidised experiments in ‘daily routine’ designed to make it easier for people to combine work and care tasks, where about 20% of end-users are lone parents. However, there are some additional, specific measures for lone parents’ childcare needs following Parliamentary debate. Thus the childcare measure (KOA) for lone parents will continue until a new act on childcare comes into force on 1 January 2005. This new act will provide financial support in a new more demand-driven system of childcare financing through tax credits for children for employed persons, with extra provision for lone parents. Municipalities will also be encouraged to stimulate the expansion of part-time work and will have access to budgets to promote the development of flexible childcare.

Though the new Dutch Act is stricter with respect to the job search obligations, there is some room for tailor-made solutions which have the potential to better address individual needs. There is also some discretion available to officials to exempt lone parents through reference to childcare availability and preferences. This discretion might be used to respect the particular circumstances and preferences of individual lone parents, but it might also be used primarily to meet other operational priorities. Thus, in the current economic situation of rising (although still low) unemployment, municipalities may be inclined to use this policy discretion to focus efforts on other claimants rather than lone parents. This may occur particularly if it is more expensive to help lone parents securePage 68 employment, thus limiting the scope for generating budget surplus.

Box 1.12. The new Dutch ‘Work and Social Assistance’ Act (Wet Werk en Bijstand, WWB) came into force on 1 January 2004

This replaces the General Act on Social Assistance (Algemene Bijstandswet) and some other related acts on subsidised work. The aim is to encourage persons on social assistance to re-enter employment through introducing sharper benefit rights and job search duties and more individualised and ‘customised’ solutions for claimants than the previous schemes. The explicit starting point is that every citizen is expected to provide for his/her own living, with employment as an important means for doing so.

• The level of benefit has not been reduced but the provision of additional specific allowances for different categories have been abolished and may only be provided on an individual, discretionary basis and local income policies are no longer allowed. This reform of the benefit is designed to reduce the poverty trap.

• The job search requirements have been made stricter – benefit claimants are obliged to co-operate in job search activities and to accept ‘generally acceptable’ employment and no one is automatically exempt from this obligation. Previously people were automatically exempt if they were older than 57.5 years or were a lone parent with either a child under the age of 5 years or an older child under the age of 12 where there was no childcare available. Exemptions are now only considered on an individual basis.

• In the Parliamentary debate there was much objection to the new job search requirements for lone parents, so the compromise reached was that municipalities must take into account childcare arrangements and the preference of parents with regard to care arrangements where developing the plan for particular claimants.

• The administrative responsibilities have also been changed and decentralised to municipal level, which now receive a fixed budget for benefits and one for re-integration services. The latter may be spent on training, subsidies for labour costs or jobs, and mediation services. As an incentive to help place people in employment, any surplus in the benefits budget arising from a successful re-integration may be kept by the municipality and spent at its discretion.

• Accompanying this reform the ‘Innovation program work and social assistance’ has been formulated by the co-operation of the Dutch organisation of municipalities (VNG) and the ‘Dutch national association of managers of municipal services in the fields of work, income and social welfare’ (DIVOSA). The program and its budget supports municipalities with policy development and innovation, for example policies to facilitate the combination of work and care.

Source: Plantenga and Remery (2004)

The increased job search requirements for a parent caring for young children while claiming social assistance in Germany and the Netherlands is part of a more general reform trend found across Europe targeted particularly at lone parents in welfare systems where this category of claimants had previously been defined as inactive and exempt from job-seeking requirements. A similar shift in policy presumption has occurred in the UK (Lewis, 2001) and in Norway (see Box 1.17 below). In the UK this policy shift has been followed by a rise in employment rates for lone parents and the UK government’s NAP/Social Inclusion 2003-5 reports the findings from evaluations which conclude that the ‘New Deal’ active labour market measures for lone parents have played a role in this development. However, the Norwegian national report notes that studies suggest that single parents’ participation in employment is strongly influenced by fluctuations in labour market opportunities, which affect the outcomes of changing policy regulations and benefit incentives (Ellingsæter, 2004).

Reforms with a narrower or more incremental focus on the unemployed: the examples of Austria, Portugal and Ireland

The reforms to enhance work incentives for the unemployed in Austria (Box 1.13), Portugal (Box 1.14) and Ireland (Box 1.15) have been more incremental and narrower in focus that the Danish, German, French and Dutch reforms.

In Austria the unemployment benefit insurance has been reduced and the qualifying period extended, but an expansionary element of the reform is thatPage 69 benefits for young and older workers have been enhanced if they participate in a training programme. The Austrian report notes that gender mainstreaming is missing in the conceptualisation of this policy, as is the case in many areas of policy formulation, despite some basic gender-differentiated impacts (Mairhuber, 2004). Thus the longer qualifying period for unemployment benefit will impact on many women with temporary or insecure contracts – such as those employed in tourism – as well as men in the male-dominated construction sector. Many of the women who do not qualify for unemployment insurance are unlikely to receive the means-tested unemployment assistance because their partner’s income is taken into account. The predominantly female category of lone parents is one of the low-income groups that will be hit by the reduction in unemployment benefit rates. It may also be the case that the introduction of longer periods of benefit support for younger and older workers while on active labour market measures may be of less benefit for women than for men because women are less likely to qualify for unemployment insurance benefits, particularly among the older cohort.

Box 1.13. Austria – Reforms to the tax/benefit system are designed to enhance work incentives

There has not been a public policy debate about ‘making work pay’ but several reforms have been made to the tax/benefit system in recent years which are designed to enhance work incentives. The problem of high tax rates for low-income families which are above the EU average was noted in the NAP/Employment in 2003.

• Unemployment benefit insurance has been reformed to tighten eligibility criteria and reduce benefit rates: The earnings replacement rates have been reduced, accompanied by a 40% reduction in family supplements for dependent family members, and the qualifying contribution period has been extended from 26 to 28 weeks.

• Enhanced benefits for younger and older workers on labour market programmes: The duration of unemployment benefit has been extended for claimants who participate in Public Employment Services training and re-integration labour market programmes (open to younger and older people aged under 25 years or over 50 years).

• The tax burden on households has been reduced through the 2004-05 tax reform, with some reductions targeted at low income families (see Box 1.6 above).

Source: Mairhuber (2004)

In Portugal there is some public debate about tax/benefit reform because the marginal effective tax rate for those leaving benefits and entering employment is among the highest in the EU, creating a stark unemployment trap for those who receive benefits (Ferreira, 2004). However, the interaction of low benefits with low wage rates in conjunction with already low tax rates in the personal tax system for the low-paid makes this problem difficult to address through the tax/benefit system (cash family benefits per person are among the lowest in Europe when expressed in Purchasing Power Standards). Reforms in 2004 reduced the eligibility requirements for receipt of unemployment insurance benefits in conjunction with introducing tighter job search requirements, in line with the direction taken in the reform of social assistance in 2002 (Box 1.14). To enhance the financial work incentives of the unemployed there is a benefit paid to those who take a low-wage (including part-time) job, and provision for assistance with childcare expenses has also been introduced, although this has not been widely implemented to date.

Page 70

Box 1.14. Portugal – Benefit reform for the unemployed and proposals to extend household-based means-testing

There have been a number of recent policy initiatives to modernise the tax/benefit system with a view to removing barriers and financial disincentives, although much of the official discourse is about ‘combating fraud and making the system more equitable’ rather than an explicit discussion of ‘making work pay’. In 2003 some measures were launched to make employment more attractive and to alleviate social expenditure, the most important of which are:

Reform to social protection for the unemployed to extend coverage while also introducing more stringent job search requirements (March 2004): The qualifying period for eligibility has been reduced but to continue to receive benefit claimants are now expected to be able to commute further for employment, must accept the first suitable job offer (previously it was possible to refuse up to three) and attend the employment centre regularly for appointments. The tighter job search requirements build on the same principle applied in the social assistance system when the Guaranteed Minimum Income (GMI) was replaced by the Social Insertion Income (SII) in 2002, with the main objective being to strengthen the activation principle of beneficiaries through more individualised plans of support. The take-up rate of SII has been much lower than GMI to date.

Employment-conditional benefits for the unemployed (March 2004): A premium of 25-35% has been introduced for claimants who take a low-wage (including part-time) job (this is in addition to the existing subsidy for employers who hire an unemployed person).

Assistance with childcare expenses (March 2004) has also been introduced, but this measure has not been widely implemented.1

Reduced sickness benefit for short absences (March 2004): New regulations designed to combat fraud.

In addition to reforms which have been implemented, proposals to extend household means-testing into the social protection system introduce the concept of the ‘family manager’ (gestor familiar), even though this appears to be a re-labelling of the notion of the ‘head of the family’, which has been outlawed since the first Civil Code in 1978.


1 This conclusion is reached by the national expert following discussions with officials from the public employment services.

Source: Ferreira (2004)

A worrying development in Portuguese reform from a gender equality perspective is the current emphasis on means-testing, on taking the household as the unit of assessment. This will tend to reinforce the traditional family model of a ‘household head’ and ‘second’ earner. Yet this has not received much attention in the public debate, instead the focus has been on how the complexity of the means-testing will exacerbate existing perverse effects given widespread tax evasion in Portugal. Gender equality issues – such as public funding of childcare services or individualisation of personal taxation – rather than distributing social assistance and tax relief to the ‘household manager’ – are not really up for debate.

In Ireland the elements of the unemployment and poverty traps are well-known, and although some ameliorating measures were introduced in 1997, the improvements have been undermined by the subsequent failure to uprate benefit rates and thresholds and by the abolition in 2004 of the transitional payment of One Parent Family Benefit during the first year of employment (Box 1.15). There have been two recent and positive measures introduced which allow recipients to retain part of their benefit while making the transition to employment or into education – the Back to Work Allowances (BTWA) and the Back to Education Allowance (BTEA) However, these positive initiatives have been restricted by the 2003 budget measure which tightened eligibility to the longer-term unemployed, thus creating new unemployment and welfare traps (Barry et al., 2004).

Page 71

Box 1.15. Ireland – Longstanding issues identified in the National Anti-Poverty Strategy to remove unemployment and poverty traps so as to ‘make work pay’

In Ireland the context to ‘making work pay’ debates is one of high rates of economic growth, falling unemployment and a political environment in which the government has been committed to job growth as the key tool for addressing poverty and social exclusion. Public policy has focussed on creating a low tax environment combined with investment in physical and productive infrastructure, and active labour market programmes for the unemployed. Women’s employment rates have risen rapidly in the context of economic growth and public and private service sector expansion, but remain largely concentrated in low-wage sectors (retail, hotel, personal services). Absolute poverty levels have declined, but relative poverty has risen. The numbers of working poor have increased, as has the level of immigrant labour. There has been an expansion of immigrant labour, which is a new phenomenon in Ireland, which traditionally has been an exporter of labour.

Successive governments have made commitments to raise welfare payments to one-third of the average industrial wage, but these commitments have not materialised. Instead, the recent slowdown in economic growth led to some public expenditure retrenchment and tightening of social protection provision in 2002-03, even though there were reduced numbers of the unemployed to provide for.

In Ireland the National Anti-Poverty Strategy has emerged from an active and developed public debate between the social actors and statutory agencies. This debate has highlighted some key issues in relation to ‘making work pay’ for there are heavy financial disincentives in the transition from welfare to low-paid employment:

• Significant secondary benefits are lost (medical cards for free health care, rent assistance, school clothing and footwear allowances, fuel allowances). A partial redress of this disincentive was made in a 1997 reform which allowed those who had been unemployed for at least a year to retain their medical card for three years after taking up employment, regardless of income (but this excludes the short-term unemployed and those who are entering from inactivity but have failed to meet the eligibility criteria to be defined as unemployed).

• The system of tapered reduction of welfare benefits introduced in 1997 is inadequate because the overall income threshold for retention of secondary benefits has not been uprated since 1994. Once earnings cross this threshold benefits are lost, so low-wage jobs incur a substantial loss of income.

• The low earnings threshold for tapered benefit withdrawal for one parent families (One Parent Families Benefit) has not been uprated since 1997. Furthermore, the policy of gradual withdrawal has been reversed; previously an OPFB recipient could continue to receive half of their payment above this threshold for 12 months to ease transition from welfare onto earnings, but this was abolished in 2004. This means that lone parents typically seek part-time and low-paid employment since it is rarely possible to obtain wages significantly above benefit levels.

• The Family Income Supplement payment for low-paid working families is inadequate. The upper threshold has not been uprated to keep pace with increases in earnings. The take-up is low due to reasons of stigma and complexity, so the Combat Poverty Agency has suggested that a tax-based system of credits akin to the UK system might improve take-up.

There have been recent positive reforms to promote the transition from non-employment into employment or education:

• The Back to Work Allowance (BTWA) introduced in 2000 allows people to keep social welfare payments on a tapered basis on taking up employment. This was identified by the European Employment Taskforce as a progressive scheme for supporting this transition.

• The Back to Education Allowance (BTEA) allows benefit recipients to pursue education with a standard payment rate which is not means tested and to keep any secondary benefits already held.

However, these positive initiatives have been restricted by the 2003 budget measure which tightened eligibility for the longer-term unemployed, thus creating new unemployment and welfare traps.

Source: Barry, Conlon & O’Connor (2004)

Page 72

Box 1.16. The household-based emphasis in the tax/benefit system in Ireland

Among the unemployed:

Many unemployed women with a partner are categorised as ‘inactive’ dependents rather than registered as unemployed because they do not meet the job search criteria of being available for full-time work. This has a number of negative impacts on their work incentives and their social protection entitlements.

• They are under-represented among those registered as long-term unemployed.

• Dependents have limited eligibility for certain labour market programmes, including the recently introduced BTWA and BTEA.

• The household-based means-test for benefits – including the important secondary benefits (medical cards for free health care, rent assistance, school clothing and footwear allowances, fuel allowances) creates a disincentive for the second earner (typically women) unless they can earn a sufficiently high wage to lift the family into a higher income bracket. The result is that many women take up low-paid, part-time and often informal employment to ensure the household retains benefits.

• A new disincentive has been created by a recent change in eligibility rules for rent assistance – from January 2004 as soon as one person is employed full-time the household loses rent assistance (‘rent supplement’) regardless of household income. This creates a disincentive for a transition from ‘no earner’ to a ‘single earner’ household, although for couples this may mean that if one member takes employment then there is no longer any rent assistance to be lost by the second entering employment as well.

• The exception is that there is some targeted provision for lone parents in receipt of One Parent Family Payment, including more flexible eligibility criteria for the BTWA and BTEA. This indicates some policy sensitivity to the specific situation of some groups of women, although this is not developed with a specific and wider gender perspective on social protection.

In the personal tax system:

• Incremental tax reform in recent years has made some progress towards tax individualisation for married women but full individualisation has not been introduced. Evaluation studies show that individualisation of personal taxation has a more positive impact on married women’s labour supply than cuts in tax rates, although overall tax reform has made only a small contribution to the large increase in women’s participation achieved over the last 20 years (an estimated 3% out of the overall 30% rise in women’s participation rates).

In the care system:

• Despite some improvements in recent years the system of child, elderly and other care services and leave provisions remains underdeveloped and is a major obstacle to employment for those with care responsibilities. A positive development is that non-means tested child income support (Child Benefit) has increased significantly in value in recent years.

• The carers’ benefits and allowances are organised on a ‘male breadwinner’ model of family life. They are only available to those carers who do not work/train more than 10 hours a week, and the allowance is means-tested on household income. These rules means many are prevented from qualifying because their household income is too high, and the value of the payment is directly affected by the spouse’s income rather than the degree of care work involved.

Source: Barry, Conlon & O’Connor (2004)

In Ireland, the gender impact is rarely discussed in relation to the measures proposed to reduce the poverty and unemployment traps, with the exception of some recognition that lone parents are predominantly women (Barry et al., 2004). Thus, for example there are more flexible eligibility requirements for lone parents to access the BTWA and the BTEA, but the impact of the household-based welfare emphasis which consigns women with partners to the category of ‘dependents’ is not addressed. Yet this fundamental presumption in the tax/benefit system in Ireland produces systematic barriers to women’s access to labour market programmes and financial disincentives for entering employment (Box 1.16). Where there are reforms with the explicit aim of increasing women’s employment these usually involve contradictions with other policy areas which undermine the efficacy of the reform. For example, the positive developments towards tax individualisation sit alongside persistently low levels of public childcare provision. Similarly, the Equality Authority has highlighted the contradictionPage 73 that employment policy is oriented to raising participation rates for all groups, including carers, yet health policy is premised on the continued availability of informal carers.

Reforms targeting employment (re-)integration of sick and disabled persons: the example of Norway

Employment rates are high for both sexes in Norway. The main strategy for integrating the non-employed into employment is active labour market programmes, rather than a focus on financial work incentives, for there is a large difference in income levels between those in and out of employment, even if rates of poverty are comparatively low (Box 1.17). Several elements of the social protection system are currently under reform in response to the financial pressures created by an ageing workforce, where the main rationale is to enhance labour supply incentives. A particular target is the non-employed in receipt of sickness and disability pensions. Tighter eligibility conditions and reduced rates of disability benefit (WAO) have also been reformed in the Netherlands to try and redress the high proportion of the working age population in receipt of these benefits (Box 1.11 above).

In January 2004 provisions for new claimants of disability pensions identified those with the potential to return to the labour market, and who will receive a lower benefit rate for a fixed period in combination with individual follow-ups to encourage resumption of employment. This follows in the steps of a widespread workplace agreement made in 2001 with the aims of reversing the rise in sickness claims and disability pensions and promoting a more inclusive workplace for workers who are older or have special needs for work adjustments (Box 1.18).

Box 1.17. Norway – Reforms to unemployment benefits and disability pensions to promote employment re-integration

Like the other Nordic countries, Norway has high employment rates for both sexes and waged work is the main pillar of Norwegian social policy, and to achieve this employment and social policy are integrated. There is a large difference in income levels between those in and out of employment. Rates of poverty are comparatively low: in 2000, 2.7% of households had incomes below 50% of median income, rising to 5.4% below 60% of median income. Between 1.7% and 2.4% of children lived in households with incomes below 50% of median income in the period 1997-99. Immigrants are over-represented among the poor and child poverty is a special concern of the ‘Action Plan for combating poverty, 2002-05’, under which child supplements have been increased for those on social assistance or pensions (mainly disability).

A greater policy emphasis on activation and the duty to take employment was developed in the 1990s by the introduction of the so-called ‘work line’ to enhance the labour market participation of certain groups (single providers, the young unemployed and the disabled). Moreover, employment and the social networks that come with it are considered as a positive good that people should be involved in. The normative shift in the 1990s reform was from a ‘right to work’ towards a ‘duty to work’, but few workfare style arrangements were introduced, with the exception that young people in receipt of social assistance can be required to work for the municipality which administers the benefit. Moreover, a premise of Norwegian policy is that employment is a positive state for non-financial reasons as well as financial ones, in particular integration into social networks and public life.

• The main strategy to integrate the non-employed into employment is active labour market programmes, and the main target groups are immigrants and refugees, long-term recipients of social assistance, young people at risk, single parents and recipients of disability pensions.

• With regard to single parents, a 1998 reform increased the job search/training requirements for this group. Prior to the reform single parents could claim benefits for up to 10 years without being required to seek employment. The reform restricted the ‘transitional allowances’ mainly to those with children under 3 years old, while those with a child aged 4 years or older could receive a full or reduced allowance on the condition that they were enrolled in education, worked part-time or were actively seeking employment.

Several elements of the social protection system are now under reform in response to the pressures of a shrinking workforce, an ageing population and rising public expenditure on early retirement, disability pensions and long-term sickness (Norway has one of the most generous sick leave arrangements, with a 100% replacement rate from the first day of sick leave). One of the main rationales for these reforms is to enhance labour supply incentives.

Page 74

Box 1.17. Norway – Reforms to unemployment benefits and disability pensions to promote employment re-integration (cont.)

• Reform of disability pensions to promote employment re-integration

Since January 2004 new claimants of disability pensions are divided into two groups: those considered permanently disabled and those with the potential to return to the labour market after temporary receipt of the benefit for 1-4 years at a lower rate. The latter group are followed up individually to encourage their return to employment.

• New unemployment benefit proposed to extend eligibility hand-in-hand with increased job search requirements

A government working group has proposed that a new unemployment benefit be introduced to encourage job searches while providing basic income security. The aim is to extend the coverage, for currently 41% of the registered unemployed are not eligible for unemployment insurance benefit – primarily newcomers (recent immigrants), persons re-entering the labour market and those in temporary or part-time work – with women less likely than men to be eligible for unemployment insurance benefits (44% of women compared to 39% of men). The proposal is to introduce a benefit for all those aged 19-66 years who have been registered unemployed for at least three months. The benefit period is two years, and the amount is twice the basic amount of social assistance (‘2G’, roughly 13 700 euro). Beneficiaries may be required to produce evidence of their job search activities to be eligible for this benefit.

Source: Ellingsæter (2004)

Box 1.18. Norway – ‘A more inclusive working life’ (the IA-agreement between the Norwegian government and the social partners in 2001)

The agreement runs from October 2001 to December 2005, covers workplaces which include about 43% of the employed, and two thirds of the workplaces covered are in the private sector. The agreement has three aims:

• To achieve a more inclusive working life to benefit the individual worker, the workplace and society.

• To reverse the growth of sickness absence and disability pensions (target: to reduce sickness absence by 20%).

• To better utilise the resources of elderly workers and persons with special needs for work adjustments (targets: to increase the actual age of retirement, and to employ more people with reduced work ability).

Companies join the agreement by signing an agreement with the Regional Social Security working life centres, which allows them to introduce particular reforms, such as more flexibility in the operating of social leave regulations. If the goals of the agreement are not achieved more punitive measures may be introduced later.

Source: Ellingsæter (2004)

The gender impact of these initiatives has not been assessed, which is common for most policy debates in Norway, where gender issues usually only enter debates indirectly in relation to particular groups of women, such as single parents (Ellingsæter, 2004). The national expert argues that a positive interpretation of this lack of gender mainstreaming is that it reflects a general sense that gender equity is largely achieved and women are integrated in working life, as indicated by the low gender differences in employment rates and qualification levels. However, there are still pronounced gender differences in working hours and earnings, and women still represent a significant part of the non-employed, so this silence on gender is problematic.

In fact there is a gender-differentiated impact of these reforms. Women are more likely to be the recipients of the new temporary disability pensions paid at a reduced rate, for they more often have medical diagnoses that make the prospect of labour market re-entry more likely (muscle/skeletal problems and mental illness), and preliminary results show that women account for 70% of those receiving the temporary benefits. The gender impact of the implementation of the workplace agreement ‘a more inclusive working life’ is as yet unknown. However, what is known is that women have a higher level of sickness absence than men, yet gender specific problems do not occupy any visible place in the agreement, with the exception of pregnant workers (almost 7% of women aged 19-45 years are pregnant at any time). Here the only device offered is to reduce employer’s contributions to sickness payment if job adjustment is not possible, which might create a disincentive for employers to establish inclusive strategies of job adjustment (Ellingsæter, 2004).

Page 75

Reforms and proposals which focus on extending social protection coverage for the unemployed and inactive: the examples of Norway, Italy, Bulgaria, Romania and Lichtenstein

Most of the reforms discussed so far in this Section 1.3 focus on ‘making work pay’ for the unemployed by introducing stricter job search criteria, reduced benefits and/or enhanced benefits for those making the transition into employment. However, there are also examples of reforms and proposals where the main emphasis is upon extending coverage.

In Norway, the social protection reforms to increase labour supply incentives and promote employment re-integration on one hand include the introduction of a reduced disability pensions for some claimants, but on the other hand, this has proceeded alongside proposals and initiatives to extend social protection and promote job searches among the unemployed and recent immigrants. A Norwegian governmental working group has proposed introducing a new non-means tested unemployment benefit with the aim that this will extend basic income security while encouraging job searches. The benefit proposed is a non means-tested benefit made at twice the basic social assistance rate to be paid to the unemployed who are not eligible for unemployment insurance benefit who satisfy job search requirements (see Box 1.17 above). This proposal has basic features in common with some parts of the German reform discussed earlier, in that the proposal is that some social assistant recipients will now be entitled to a higher benefit provided they fulfil certain job search criteria. However, in contrast to the thrust of the German reform, the Norwegian proposals are made in the context of maintaining existing levels of social assistance and unemployment insurance benefits, rather than benefit reductions. From a gender perspective this reform will benefit slightly more women than men, for among the registered unemployed fewer women are eligible for unemployment insurance benefit than are men (44% compared to 39%). There is also a new initiative in Norway to integrate immigrant ethnic minorities into working life through a training programme and a benefit which is twice the basic rate of social assistance, and equivalent to the amount proposed for the new unemployment benefit (Box 1.19). This scheme is also likely to be of particular benefit for women, whose needs are seldom addressed despite their lower employment rates compared to immigrant men.

In those countries with limited social protection coverage for the working age population the emphasis of policy debates and reform is largely centred on extending social protection, as is the case in Greece and Italy (Karamessini, 2004; Villa, 2004). A similar situation of limited social protection coverage also applies to many of the new Member States, as indicated in their 2004 NAP/Inclusion reports and in Bulgaria and Romania in their preparation for joining the EU. In Bulgaria and Romania there is no sustained debate about ‘making work pay’; instead in the preparation for EU membership the focus is on preparing the legal and institutional processes and developing economic and social policy in line with EU guidelines and requirements. Here the main issue is about extending social protection and introducing labour market programmes to enhance employability (Boxes 1.20 and 1.21).

Box 1.19. Norway – A new initiative to integrate immigrant ethnic minorities into working life

The integration of ethnic minority immigrants has been a topic of much public and political debate in recent years, with various education, training and anti-discrimination measures introduced. However, at the workplace level in practice the social partners have devoted less efforts to developing initiatives in this area than for older workers, for example.

From 1 September 2004 all Norwegian municipalities are responsible for an introductory training programme for all newly arrived immigrants, who will receive an annual amount broadly in line with the ‘2G’ rate of twice the basic amount of social assistance proposed for the new unemployment benefit. Participants will receive language training, education about Norwegian society and training in basic qualifications if needed.

This scheme is likely to be of particular benefit for women, whose needs are seldom addressed despite their lower employment rates compared to immigrant men. However, persons immigrating due to family reunification are exempted from this programme. This applies to women in particular, as there is a growing proportion of Norwegian men who marry non-western women.

Source: Ellingsæter (2004)

Page 76

Box 1.20. Bulgaria – There is no sustained debate about ‘making work pay’ through tax/benefit reform in the preparation for EU membership

In the preparation for EU membership the focus is on preparing the legal and institutional processes and developing economic and social policy in line with EU guidelines and requirements, including the development of gender mainstreaming policy processes.

Low benefit rates and high unemployment mean that incentives in the tax/benefit system is not the issue, rather the dilemma people face is how to find and maintain employment.

Women account for just over half of the registered unemployed, and in recognition of this a number of programmes have been initiated by the Ministry of Labour and Social Policy targeted at enhancing women’s work incentives and employment re-integration. These programmes are designed to support and promote job searches, to reconcile employment with family responsibilities, to provide access to training and to promote entrepreneurship, including business start-ups in relation to childcare services.

However, most of the firms in Bulgaria are small (56% of employment is in the SME sector in micro-firms) and a number of studies show that employers avoid hiring women who have children or are likely to in the future because of the fear of disruption and costs if the children are ill.

Source: Beleva (2004)

Box 1.21. In Romania a number of reforms have been implemented in relation to the EES guideline 8 measures to ‘make work pay’ and the National Anti-Poverty and Social Inclusion Plan

• The social assistance system was reformed in 2001 to provide a guaranteed minimum income and increased administrative support, with compulsory community job placements for those in receipt of the GMI who are able to work.

• Minimum wage rates are being gradually increased.

• The unemployment benefit system was reformed in 2002. This is paid at a fixed level equivalent to 75% of the minimum gross wage. The duration of the benefit is related to the contribution record, and ranges from 6 months up to 12 months. A number of labour market programmes have been developed to enhance employability.

• Employee and employer social insurance contribution rates were reduced in 2003 to reduce the tax wedge on earnings and labour costs.

• In 2004 a range of tax allowances were introduced to encourage people to contribute to optional pension schemes, private health insurance and for home insurance and insulation improvements.

Source: Zamfir (2004)

In Italy a reform has been proposed which would extend social protection and benefit coverage to a higher proportion of the unemployed (Box 1.22). The reform would be particularly beneficial to women since they constitute the majority of the unemployed. However, the reform has been repeatedly postponed on the rationale of budgetary restrictions. The Italian expert argues that while the gender impact has not been an explicit part of the debate, the fact that the reform is particularly beneficial for women may be part of the reason why it has less priority attached to it (Villa, 2004). While the Italian government has expressed the need to reform the labour market and the welfare system together in the White Paper on the Welfare State (2003), there has been little advance on the latter in terms of how to promote security alongside flexibility. The implicit assumption is that labour market flexibility will increase the number and range of employment opportunities, so that benefit reform can be postponed to ‘better times’. In connection with this the regulation of part-time work has been further modified (Law 30/2003 and Decree 276/2003) with the explicit aim of encouraging employers to create part-time jobs and thus encourage unemployed and inactive women to enter employment. The reform contains no financial incentives for employers to promote part-time work, rather it introduces regulatory changes to make part-time arrangements more flexible for employers. Consequently, the ‘family’ remains the key actor as the solution for welfare, as is stressedPage 77 in numerous official documents. Yet this sits alongside government recognition that the family as an institution is beset with a number of problems concerning its sustainability – low fertility, the lack of childcare services, high costs of raising children, reconciliation difficulties and poor access to good employment.

Box 1.22. Italy – Extension of social protection for the unemployed is postponed again

Poverty in Italy is regionally concentrated (two thirds of poor households are in the Mezzogiorno) and concentrated on families with children.

Similar to the situation in Greece, the focus of concern in the social protection system is not so much about ‘making work pay’ as about extending protection through implementing a benefit system reform (‘riforma degli ammortizzatori sociali’) and introducing measures to combat poverty.

The proposed reform should guarantee broader coverage and eliminate differences of treatment between categories of workers while also encouraging participation in employment. Such a reform would be particularly beneficial to women since they constitute the majority of the unemployed. This reform is more urgent than ever given the recent changes in labour market flexibility introduced by the Legge Biagi. The benefit reform was announced in 2002 with a pledge to introduce it in 2003, yet it was then postponed in 2003 and 2004 and given the current budgetary restrictions it is very unlikely to be implemented in the near future.

Source: Villa (2004)

This focus on the ‘family’ lacks a gender mainstreaming perspective, yet a greater gender awareness might yield a more effective focus for designing policies to support families and tackle poverty and social exclusion. For example, the Italian report (Villa 2004) notes that while there has been recent renewed public and academic debate about poverty and the ‘working poor’ in Italy, the emphasis in the debate is upon how to measure relative poverty rather than understanding the dynamics. The potential role of female employment in reducing the risk of household poverty is missing from debates.

Finally, the extension of social protection coverage rather than an emphasis on ‘making work pay’ is also the focus of reform in Lichtenstein, in the context of an affluent economy (Box 1.23). Here there have been several recent reforms targeted at improving the situation of women that focus on their family roles. Thus family benefits have been extended and a supplement introduced for single parents and pension rights have been improved for married women. However, the thrust of social security reform to redistribute resources towards women is built upon maintaining a traditional ‘male breadwinner’ gender arrangement and supporting women who exit the labour market for domestic reasons, rather than on promoting women’s employment integration. This is reinforced by the limited development of work-family reconciliation policies, where some recent, but modest improvements have been made – there is a new right to unpaid parental leave and childcare has expanded but remains expensive. Similarly, anti-discrimination legislation is very recent and remains to be implemented effectively.

Box 1.23. Lichtenstein – The buoyant economy means there is little pressure to stimulate debate for tax/benefit reform to ‘make work pay’, however employment rates for women are low

The principality is affluent, with high employment and income levels and a large number of cross-border commuters (45% of all employees – mainly from Austria, Germany and Switzerland). Unemployment and the problem of the ‘working poor’ is largely absent, although as elsewhere particularly life situations – lone parenthood, families with many children, retirement – are associated with increased risks of low-income and relative poverty. Tax rates are low in comparison to neighbouring Austria, and low-income families benefit from a combination of additional tax relief plus social assistance. Family benefit has recently been extended and supplemented for single parents, and this reform was explicitly introduced to improve the situation of women.

Despite the buoyant economy, the female participation rate is low: just above 40% and almost half of employed women work part-time. Most of the recent increase in women’s participation is in the form of part-time work. Women only gained the vote in 1984, in 1996 a Gender Equality Office was created.

Page 78

Box 1.23. Lichtenstein – The buoyant economy means there is little pressure to stimulate debate for tax/benefit reform to ‘make work pay’, however employment rates for women are low (cont.)

• Social policies remain oriented largely towards maintaining traditional ‘male breadwinner’ gender arrangements. Measures are targeted at distributing resources to support women who have exited the labour market after childbirth, with an emphasis on supporting marriage.

• A major emphasis of the annual ‘Action Plan’ for gender equality is the implementation of gender equality in the area of social security. However, gender equality is interpreted mainly in terms of providing for (married) women who have left the labour market for domestic responsibilities rather than addressing women’s labour market exits.

A concrete example is pension reform. In 1996 the married couples’ pension was replaced with an individual pension, with individual pension insurance coverage extended to non-active women based on the introduction of the principle of ‘pension-sharing’. This principle has subsequently been extended to give divorced women access to their ex-husband’s company pension schemes by introducing pension splitting according to the number of years of marriage. While more women have gained individual entitlements to pensions through marriage, access to a pension through employment has become more onerous: employed women have to work more years than before because the retirement age has been equalised upwards to 64 years.

• In relation to supporting women’s integration into employment after childbirth unpaid parental leave provisions have been introduced recently, despite strong opposition from businesses which led to some modifications to limit provisions.

• Childcare services have been expanded and more flexible opening hours introduced since the mid-1990s, but full-time year-round care remains expensive.

A particular problem is the lack of implementation of the anti-discrimination legislation. No woman has filed a complaint of discrimination, so the mediation arrangements have not been put into use, although women’s organisations are aware that many cases of discrimination exist. Women are reluctant to come forward for fear of victimisation and stigmatisation in a small country. Anonymity is only possible if the complaint is brought under the umbrella of an organisation such as the Association for Working Men and Women, but this is only possible if there is more than one complainant from the same company.

Source: Papouschek (2004)

1.4. Other measures to encourage the unemployed to take part-time or low-waged jobs

We have already reviewed a number of recently reformed tax/benefit measures which contain some explicit financial mechanisms to encourage the unemployed to take part-time or low-waged jobs. These include the tax credit reforms in Belgium, France and the UK (Section 1.1), and provisions for the unemployed to retain an element of their benefit to supplement low earnings from either full- or part-time employment, for example in the cases of Germany, France, Ireland and Portugal discussed in the previous Section 1.3. Some of these supply-side reforms to unemployment benefit/assistance have been made in conjunction with policy provisions and budgets for demand-side job creation measures, notably the detailed job creation plans in Germany and France as well as the newly devolved municipality responsibilities in the Netherlands, which include budgets that can be used to subsidise labour costs or jobs (for details see the relevant national boxes in the above sections).

In a number of countries demand-side efforts through employer subsidies to encourage job creation for the unemployed and low-earners has become a key element of policy reform directed at ‘making work pay’. France and Belgium are two examples of where reductions in employers’ social security contributions have been made in a series of measures since the early 1990s to stimulate labour demand (Silvera, 2004; Meulders and O’Dorchai, 2004). The rebate for employers’ contributions in Belgium was raised again as of 1 January 2004, and it means that no contributions are paid for many of the lowest-paid employees. The Belgian report notes that the efficiency of such measures (reduced employers’ contributions) are often questioned because they do not appear to be effective in terms of job creation in relation to the cost of foregone social security revenue, but that in their favour the Belgian government considers that these measures support business competitiveness and promotePage 79 employment stability during economic downturns (Meulders and O’Dorchai, 2004).

In a similar vein, a demand-side policy emphasis on wage subsidies and the creation of part-time jobs are the focus of ‘making work pay’ policy reforms in Greece (Karamessini, 2004). Since 2000 five relevant measures have been introduced (Box 1.24). Three are directed at employees: full-time employees on minimum wage have been exempted from social security contributions, the long-term unemployed who take part-time work receive a transitional monthly allowance as an incentive, and a wage supplement has been introduced for ‘mini’ part-time jobs to make them more financially attractive. The other two initiatives are directed at employers: private sector employers receive wage subsidies for hiring unemployed persons in receipt of benefits to either full-time or part-time work, and in the public sector there is a job creation programme for fixed-term part-time jobs.

There has not been a gender assessment, or any other type of evaluation of this body of reforms in Greece (Karamessini, 2004). The wage subsidies for hiring the unemployed aims to make work pay for the low-paid by attempting to enhance work incentives without reducing benefit levels. This indirectly discriminates against women for two reasons. First, more unemployed men are in receipt of benefits than unemployed women. Secondly men also have longer benefit insurance entitlements due to their better paid and more stable jobs in their work histories, and this offers more incentive for employers to hire them. There is little variation between the minimum and maximum unemployment insurance benefit rates so in practice there is no gender difference in this element of the amount of subsidy.

Box 1.24. Greece – The focus of ‘making work pay’ policy reform is on wage subsidies and the creation of part-time jobs

The NAP/Employment 2003 explicitly stated that the tax/benefit system does not include work disincentives since benefits are low, as is the tax burden on the low-paid wage earners. High rates of social security contributions are identified as a potential disincentive. The policy emphasis is upon improving labour demand, active labour market schemes (training, counselling and subsidies for business start-ups), work-family reconciliation and raising social protection to the levels found in EU countries with more developed systems, rather than removing work disincentives. The social policy proposals of the new government are expected to be announced later this year.

Since 2000 five employment measures have been taken which relate to the ‘make work pay’ focus of debates:

• Tax break for low wage employees: Full-time employees on minimum wage have been exempted from social security contributions (2000).

• Private sector wage subsidies for hiring the unemployed: The unemployed on benefits who secure a job in the private sector can have their benefit used as a wage subsidy for the employer for either a full-time or part-time job if paid at the normal wage rate. The employer makes up the difference between the benefit and the wage. If the person is dismissed before the end of the entitlement period they continue to receive their unemployment benefit.

• Public sector creation of fixed-term part-time jobs for the unemployed/inactive: To fill vacancies in social services and related urgent recruitment needs. Public sector jobs have good working conditions and this helps to make part-time work more attractive to the unemployed and inactive, given the low preferences for this form of employment in Greece.

• Promotion of part-time work for the long-term unemployed: They receive a monthly allowance for one year if they take a part-time job of at least 4 hours/day (2000).

• Wages improved for those in low-paid ‘mini’ part-time jobs: Pay premia introduced in 2000 for the low-paid working less than 4 hours/day on minimum hourly wage rates.

Source: Karamessini (2004)

Page 80

Box 1.25. Germany – Expansion of ‘mini-jobs’ excluded from social protection coverage

Most of the increase in women’s employment in Germany in recent years has come from the expansion of private sector ‘mini-jobs’ (400 euro/month) introduced in 2003. The ‘mini’ or ‘marginal’ job definition is no longer connected to the previous 15 hour upper limit on working hours, so it is now possible to create jobs which involve longer hours at a lower hourly rate of pay which fall within the ‘mini-job’ definition. By June 2004 8 million had been created, an expansion of 1 million in 1 year. Two thirds of these employees are women.

The expansion of these jobs incurs some major disadvantages: employees do not have social protection coverage from this form of employment (in particular unemployment and pension insurance), the state forgoes the revenue from social security contributions, and the expansion of this form of employment undermines the maintenance and development of regular part-time jobs.

Source: Maier (2004)

Box 1.26. Finland – controversial proposals to subsidise low-wage jobs and expand the ‘working poor’ are deferred

The emphasis in the Finnish NAP/Inclusion is upon job creation measures for those with low skills and limited qualifications. To boost demand for low-wage jobs reductions in indirect labour costs as a subsidy for ‘low-productivity’ work were planned, along with an extension of training measures. This was based on a report by the Pellervo Economic Research Institute (Holm and Vihriälä, 2002) which proposed reduced payroll taxes for low-wage jobs, drew inspiration from the French reforms in this area and was debated widely in the 2003 election. However, after the new government costed the measure they concluded it would be too expensive and so have deferred implementation until 2006 at the earliest.

The measure was considered a controversial one because of the concept itself of ‘low-productivity’ which in the Finnish context is reminiscent of the US term the ‘working poor’, and measures to support the expansion of low-paid jobs runs counter to the Finnish solidarity and narrow income differentials in the welfare state regime.

Source: Lehto (2004)

Another important example to note is the promotion of ‘mini-jobs’ in Germany through such employment being exempt from social security contributions (Box 1.25). This development was given an additional stimulus in 2003 where the definition was reformed so that the previous 15 hour upper limit on working hours was abolished, leaving the definition of a ‘mini’ job resting on an earnings threshold. There has been a rapid expansion of ‘mini-jobs’ reaching a total of 8 million by June 2004. Two thirds of these employees are women, and the expansion of this form of employment accounts for much of the recent increase in women’s employment. The national report identifies some key problems with the expansion of this form of employment: such jobs are excluded from social protection coverage, the foregone revenue from social security contribution exemptions, and that the growth of ‘mini jobs’ undermines the development of regular, part-time jobs (Maier, 2004).

In contrast, proposals to introduce subsidies for low-wage jobs in Finland have been deferred (Box 1.26). This decision was made on the grounds of cost, but the proposal had been controversial because the idea of introducing measures to stimulate the creation of low-paid jobs runs counter to the principles of the Finnish welfare state regime and the broader ethos of Finnish society (Lehto, 2004).

The Finnish expert notes that the gender impact of the proposal was not assessed in the public debate, but that the reform would primarily impact women for they dominate in the low wage service sectors, as well as the small proportion of immigrant men and women in the population (Lehto, 2004). Critics of the proposal point to a range of issues which suggest that the proposal was poorly conceived: the evidence from employers’ surveys is that the demand for low-skilled labour will continue to fall while that for qualified labour will increase; the pool of low-educated people searching for work is small once students looking for vacation work are excluded from the count; and that many of the unemployed are highly skilled or have a disability and the employment integration of both groups is more easily achieved by different forms of intervention than by a general subsidy for the creation of low-wage jobs.

Page 81

1.5. ‘Making work pay’ in relation to pension reform and active ageing

The issue of pension reform in relation to work incentives and ‘making work pay’ was not included as an explicit element of the work programme, but a number of the national reports highlight reforms to pension systems which are largely designed to tighten eligibility conditions and to create financial incentives in favour of ‘active ageing’ and later retirement (France, Portugal, Austria, Greece, Italy, Norway, the Netherlands)7. Systematic gender impact assessments were not made for these reforms although there are likely to be important gender differentiated impacts of these reforms (Box 1.27). Firstly, this is because women typically have less continuous employment profiles and lower earnings. So, for example in Austria the new provisions for child-related breaks in the contribution record have been criticised as inadequate by the Federal Chamber of Labour. Secondly, a higher proportion of women’s employment is concentrated in the public sector so reforms which target either the private or public sectors are likely to have a gender impact, for example the recent reforms in France and Portugal8.

Box 1.27. Examples of pension reforms to tighten contributory requirements and deter early retirement

In Austria the pension system has been reformed to encourage people to prolong their working life (along with some reduction of non-wage labour costs for older workers): early retirement has been abolished, increased increments to pension for working beyond statutory retirement age have been introduced, and the contribution service requirement has been increased from the ‘best 15’ to 40 years. To reduce the negative impact of (mothers’) child raising breaks on the new 40 years contribution rule the contribution service requirement is reduced by 3 years per child, with enhanced contributory credits for a 2-year period of childcare per child. The additional measures for child-related breaks in the contribution record are insufficient to compensate women who take longer breaks or for periods of part-time employment, and will exacerbate the gender pensions benefit gap according to the Federal Chamber of Labour (Mairhuber, 2004).

In France the 2003 pension reform was passed despite major public opposition and mobilisation. This increases the contribution requirements and harmonises the public and private sectors. The number of contribution years is to be raised in the public sector (currently 37.5) to match that of the private sector (currently 40), and then increased for all to reach 42 years by 2020. Pension reductions are introduced for missing contribution years in the public sector, while existing reductions in private sector will be reduced to be harmonised at 5% for each missing year. The reform introduces the possibility to buy up to 3 years of contributions for missing years due to education. Some advantageous measures were introduced for part-timers, but short periods of part-time work are not counted. There is also some harmonisation of child-related contribution credits by tightening eligibility conditions for civil and public sector employees while preserving arrangements which exist in the private sector (Silvera, 2004).

In Portugal new regulations (January 2004) for civil servants’ pensions introduce tighter contribution requirements through extending the qualifying service period and the reference period for calculating the pension level, which penalises early retirement (Ferreira, 2004).

In Italy the pension system was modified (July 2004) to encourage older workers to postpone retirement age. From 2008 the minimum contributory history for early retirement (independent of age) is increased from 35 to 40 years, and the minimum age for early retirement is due to rise. The law provides for a different treatment for women, which means that if women take early retirement they will have a lower pension (Villa, 2004)

In Norway, in spring 2004 the Pension Commission presented a proposal to restructure the pension system in response to the challenges of an ageing population and trends towards earlier retirement. The main elements of the proposal are firstly to relate pensions more closely to overall lifetime income by counting all years instead of the best 20 years as occurs presently, and by linking the pension level more closely to the amount of contributions paid, so that later retirement is rewarded with a higher pension. Tax credits for periods of unpaid care work are to continue as before. Secondly, to replace the current early retirement schemes for some groups of employees with a flexible retirement age for all employees (between 62 and 67 years in contrast to the current retirement age of 67 years) (Ellingsæter, 2004).

Page 82

Box 1.27. Examples of pension reforms to tighten contributory requirements and deter early retirement (cont.)

In the Netherlands the proposition of the government to abolish the tax relief for early retirement was fiercely opposed by the trade unions. After a period of social unrest, the compromise stated that for persons not yet 55 at the first of January 2005, a new early retirement policy will be developed. In more general terms, it seems likely that the financial incentives for older workers to leave the labour market will change in order to increase the participation rate among the 55+ (Plantenga and Remery, 2004).

In Norway, the potential effect on gender equality was a central issue in the debate about the pension proposal due to women’s over-representation in low income groups. Some have criticised the removal of the ’20 best year’ rule, however, in the case of Norway there is evidence that this has in practice favoured men, due to the increasing and higher income of men over the lifecourse. Both the Gender Equality Ombud and the Ministry of Children and Family Affairs support its removal. The Ombud argues that pension credits for unpaid childcare should be extended from the current 6 years to 10 years, and backdated (they were only introduced in 1992). The lifetime income requirement for being eligible for flexible retirement is so high that women with an average wage will not reach it, and so they and other low income groups will not be able to retire early. Proposals for spouse sharing of pensions in divorce and survivor’s pensions have also been criticised for not being sufficiently worked out.

The extension of compensation for periods of non-employment due to childraising were also an issue in the recent pension reforms in Luxembourg and Lichtenstein, and in both cases the reforms have generally improved the protection offered to women. The 2002 reforms in Luxembourg increase the credit for periods of child-rearing from 3 years to up to 6 years per child, and the minimum reference wage for computing pension rights during the ‘baby-years’ (années-nourisson) to a level of 1.5 the social minimum income (SSM). Further, the reference wage is multiplied according to the number of children, so that a non-employed parent with two young children has a double credit (Plasman and Sissoko 2004). In Lichtenstein the individualisation of pension rights and the introduction of pension splitting has improved married women’s pension rights (see Box.1.23 above).

1.6. Work-family reconciliation as the main focus on ‘making work pay’

In Spain and Sweden the main policy reform which relates to the ‘making work pay’ debate in recent years is to improve work-family reconciliation. These countries are quite different on many economic, employment and social policy indicators, including that of work-family reconciliation. Thus in Spain, work-family measures are limited, although expanding and this condition – as well as others, not least high unemployment – contributes to the low female employment rate in Spain. In contrast, Sweden has the highest female employment rate among the EU-25, and the extensive work-family measures that exist in Sweden are well known.

In Spain women and young people are among the groups most at risk of social exclusion due to difficulties in accessing employment and changing family structures and relationships. Women with low education levels and those who are single parents are particularly vulnerable to poverty. However, the recent work-family reconciliation measures introduced in Spain (Box 1.28) have a limited impact on the employment integration of non-employed women (Moltó, 2004). The NAP/Social Inclusion does include some proposals to improve childcare services for lone parents, Romany mothers and female prisoners but only in general terms. The law on domestic violence is more specific, and here conciliation measures include leave of absence from work and geographical mobility due to physical or psychological violence along the same lines as parental leave, in conjunction with a new special employment and training programme to be introduced. Mothers of disabled children face particular reconciliation obstacles which are not addressed in current measures. Furthermore,Page 83 there is a growing population of immigrant women, mostly employed as domestic servants providing care-related services for other women. This group of women face precarious employment conditions and limited social protection with little access to the recent work-family conciliation measures introduced, yet their need is often more pressing because of their higher fertility rates.

Box 1.28. Spain – Work/family reconciliation measures are at the heart of the ‘making work pay’ debate

Work-family reconciliation is an objective of at least four different plans mentioned in the NAP/Employment and NAP/Social Inclusion, in order to tackle the problems of a shortage of public childcare, low levels of child-related payments, and to reconcile working hours and other features of employment with family life. These different plans are promoted by different ministries and at national, regional and local levels, which has had a positive effect on raising the level of debate, public awareness and stimulating initiatives. However, on a more negative note the developments have been uneven and the lack of specific budgets means that targets for childcare provision and other objectives are unlikely to be met. The most recent national initiatives include:

• Law 39/1999 on the conciliation of work and family life (leave provisions, working hours compatibility).

• Measure to pay employed mothers of children aged 3 years or under 100 euro/month. This is a measure of family promotion, not a measures against social exclusion per se since most non-employed are women not covered by this measure.

• Some recent expansion of childcare.

Source: Moltó (2004)

If the achievement of high employment rates for both sexes is one of the indicators of a successful ‘making work pay’ emphasis in social and economic policies then Sweden (Box 1.29) and Norway (see Box 1.17 above) have succeeded both in terms of job creation and maintaining work incentives in the context of a relatively generous social protection system for people who are unemployed or otherwise unable to work. The high female employment rate in Sweden is attributed to five factors which originated in the 1970s: individualisation of taxation, the expansion of education (including continuing adult education programmes), a narrowing of the gender wage gap due in particular to the solidaristic wage policy, the investment in parental leave, childcare and other social infrastructure, and the expansion of public sector employment opportunities.

Box 1.29. Sweden – High employment rates for both women and men suggest that social and economic policies have succeeded in ‘making work pay’

Sweden has the highest employment rates for both women and men (75.6% and 72.8% respectively in 2003) in the EU-25, although still lower than the peak achieved in 1990 prior to the economic recession. Hence, if an indicator of ‘making work pay’ is high employment rates for both sexes (and comparatively later retirement ages) then Swedish social and economic policies have succeeded both in terms of job creation and work incentives in the context of a relatively generous social protection system for people who are unemployed or otherwise unable to work.

Research has shown that the high female employment rate is attributed to five particular factors:

• The introduction of individual taxation at the beginning of the 1970s.

• The expansion of education in the 1970s. The EU lifelong learning target of 12.5% is already exceeded in Sweden, with women’s participation rates exceeding those of men.

• Rising relative wages for women, for which the solidaristic wage policy played an important role in narrowing the wage dispersion between higher and lower paid workers.

• The investment in social infrastructure, particularly childcare and parental leave.

• The expansion of public sector employment (50% of women and 20% of men are employed in the public sector).

Following the recession in the 1990s there has been some economic recovery but rates of unemployment and long-term sickness remain relatively high compared to earlier periods and both are a focus of economic policy. However, the problem seems to be harder to tackle now than in previous periods – this is partly due to the extent of the problem but it is also due to the extent of political disagreement about what should be done.

Page 84

Box 1.29. Sweden – High employment rates for both women and men suggest that social and economic policies have succeeded in ‘making work pay’ (cont.)

• In 2002 childcare costs were reduced for parents to enhance work incentives and labour supply

Day care provision is subsidised according to parents’ income and provision is comprehensive in terms of places. For example, in 2003 day care facilities catered for 45% of babies aged 1 year and 86% of children from 2-5 years old. In addition, every child is offered a free place in public pre-school from the Autumn term of the year in which the child becomes 4, equivalent to part-time hour provision (525 hours/year).

In 2002 additional steps were taken to reduce the marginal tax effects of childcare costs by the introduction of a maximum fee. This has contributed to the decline in the average marginal tax effect from an increase in labour supply over the period 1997-2003 in Sweden.

Source: Löfström (2004)

The public investment in subsidised childcare in Sweden was further extended in 2002 through the introduction of a maximum fee for parents, which has contributed to a reduction in the marginal tax effect on the labour supply of parents. The impact will be primarily on women’s labour supply, particularly those who are lower-paid, for the national report notes that Swedish studies have shown that childcare costs have a greater impact on women’s labour supply than on men’s, and that directly subsidised childcare has a more positive impact on increasing labour supply than does personal tax relief for childcare (Löfström, 2004).

Sweden, and the other Nordic countries demonstrate how long-term investment in social protection systems and social infrastructure can provide an inclusive policy model of ‘making work pay’. In this model comprehensive social protection provision is combined with an expectation that all men and women participate in the labour market, facilitated by social infrastructure which supports the family care responsibilities – children and elders – that most adults take on at different stages in their lives.

Work-family reconciliation measures have been expanded in most countries in recent years targeted primarily at promoting women’s employment integration and labour supply (OECD 2001). We now turn to discuss two measures which relate to young children – maternity/parental leave and childcare services.


[2] General tax reductions in Italy were approved at the end of 2004 and implemented in January 2005. This reform took place after the national reports were prepared. The reform was introduced to reduce the high tax burden, but in practice it only reduces tax payments for those with either extremely low or extremely high personal income, leaving the vast majority of employed persons almost unaffected.

[3] ‘Tax and benefit reform in the UK – making work pay’ was the subject of a peer review meeting in November 2000. The review focussed on the UK Government’s strategy to move people from welfare into work. For further details of the peer review programme remit and meetings see

[4] The UK NAP/Inclusion 2003-05 estimates that the introduction of the new tax credits will yield a transfer of resources from men to women of up to 2 billion pound (pp.35, para 24). There has also been a parallel reform of the child support system which benefits carers. If the parent with care responsibilities enters work they are allowed to retain 10 pound a week, which is seen as enhancing work incentives particularly for lone mothers. The focus of the scheme on shifting support from the state to fathers is not identified – the retention of the 10 pound a week is the first time the child support scheme is being used to support lone mothers and children (Rubery, 2003a).

[5] This emphasis on joint assessment focus has also been extended in the ‘New Deal for Partners of Benefit Claimants’ (introduced in 2001) which extends job-seeker requirements and employment services to many women previously excluded from the New Deal schemes but now included on the basis of the status of their partner (85% of participants are women). The rationale is aimed at reducing the number of workless households rather than expanding individuals’ opportunities. Joint claims for jobseeker’s allowance for childless couples have also been introduced, with work-focussed interviews which involve increased responsibilities for the actions of partners but no right of access to individual income, with income reduced pound for pound as partners’ earnings increase beyond a small earnings disregard. In most cases the nominated claimant for receipt of the benefit is the man, and research shows that many couples did not realise they had the option – reinforces the old male breadwinner model. Furthermore, as Rubery (2003b) notes, there is little attention paid to mobilising non-employed women into employment if they have an employed partner, even if he is in receipt of in-work tax credits.

[6] Even though they can claim tax credit to offset some childcare costs they can face a high marginal withdrawal rate (> 60%).

[7] In the UK a major new government report was published in Autumn 2004 which shows that current pension arrangements – the combination of basic state pensions plus occupational and personal private pensions – will deliver inadequate pensions for a large proportion of the population. Options for increasing contribution arrangements and other savings measures are being discussed, and most of these imply a longer working life and/or saving a high proportion of earnings than occurs currently.

[8] Equal treatment measures for access to pension schemes are proceeding alongside pension reforms but not necessarily integrated into the overhaul of pension systems. One example of a new equal treatment measure has occurred in Belgium where a new law (April 2003) has banned gender differentiated treatment (except in relation to the treatment of different life expectancies) in the taxation of supplementary pensions and other additional social security benefits, and established equal treatment in access to pension schemes and pro rata payments for full-time and part-time employees (Meulders and O’Dorchai, 2004).