This chapter sets out four reform scenario proposals for the solidarity benefit system in Portugal. Solidarity benefits, which form part of the broader system of all social benefits, provide support for low-income individuals and households typically with very low or no income. Existing solidarity benefits include minimum-income benefits, alongside a dozen other means-tested benefits targeted at specific needs or circumstances. All of the proposed reforms would simplify these existing solidarity benefits into a single simplified benefit. Two of the proposed reforms then extend this simplified benefit to i) increase support levels and ii) strengthen work incentives. This chapter describes these four reform scenarios in detail and sets out the anticipated impact of the proposals on household incomes, poverty, work incentives, and fiscal costs.
Towards a Unified Social Benefit in Portugal
Abstract
Proposed reforms to solidarity benefits, their impacts on employment outcomes and poverty, and their estimated fiscal cost
This chapter sets out four reform scenarios to improve the social benefit system in Portugal. These scenarios reform solidarity benefits and, in line with the project objectives, aim to enhance work incentives, increase the capacity for poverty reduction and improve the targeting of support while respecting fiscal constraints.
All the solidarity benefit reform scenarios simplify the existing benefit system, with each scenario providing a different additional focus
Scenario A, Simplified solidarity benefits: Reduces the number of benefits and harmonises the calculation of support.
Ten existing solidarity benefits are replaced by a single unified solidarity benefit with a single set of entitlement rules. The amount of benefit would depend on the household size – using the same rules as the existing minimum-income benefit – alongside entitlements to supplements for specific needs – such that overall support would be similar to other existing social benefits being replaced. A single means-test would reduce the level of support for households with other income sources or assets, while a benefit cap would limit the maximum amount of support.
Scenario B, Poverty reduction: Builds on Scenario A by increasing support levels.
A 10% increase would be applied to the basic support levels and the additional elements and would be proportional to recipients’ entitlement under Scenario A.
Scenario C, Improved work incentives: Builds on Scenario A reducing the earnings deduction rate.
This scenario would reduce the earnings deduction rate to 70 cents per euro, compared to the 80 cents deduction rate in Scenario A (and the current minimum-income benefit).
Scenario D, Extended harmonisation of solidarity benefits: Extends Scenario A to all solidarity benefits.
In addition to the ten benefits reformed under Scenario A, this scenario would also replace five more existing solidarity benefits for pensioners, the unemployed, and disabled persons.
Microsimulations based on EUROMOD suggest that the proposed reforms are likely to reduce poverty and improve work incentives, but further investment has the potential to achieve greater impacts
Coverage of benefits would improve most under Reform Scenarios B and C with higher benefit levels and improved work incentives. Improved support levels (Scenario B) and expansion of support to more low-earning households (Scenario C) would both independently increase coverage of solidarity benefits among low-income individuals from 12% to 13% by extending support to more low-earning households. Intended improvements in benefit take‑up would increase this further.
Simplification of benefits would improve support for households with additional needs. Under the proposed reforms households entitled to support for at-risk pregnancies, pregnancy termination, parental leave, adoption, hospital travel and new-born hospital care would be entitled to a single benefit which cumulates all additional need support rather than seeing their eligibility limited to one benefit irrespective of their needs.
The poorest individuals would gain most from the proposed reforms. Under Scenarios B, C, and D individuals in working-age households in the poorest income decile are most likely to benefit from the reforms. In total, average incomes in the poorest income decile are EUR 73 per year higher under Scenario B, EUR 33 higher under Scenario C, and EUR 21 higher under Scenario D. Pensioners would continue to be eligible for the pensioner minimum income benefit, but differences in entitlement rules would mean some individuals, fewer than 0.5% of pensioners, would be worse‑off as a result of the closing the Social Pension.
Work incentives would be more consistent and smoother under all reform options, and further improved under Scenario C. Under all scenario’s, the new single deduction rate, of 80 cents for every euro of earnings, would ensure benefit claimants always have an incentive to work more hours. This would replace fixed earnings threshold rules for many existing solidarity benefits which can discourage full-time work. Scenario C would further improve work incentives by reducing the earnings deduction rate to 70 cents per euro earned and lowering participation tax rates towards the EU-average.
Increasing benefit levels by 10% would reduce deep poverty by 0.1 percentage points, while reducing the earnings deduction rate by 10 percentage points would reduce deep poverty by 0.06 percentage points. Scenario B, increasing benefit levels by 10% would reduce the at-risk-of-poverty rate by 0.10 percentage points – around 10 000 persons (defined as person with income below the 40%-of-median-income threshold). Scenario C would only affect those in work (with low earnings) and by reducing the earnings deduction rate by 10 percentage points would reduce the poverty rate by 0.06 percentage points – around 6 000 persons. Further reductions in poverty would occur under if the reforms resulted an increase in benefit take‑up.
Improvements in targeting would be marginal, with greater equalisation of support between working-age and pensioner households required for more substantial improvements. Under the reforms proposed, the share of solidarity spending going towards the poorest households would increase by between 0 and 3 percentage points across the four reform scenarios (from a current 48% share).
Simplification alone is anticipated to reduce fiscal costs. Additional investment, at relatively minor levels, to increase support levels (Scenario B), or improve work incentives (Scenario C) comes with a minor fiscal cost. Scenario A is estimated to reduce spending by EUR 6 million per year, mostly from reforming the Social Pension. Scenario B would increase spending by EUR 42 million per year, as a result of increased benefit levels. Scenario C would increase spending by EUR 15 million per year, due to increased benefit support for in-work solidarity households. Scenario D would reduce benefit spending by EUR 18 million per year, mostly because of how the smoother earnings deduction rate reduces support for existing recipients of the Social Unemployment Benefit who are living in working households.
This section outlines four reform proposals for the solidarity benefit system in Portugal. The first proposal (Scenario A), discussed in Section 2.1.1, focuses on simplifying most of the existing working-age solidarity benefits by combining support into a single benefit. The second proposal (Scenario B), discussed in Section 2.1.2, builds on the first reform by increasing benefit levels for solidarity benefits. The third proposal (Scenario C), discussed in Section 2.1.3, also builds on the first reform but looks at improving work incentives. The fourth proposal (Scenario D), discussed in Section 2.1.4, considers a more comprehensive reform of all solidarity benefits, combining all working-age and pensioner support into a single solidarity benefit.
2.1.1. Scenario A: Simplified solidarity support
Scenario A, simplification of solidarity support, combines ten of the existing working-age solidarity benefits, set out in Figure 2.1, into a single solidarity benefit (Prestação Social Única, PSU, in Portuguese). The existing benefits that would be combined into the single solidarity benefit are:
Minimum Income Benefit (Rendimento social de inserção, RSI),
Social Parental Benefit (Subsídio social parental),
Social Adoption Benefit (Subsídio social por adoção),
Widow’s Social Pension – for working-age individuals (Pensão de viuvez),
Orphan’s Social Pension (Pensão de orfandade),
Social Pregnancy Termination Benefit (Subsídio social por interrupção da gravidez),
Social Specific Risks Benefit (Subsídio social por riscos específicos),
Social Pregnancy Risk Benefit (Subsídio social por risco clínico durante a gravidez),
Social Hospital Travel Benefit (Subsídio social por necessidade de deslocação a unidade hospitalar), and
Social Newborn Hospital Care Benefit (Subsídio social específico por internamento hospitalar do recém-nascido).
Existing solidarity benefits that would be replaced by a working-age solidarity benefit under Reform Scenario A
Source: DGSS and OECD.
Separate to these working-age solidarity benefits the Social Pension (and by implication the Widow’s Social Pension for pensioners) should also be discontinued.1 Recent increases in the levels of the minimum income benefit for pensioners (CSI) have made the Social Pension benefit effectively redundant, as almost all recipients entitled to the Social Pension are entitled to more support under CSI, regardless of whether they receive the Social Pension or not. Exceptionally, some pensioners with significant assets but very low income may lose out because they are entitled to the Social Pension but not the minimum income benefit for pensioners. Overall, discontinuing the Social Pension will have almost no impact on pensioner households (discussed more in Section 2.2.2).2
The Social Unemployment Benefit (Subsídio social de desemprego), the Carergiving Benefit (Subsídio de apoio ao cuidador informal principal) and Pensioner Social Assistance Benefit (Complemento solidário para idosos, CSI) would not be affected by this reform and would continue to operate under their current rules.3
The level of entitlement under the simplified working-age benefit would be calculated at the household level and would depend upon the number of household members, their circumstances, and their income. Calculation of benefit entitlements would proceed in four steps: i) defining the benefit unit household, ii) calculation of the basic level or support, iii) calculation of the value of any additional supplements for specific circumstances, iv) deduction of earnings and other income from the amount of support, and v) application of a benefit cap to limit the total level of support (Figure 2.2).
Proposed simplified solidarity benefit calculation steps
Source: OECD.
Entitlements for many recipients would remain unchanged from the existing system – especially workless households claiming only minimum income benefit or single adult households claiming one of the other benefits. This is because the equivalence scales and the means-testing approach to household income for the proposed new benefit would be the same as the existing minimum income benefit (RSI). In cases where entitlements do differ for workless households, this is to the advantage of recipient households currently entitled to multiple solidarity benefits. Under the current system additional support from additional need benefits (such as Pregnancy Risk Benefit) can be offset by lower entitlement to RSI for households with multiple adults and/or children. The proposed new system would keep the higher amount of RSI support for these larger households in addition to higher support for extra needs through a supplement. Section 2.2.2 discusses this effect in more detail.
The harmonisation of the income and asset means-test to align with the existing RSI means-test theoretically results in higher and lower entitlements for existing benefit recipients in working households. In practice, however, few working-age households are directly affected under Scenario A according to EUROMOD modelling. This can be explained by few recipient households with earnings in the relatively narrow entitlement band or with assets that are affected by the changing asset entitlement rules. The numbers of winners and losers from each reform are examined in detail in Section 2.2.2.
Step 1: Household benefit unit definition
The new single solidarity benefit would be calculated at the whole household level (up to and including relatives of the second degree). This definition, in line with the existing household definition for RSI, includes all adults and children living at the same address who are sharing common meals. Notably, this “whole household” approach for the simplified solidarity benefit would include any pensioners living in the household. This means that CSI income is included in the income deduction as described below.
Annex 2.A assesses different possible household benefit unit definitions in more detail, including an alternative household benefit unit definition for the simplified single solidarity benefit. This alternative household benefit unit would exclude pensioners and their income from the entitlement rules, leading to a more generous entitlement for households containing both working-age adults and pensioners.
Step 2: Calculation of basic support levels
The basic level of single solidarity support would be calculated in the same way that RSI is calculated in the current system and would use the same levels. As set out in Table 2.1, the household would be entitled to EUR 237.25 per month for the first adult, EUR 166.08 for each subsequent adult (70% of the first adult level), and EUR 118.63 for each child aged 18 or under (50% of the first adult level).
Basic monthly levels of single solidarity benefit per household member
Source: DGSS and OECD.
The basic level of support available under the new solidarity benefit would, in effect, be the minimum income level any given household would have in Portugal (assuming the household claims the benefit). The total amount of basic entitlement for a household is the sum of the entitlement levels for each member. For instance, a couple with two children is entitled to EUR 640.58 (237.25 + 166.08 + 2 x 118.63).
Step 3: Calculation of supplements
Under certain circumstances households would be entitled to short-term supplementary amounts in addition to the basic support levels – these supplements would reflect, and replace, many of the current situational benefit streams that currently operate alongside the existing minimum income benefit in the solidarity system. They would be intended for temporary additional needs that would last six months at most.
Short-term supplements
Households would be able to receive multiple short-term family-related supplements with no limit. These short-term supplements would be cumulative and paid in full, on an equivalent daily basis, for every person that fulfils the criteria, for as long as conditions were met. The conditions would remain the same as their equivalent existing benefits and would be as follows:
Pregnancy Risk Supplement: available when a pregnant mother is unable to work their normal job because doing so would create a health risk for the mother or the unborn child.
Specific Risk Supplement: available when a mother of a new-born child is unable to work their normal job because doing so would create a health risk for the mother or the newborn child.
Hospital Travel Supplement: available when a pregnant mother is advised to travel to a hospital to give birth.
Newborn Hospital Supplement: available when a newborn baby is required to stay in a hospital due to medical reasons.
Pregnancy Termination Supplement: available when someone is unable to work due to the recent termination of a pregnancy.
Parental and Adoption Supplement: The household is entitled to this supplement if an adult has recently become a parent or has recently adopted a child. The supplement is valid for up to 150 days across both parents of any newborn child or newly adopted child. In the case of a newborn child, the mother is entitled to an additional 72 days, and the father an additional 35 days.
Table 2.2 lists all the supplements that would be available under the working-age single solidarity benefit, and their monthly values. The combined amount of support of basic support and supplements for a single adult under the reformed system has been calibrated to be equivalent to the support available under the current system. For instance, an expectant mother entitled to the Pregnancy Risk supplement would be entitled to a total single solidarity benefit of EUR 407.41 (EUR 237.25 basic and EUR 170.16 supplement). That is the same as the monthly value of the existing Pregnancy Risk solidarity benefit. This is true for all the supplements in the case of a single adult. Larger households with multiple adults or with children would receive more solidarity support overall under the reform than compared to the current system. Section 2.2.2 discusses this is more detail.
Single solidarity benefit supplements and monthly values, euros
Note: Short-term supplements would be paid at the full rate for multiple supplements in a household.
Source: OECD.
Step 4: Income deduction
The combined total of the minimum income support and the supplements would give an overall total maximum for the single solidarity benefit. The proportion of this maximum to which the household would be eligible would then be calculated on the basis of a means-test, on the income and assets of the household. The means test would reduce the level of entitlement for those households with higher incomes and more assets. For households with sufficient income and assets, entitlement to solidarity support would be reduced to nothing.
The deduction rate would be calculated as follows: for every euro of employment income (as an employee or self-employed) the single solidarity benefit entitlement would reduce by EUR 0.80. In line with the current minimum income benefit, there would continue to be a temporary reduced deduction rate of EUR 0.50 for new employees who were returning to work after 12 months of unemployment for the first six months of the new employment. Earnings deduction rates below EUR 1.00, enable workers to keep some of their benefits alongside their earned income and, as such, incentivise benefit recipients to work.
Non-employment income would also be deducted from the benefit entitlement, but at a rate of EUR 1.00 for every euro of income. This would include: income from other benefits (but not income from child benefits, the individual disability benefit or other family protection benefits); income from rental properties; savings income, and private pension income. As pensioners are included in the benefit household, any Pensioner Social Assistance Benefit they are receiving would be included as deductible income. Annex A considers this interaction in more detail in terms of how it affects pensioners living in different households.
The single solidarity benefit would also be means-tested against assets by calculating an imputed income based on the value of the assets. The approach would be similar to the current system but would remove some disincentives to save that are present in the current system.4 The primary residence of the household and the value of any car would not be included in the valuation. All other assets over the threshold of EUR 16 000 would be converted into an imputed monthly income by multiplying by a fixed flat rate of 5%. This amount would also be deducted from the single solidarity benefit entitlement. The expectation is that the household support themselves using their savings until their savings fall below the threshold.5
Table 2.3 illustrates an example income deduction calculation for a household that would be entitled to the new single solidarity benefit. A household with a single pensioner, three other adults, and two children would be entitled to EUR 162.06 per month after their earnings of EUR 200, savings of EUR 18 000, and pension benefit of EUR 550.67 are taken into account.
Example income deduction calculation for the proposed new working-age solidarity benefit for an example household with a single pensioner, three working-age adults, and two children.
Note: There is a savings allowance of EUR 16 000.
Source: OECD secretariat calculations.
Step 5: Benefit cap
A benefit cap – as specified according to the request of the Portuguese administration – has also been applied to the simulated scenarios. Thus, once the amount of support net of income deductions has been calculated, the resulting level of support would be subject to this benefit cap. The application of the benefit cap would be the final calculation step to determine entitlement to the proposed new working-age single solidarity benefit.
The benefit cap would be set equal to the monthly equivalent value of the minimum wage, EUR 820, multiplied by the number of adults in the household. For example, a single‑adult household would have a benefit cap of EUR 820 per month, a two‑adult household of EUR 1 640 and so forth. If the households combined income, including earnings and solidarity benefit income but excluding Child Benefit and other family protection benefit income, is above this cap, then their entitlement to the proposed solidarity benefit would be reduced until the income level reaches the cap. For example, a household with one adult and five children would be entitled to EUR 830 if they had no other income. This household would have their entitlement to a single solidarity benefit reduced to EUR 820.
Such a benefit cap is unlikely to affect many households, and risks complexifying the benefit calculation with little practical effect. In practice the cap will only apply to single parent households with four or more children, or couple households with more than eight children. Moreover, if the household is in receipt of any of the supplements listed above, the benefit cap would not apply. Administrative data on minimum income benefit (RSI) shows that there were fewer than 1 000 households that were large enough for the benefit cap to apply in 2023.6 Applying only to larger households, such a benefit cap will disproportionately increase poverty among children, who are over-represented in such households. Nevertheless, the number of affected households is sufficiently small that this will not come across in the simulations and is unlikely to impact on child poverty levels.
Beyond the simplified solidarity benefit reform, two additional reform options are outlined below which would provide additional support to low-income households and improve work incentives. The first of these is to increase basic levels of benefit support by 10% – the poverty reduction scenario (Section 2.1.2). The second is to reduce the employment income deduction rate by 10 percentage points from 80% to 70% – the work incentives scenario (Section 2.1.3). Both of these reform scenarios are considered in addition to the simplified solidarity benefit reform described above.
2.1.2. Scenario B: Poverty reduction – Increasing single solidarity benefit levels
With the goal of strengthening the poverty reduction capacity of the benefit, this alternative reform scenario simulates an increase in benefit levels in addition to the proposed simplified solidarity support proposed in Scenario A. This scenario increases the basic single solidarity benefit level by 10% and would affect all households receiving the benefit, providing more support to both workless and in-work households (with very low earnings). All households currently entitled to support would see the same cash gain relative to their household size. A two‑adult, two‑child household would receive a larger cash increase than a single adult.
Figure 2.3 illustrates the impact of a 10% increase to basic solidarity benefit levels for an example household with a single parent and two children. As Panel A shows, for a parent who is working the equivalent of 0 to 27 hours per week and earning at the minimum wage, the increase in solidarity benefit is passed on in full and is worth EUR 47 per month. With earnings of between 28 and 30 hours at the minimum wage, the household will be newly entitled to some support (worth up to EUR 47 per month). Panel B shows the effect of the reform on total income. If the household is not working the higher level of solidarity support will increase their overall income by 7%, with smaller proportional increases for households with higher earnings. The overall income effect when the household is not working is lower than 10% because Child Benefit support remains unchanged. This, in effect, dilutes the overall income effect for households with children. In addition, higher solidarity benefit income reduces entitlement to Child Benefit between certain earnings levels (between 15 and 21 hours at the minimum wage in Panel B), which partly offsets the income gain.
Single solidarity benefit and total income under proposed reforms with current benefit levels and a 10% increase, for a household with a single parent and two children by the number of hours worked at the minimum wage, 2025
Note: Total household income includes single solidarity benefit, net earnings after social contributions and income tax, Child Benefit and Child Guarantee. Children aged over 3.
Source: OECD calculations.
Increasing the basic level of solidarity benefit support would provide higher incomes to all low-income households in receipt of the benefit, thereby reducing levels of poverty. Importantly, the increase in support reflects the size of the household, so larger households and households with children will see a proportionally similar reduction in poverty as smaller, single‑adult households.
2.1.3. Scenario C: Work incentives – Reducing the earnings deduction rate
This third reform scenario is focussed on enhancing work incentives; strengthening the financial return to work, to encourage benefit claimants to move to self-sufficiency. The proposed changes are also in addition to the Scenario A proposals. This scenario would reduce the main earnings deduction rate from 80% to 70% and would allow single solidarity recipients to retain a larger share of their benefit income when entering work, thereby experiencing a more noticeable increase their earnings. Because the increased work incentive results from higher benefit support for households with some earnings the financial gain is greater for higher-income solidarity benefit recipient households.
Figure 2.4 illustrates how such a 10 percentage point reduction in the earnings deduction rate would affect benefit entitlement (Panel A) and household income (Panel B) for an example single parent household with two children. There would be no change in income for a household with no earnings. However, as the household increases their earnings up to the equivalent of working 27 hours per week at the minimum wage, their benefit support and income would be EUR 50 higher per month under the lower deduction rate.
The maximum percentage increase in income for this example household is a 6% increase, although this would be higher for larger households with higher earnings levels. This is because these households would see entitlement to the single solidarity benefit, extend further up the earnings distribution. As with Scenario B, higher solidarity benefit reduces the level of Child Benefit at some points in the earnings distribution (between 16 and 22 hours at the minimum wage as shown in Figure 2.4 Panel B).
Single solidarity benefit and total income under proposed reforms with current 80% earnings deduction rate and lower 70% rate, for a household with a single parent and two children by the number of hours worked at the minimum wage, 2025
Note: Total household income includes single solidarity benefit, net earnings after social contributions and income tax, Child Benefit and Child Guarantee. Children aged over 3.
Source: OECD calculations.
2.1.4. Scenario D: Extended harmonisation of solidarity benefits
This fourth, and final, reform scenario expands upon the first reform option. Instead of simplifying ten solidarity benefits, this reform proposal consolidates all 15 of the current solidarity benefits into one single solidarity benefit, including pensioner support. While this complete single solidarity benefit would operate under most of the same calculation rules as set out above, there would be some additional supplements to reflect the other benefits that are replaced, and it would have a different household definition.
Figure 2.5, below, lists all 15 existing solidarity benefits that would be replaced by a simplified solidarity benefit under a more extensive reform. In addition to the benefits reformed under the first simplification reform scenario (Section 2.1.1), the following four remaining solidarity benefits would also be included in reform Scenario D:
Social Unemployment Benefit (Subsídio social de desemprego),
Disability Benefit household element (Prestação Social para a Inclusão o Complemento),
Pensioner Social Assistance benefit (Complemento solidário para idosos), and
Caregiving benefit (Subsídio de apoio ao cuidador informal principal).
Existing solidarity benefits that would be replaced by a complete single solidarity benefit
The calculation rules for the broader single solidarity benefit would be the same as set out above in Section 2.1.1 with two exceptions. First, the household benefit unit definition would distinguish between all working-age adults and pensioner couples into separate benefit households. Dependent children would be included in the benefit household of their parent.7 A pensioner or a pensioner couple would therefore be entitled to the single solidarity benefit on the basis of only their circumstances and income. All working-age adults would be entitled to support on the basis of the circumstances and income of all working-age adults in the household. This change to the household benefit unit definition would ensure pensioners receive the same level of support, while also making solidarity benefits fairer for mixed-aged households (under the current system mixed-aged households have pensioner support deducted from working-age support leaving the household worse‑off than if they were living separately).8
Second, the new single solidarity benefit would have long-term supplements in addition to the short-term supplements as part of Scenario A. These long-term supplements would provide support for long-term additional needs and would reflect the higher amount of support provided by the current system for pensioners, disabled persons, social unemployment benefit recipients, and carer benefit recipients. These would be subject to a reduced rate where the household qualifies for multiple supplements. The household would receive the full value of the first of these supplements, 70% of the value of the second supplement, and 30% of the value of the third or subsequent supplement.9 The conditions for these supplements would be as follows:
Unemployment Supplement: Entitlement depends on whether adults in the household meet the conditions for social unemployment support. Specifically, the adult has met the contribution requirements of having worked 180 days in the last year,10 or having just exhausted insurance‑based Unemployment Benefit.
Care Supplement: The household is entitled to this supplement for each adult that is performing unpaid care duties for another disabled adult who is receiving Assistance Benefit. The adult receiving the care can be in the same household or a different household.
Pensioner supplement: This supplement is payable if any adults in the benefit household are over the pension age of 66 years and 4 months.
Disability Supplement: The household is entitled to this supplement for each adult in the household that meets the disability threshold of 60% as currently set out under the Disability Benefit.
Table 2.4 shows the values of all the supplements under the proposed extended solidarity benefit, including those set out in Section 2.1.1 above. Households would only be entitled to one long-term solidarity benefit per adult in the household. For instance, a pensioner would only be eligible for the pensioner supplement. The reduced values for long-term supplements would depend on total number of long-term supplements the household is receiving. Because of the higher value of the pensioner supplement, it would always be considered the first supplement in the benefit household (and second when two pensioners).
Single solidarity benefit supplements and monthly values, euros
Note: Short-term supplements would always be paid at the full rate for multiple supplements in a household. Reduced supplement values for long-term supplements depend on the total number of supplements received in the household. The first supplement is paid at the full rate, the second at the second rate, and other supplements at the subsequent rate where relevant. Pensioner supplements are considered first.
Source: OECD.
These additional supplements would mean that most households currently receiving existing benefits would receive a similar level of support under this extended single solidarity benefit. For instance, single and couple pensioner households would receive the same level of support under the new benefit as they currently receive from CSI. Larger households, that receive multiple solidarity benefits under the current system, may receive more or less support depending on their exact circumstance and other income in the household. Typically, these more complicated households with no other income will receive more support under the reform scenario, while those with more income may receive less support. The empirical analysis of Section 2.2 outlines the impact of these changes – both fiscal and distributional.
The purpose of the reforms set out in this chapter is to strengthen the current solidarity benefit system through simplification and changes to the benefit levels and entitlement rules. These improvements are designed to increase take‑up, harmonise work-incentives, reduce poverty, and improve the targeting of support. Beyond this immediate goal, the simplification of the benefit system will facilitate a separate reform project in Portugal which is modernising the digital infrastructure that manages and processes benefit payments.11
This section examines the anticipated impacts of the solidarity benefit reforms presented in Section 2.1:
Scenario A: Simplified solidarity support – working-age single solidarity benefit reform.
Scenario B: Poverty reduction – working-age single solidarity benefit reform with a 10% increase in benefit levels.
Scenario C: Work incentives – working-age single solidarity benefit reform with a 10 percentage point reduction in the earnings deduction rate.
Scenario D: Extended harmonisation of solidarity benefits – complete single solidarity benefit reform (including pensioner benefits).
Table 2.5 summarises the immediate impacts of each reform against key measures including household incomes, poverty, work incentives, and fiscal costs. The potential indirect impact of a simplified benefit architecture on poverty and fiscal costs that may occur through higher take‑up of solidarity benefits is also considered.
Impacts of solidarity benefit reforms scenarios on selected household incomes, at-risk-of-poverty rates, work incentives, and fiscal spending in 2024.
Note: Analysis excludes behavioural impacts, except where higher take‑up of support is specified.
Source: OECD analysis using Euromod.
Looking further ahead the simplification of solidarity benefits is likely to have three additional advantages.
It creates a framework from which to harmonise other parts of the benefits system (such as child and housing benefits) to further improve work incentives and remove inconsistencies in support.
It allows for the extension of employment support work search conditionality to more individuals.
A simplified solidarity benefit could be expanded quickly and easily to provide more in-work support to respond to rising inequalities, and immediate additional income support during future economic crises.
The remainder of this section discusses these, and other impacts, in more detail. The impact analysis builds on microsimulation modelling, using European Union’s microsimulation model, Euromod, as well as data from the EU income and living conditions survey from 2022 and the OECD Tax-Benefit Model. Results are calibrated to administrative benefit data from 2023 to provide more accurate results that reflect the partial take‑up of existing benefits. For ease of understanding and comparison, the analysis considers impacts as if the reforms were fully enacted in 2024. In practice, the impacts will vary in future years depending on the specifics of the implementation of the reform, as well as on macro trends – both short- and long-term. Short-term factors, such as labour market tightness, will impact on household incomes in turn affecting the number of households eligible for solidarity benefits. Longer term trends such as technological progress and demographic changes will also impact upon benefit eligibility and spending. Box 2.1 describes the methodological approach of the impact analysis in this section in more detail. Annex 2.B provides more information about the Euromod microsimulation modelling, including important limitations of the analysis.
The simulations of the proposed reforms to the social benefit system that form the basis of this analysis will shed light on the expected differences between the four reform options outlined in Section 2.1 Performance will be assessed both from the perspective of the individual claimant – who would be eligible for how much, and for how long? And from the perspective of the country as a whole – the social and fiscal impact of the reforms. To inform these ex-ante simulations this note draws on the following technical work.
Impact of reforms on key benefit parameters in an international comparison
The OECD Tax-Benefit Model (TaxBEN) compiles the complex benefit policy architectures of countries across the OECD into an internationally comparable format. The model can be used to provide insights into the performance of the benefit system along a number of tangents such as income adequacy, net replacement rates and work incentives. The model draws upon national benefit rules provided by relevant countries’ authorities, as well as their interaction with other relevant elements of the tax and benefit system including social contributions and taxation. Because these rules vary across jobseeker profiles and household characteristics, international comparisons modelled with TaxBEN are based upon vignettes; that is, they are based upon a number of illustrative cases of specific household compositions and employment histories.
As such, TaxBEN gives a picture of what an individual in these hypothetical circumstances should expect to receive under each reform scenario (OECD, 2022[1]). However, TaxBEN is not based upon actual data regarding the circumstances of the Portuguese population, and thus, it is unable to shed light on the likely economy-wide impact of the reforms – be that in terms of the fiscal or the distributional impact. To complement the TaxBEN analysis therefore, microsimulations informed by Portuguese data delve more deeply into questions of the likely impact of reforms on the Portuguese population.
Anticipating the distributional and fiscal impact of reforms
Alongside the cross-country comparative policy vignettes, microsimulation methods apply the proposed reform parameters to the Portuguese data. The microsimulations that inform this report employ the European Union’s Euromod model, utilising the Portuguese component of the European Union’s Statistics on Income and Living Conditions (EU-SILC).
Euromod is essentially a calculator of tax liabilities and benefit entitlements for the EU Member States. It combines detailed legislative information on tax and benefit rules with a representative sample of the population to calculate benefit entitlements and income tax payments and social security contributions for the households that appear in the data. The simulated components include most direct taxes (especially income taxes on all sources of income including tax credits, payroll taxes and social insurance contributions) and benefits (e.g. welfare benefits, social assistance and some transfers based on previous contributions, e.g. unemployment benefits). Indirect taxes and taxes on corporate profits are not included in the model, likewise in-kind benefits.
To run the simulations, Euromod creates a synthetic dataset representative of the Portuguese population in 2024. The synthetic dataset is largely built from information contained in EU-SILC 2022, which refers to 2021 income. A number of external sources, mostly administrative, are also used to adjust and extrapolate some parameters to the 2024 context. Information provided by MTSSS plays a key factor to improve (or impute) parameters needed to simulate eligibility benefits, as well as their level. Some examples of the use of external sources to improve EU-SILC information are:
Administrative data on benefit spending on social benefits is used to calibrate the model (through benefit take‑up) so that total spending on social benefits matches that provided by the Portuguese Government.
Macroeconomic parameters used to extrapolate 2022 figures to 2024.
The rules of different reform scenarios, as well as the rules reflecting the current situation (baseline) are implemented as independent “models” and run over the synthetic population to produce the results presented here. This approach guarantees comparability between the baseline and different reform scenarios.
2.2.1. Coverage of solidarity benefits
The proposed reforms in this chapter would improve benefit coverage through two main channels: closing support gaps for the working poor and encouraging higher take‑up of benefits. Benefit coverage is defined as the number of households receiving given benefit support. For income‑replacement and minimum income benefits targeted at low-income households – such as solidarity benefits in Portugal – this is typically expressed as a percentage of the low-income population. Existing benefit coverage levels in Portugal are low. As shown in Chapter 1, just 34% of single workless adults receive social benefits, and just 3% of low-paid adults living alone receive any social benefits, among the lowest rates across comparator OECD countries.
Proposed reforms could close support gaps for the working poor
Poverty is not just an issue confined to workless households. Across the EU, 8% of employed persons are in poverty, in Portugal this figure stands as high as 10%.12 Social benefit systems have an important role to play in reducing in-work poverty by providing a top-up to households with very-low earnings, while of course continuing to incentivise those households to increase their earnings. In-work poverty is especially relevant for single parent households who have reduced capacity to work full-time and have a higher support ratio of children to adults than other households.
Reform Scenario C, improving work incentives, not only improves the financial incentives for solidarity benefit recipients to work more hours, but it also provides more support to low-earning households. Figure 2.6 shows that reducing the earnings deduction rate of solidarity benefits extends support further up the income distribution while also providing more support to low-earning households. Under Reform Scenario C, a single parent with two children will be entitled to some solidarity support when working 30 hours per week at the minimum wage compared to 21 hours under the current system (and compared to 26 hours under Reform Scenario A). Under a reform Scenario C, the same single parent working 27 hours per week at the minimum wage would be newly entitled to solidarity benefit income of EUR 58 per month, a 7% increase in income compared to the existing system. In total Scenario C would increase solidarity benefit coverage of very-low-income individuals below pension age by 1 percentage points (from 12% of the poorest 10% of individuals to 13%).13
Reform Scenario B, improving poverty reduction, also improves coverage of solidarity benefits among individuals below pension-age, also by 1 percentage point. This is because increasing the level of benefit support mechanically increases the maximum earnings threshold for solidarity benefits. For a single parent with two children, entitlement to solidarity benefits would extend up to 29 hours at the minimum wage.
Solidarity benefit income (Panel A) and total household income (Panel B) for a single parent, two‑child household under existing system and proposed Reform Scenario C, 2025
Note: Income includes solidarity benefit, child benefit and earnings net of social contributions and income tax.
Source: OECD calculations.
Benefit simplification could encourage higher coverage through higher take‑up
The proposed reforms may increase coverage, beyond those mechanisms outlined above, if they are successful in improving take‑up of solidarity benefits. Estimates currently suggest a take‑up rate for solidarity benefits of 71% in Portugal. In effect, 29% are not claiming support to which they are entitled. While take‑up of solidarity benefits in Portugal is low compared to some other benefits (such as child benefits), this is typical for minimum-income benefits across the EU (Eurofound, 2015[2]). Take‑up of minimum income benefits can be low for several reasons. These include: i) a lack of information, individuals may not be aware that they are entitled to support; ii) stigmatisation, individuals may avoid claiming benefits they are entitled to because they do not want the negative association of claiming low-income benefits; iii) willingness to work, individuals may not be willing to engage with the work search conditions required to access benefits; and iv) financial need, some individuals have other resources available to them in the short-term and don’t need or want to claim.
Simplifying support can increase take‑up of benefits by making it easier to understand the benefit rules, and easier to claim the benefit. By streamlining entitlement conditions, and simplifying application procedures, the reform would reduce the informational barriers that undermine applications. Under a single solidarity benefit there would be one set of entitlement rules based on a single definition of a household for the purposes of calculating the level of benefit support as well as the relevant income and assets which should be taking account in the means-testing of the support. With fewer social benefits, less knowledge is required to claim a single working-age solidarity benefit compared to deciding which one of 12 different solidarity benefits to claim. Moreover, if additional supplements are automatically paid under the new solidarity benefit, then take‑up of benefit entitlement would further improve as it would be impossible to miss out on this additional support. For instance, an existing minimum income benefit (RSI) recipient may not be aware of their entitlement to social pregnancy benefit. Under a new single working-age benefit, the additional pregnancy supplement could be paid automatically if the relevant government systems are connected.
At the same time harmonising support by combining benefits may also increase take‑up by limiting the stigma associated with claiming minimum income support. This is because applications for minimum income support would become inseparable from applications from what are currently separate social benefits. Scenario D, which combines all solidarity benefits under the proposed reform, goes furthest in this regard.
2.2.2. Adequacy and household incomes
Alongside entitlement rules a key feature of income‑replacement benefits is the amount of support they provide, and the effect this support has on household income. More generous benefits will have a greater impact on poverty reduction, albeit with trade‑offs for work incentives and fiscal costs. The exact effect of the reforms on incomes will vary according to the reform and the circumstances of each household, some would be better-off and others worse‑off.
Reform options could increase support levels for working-age households
Reform Scenario B, improving poverty reduction, would directly increase the level of solidarity benefits for all working-age households eligible for support. By themselves, other scenarios would not uniformly increase support levels (Scenario C, work incentives, only increases the level of support for poor working households as described in Section 2.2.1).
The 10% increase in base support (described in full in Section 2.1.2) raises support levels for all working-age solidarity benefit recipients.
Figure 2.7 illustrates existing benefit levels and those proposed under Scenario B for a selection of households. A workless, single‑parent, one‑child household would see their equivalised income rise from EUR 375 to 403. A workless couple, two‑child household would see their equivalised income rise from EUR 381 to 413. In general, larger households have higher equivalised incomes from basic out of work support because of the additional amount of support per household member is greater than the equivalisation factor used to calculate equivalised incomes. The exception, of course, is large households that would be affected by the proposed benefit cap in Section 2.1.1. For all households, the higher equivalised incomes under Reform Scenario B would still fall short of pensioner solidarity support levels.
Equivalised income levels with existing and proposed solidarity benefit (Reform Scenario B) for selected out-of-work households compared to poverty thresholds, 2025
Note: Out-of-work benefit income includes basic solidarity support (minimum income support) and Child Benefit. Households are not entitled to any other benefits or any solidarity supplements. Children in selected households are aged between 8 and 14. Incomes equivalised using OECD-equivalence scale which assigns a weight of 0.5 to adults and children aged 14 and over, and 0.3 to all other children. Reform Scenario B simplifies solidarity benefits and increases the level of support by 10%. Poverty thresholds are set at a proportion of median equivalised household income. At-risk-of-poverty thresholds uprated in line with real GDP growth projections.
Source: OECD calculations; Eurostat, at-risk-of-poverty thresholds (2023); OECD Economic Outlook, December 2024.
Households with additional needs would receive more support under all scenarios
The simplification of the solidarity benefit under all reform proposals would ensure that all eligible households would be entitled to additional support when facing additional needs. Under the current system, households entitled to more generous solidarity benefits (pregnancy risk benefit, specific risk benefit, hospital travel benefit, newborn hospital benefit, pregnancy termination benefit, social parental benefit and social adoption benefit) can have this benefit completely offset by a lower amount of minimum income benefit (Rendimento Social de Inserção, RSI). This results from the fact that larger households are entitled to a higher amount of RSI compared to the other solidarity benefits listed above which are paid a single rate. In essence this limits receipt of these higher-value additional support benefits to single adult households. For example, as Table 2.6 shows, a single parent with two children is currently entitled to EUR 474.51 from the current minimum income benefit (RSI). However, if this household is also entitled to the Pregnancy Risk benefit, any Pregnancy Risk benefit received would be deducted from their RSI award. In effect, under the current system, they receive no additional income compared to any other single parent with two children and no pregnancy risk.
Under the proposed reforms, a single benefit with supplements would mean that larger households would not be unfairly penalised when entitled to support for additional costs. Combining the different existing solidarity benefits as supplements under a single benefit means that the additional support that someone receives with a supplement under the proposed new benefit would not be deducted from other benefit support as is the case under the current system. Table 2.6 shows that under the proposed reforms (for all scenarios), the single parent entitled to the pregnancy risk supplement would be entitled to a higher amount of support: EUR 644.67 (EUR 474.51 basic support and EUR 170.16 supplement).
Solidarity benefit entitlements for a single parent with two existing children and a pregnancy risk and no other income, current system and proposed simplified solidarity benefit
Source: OECD calculations.
Higher benefit levels and improved work incentives would boost incomes the most for the poorest individuals
The reforms would see some individuals better-off, with higher monthly incomes, and other individuals worse‑off. While the reforms are, designed to provide increased support, certain aspects of the simplification of benefits lead to some households being worse‑off, although in all cases households would always have income equal to or higher than the minimum income benefit level.
The poorest individuals, with incomes in the lowest income decile, are most likely to benefit from all of the proposed reforms, as shown in Figure 2.8. Almost all working-age households affected by the reforms in Scenarios A, B, and C would either see their household incomes unchanged or increase as a direct result of the reforms (Panels A1, B1, and C1 respectively in Figure 2.8). Under Scenario B, increased levels of support, 4% of individuals in working-age households would be better-off, rising to 36% of individuals in the poorest income decile. Those affected in the poorest income quintile would be better-off by an average of EUR 61 per month. Under Scenario C, improved work incentives, 2% of individuals in working-age households would be better-off, rising to 21% in the poorest income quintile (who would be better-off by an average of EUR 47 per month). Scenario D, a broader harmonisation of solidarity benefits, would result in a greater number of winners and losers compared to the other scenarios. Specifically, 2% of working-age individuals would gain from the proposed reform with an average gain of EUR 117, mostly as a result of the new benefit-household definition (as described in Section 2.1.4). In contrast, 3% of working-age individuals would lose‑out, mostly as a result of the harmonisation of social unemployment benefit with minimum income support, with an average income loss of EUR 103 per month, although with a lower average loss of EUR 57 per month for individuals in the poorest income decile. Harmonisation of support will leave households worse‑off where they have a small amount of other income. This is because of the smoothing of work incentives through a single income deduction rate (discussed in more detail in Section 2.2.3).
A small number of pensioner-only households may see their incomes fall somewhat
While pensioner incomes will never fall below the existing CSI benefit levels under any of the proposed reform scenarios, reforming the Social Pension does lead to some income loses in all scenarios. This is due to minor differences in the entitlement rules for the Social Pension and pensioner minimum income benefit (Complemento Solidário para Idosos, CSI).
The minor income loses are driven by the fact that, under the proposed reforms additional resources are taken into account when determining eligibility for CSI which are not taken into account for the Social Pension. For instance, under the current rules, individuals or couples can have a limited amount of earnings (up to EUR 209 per month) alongside receiving the Social Pension and in a rare number of cases can be better-off claiming the Social Pension compared to CSI. In addition, some other solidarity benefits can be accumulated with the Social Pension, such as the Widow’s pension, which, again in rare cases, can result in a higher income compared to CSI alone.
The majority of pensioner-only households (Panels A2, B2, and C2 in Figure 2.9) experience almost no income changes under Reform Scenarios A, B and C, with the small number of losers (less than 0.2%) due to the reform of the Social Pension.
Less than 0.5% of individuals in working-age households would see their income fall under Reform Scenarios A, B and C as a result of changes to the Social Pension impacting mixed-aged households which are included in the working-age household category. All mixed-age households that would experience income falls are in the third-lowest income decile, and the average income loss would be EUR 39 per month.
Scenario D has some mitigating positive effects on pensioner households, with around 1% of pensioners gaining from the reform (with an average gain of EUR 74 per month respectively). The losers are the same as under Reform Scenarios A, B, and C. The small number of pensioner households gaining are a result of the changed earnings deduction rate. Under the current CSI benefit, earnings are deducted at a rate of 100%, however under the simplified benefit earnings would be deducted at the same rate as for working-age households: 80%. For the small number of pensioner households with earnings, this leaves them with more income.
Share of pensioner and non-pensioners by income decile that gain or lose from the proposed reforms, 2024
Note: Deciles 6 to 10 shown together. OECD equivalisation scales. Effect on household income per individual shown. Working-age households include mixed-age households that include both pensioners and working-age adults. Estimates do not include behavioural effects. See Annex 2.B for full details on modelling process.
Source: OECD analysis using Euromod.
Increasing benefit levels provides the most progressive impact on incomes
Taking together the number of winners and losers of the proposed reforms, and the average gain and loss they would experience, provides an overall distributional income effect. Figure 2.9 summarises the simulated income effects of the reform scenarios across the income distribution. It shows the average income change for the population, by income decile, relative to the current benefit system.
Scenario A, simplification of benefits, has little impact on average incomes.
Scenario B, increasing solidarity benefit levels by 10%, has the largest income effect on the poorest decile of households, increasing their average annual incomes by EUR 73 per year.
Scenario C, reducing the earnings deduction rate by 10 percentage points, also has a substantial impact on the lowest-income households. The proposed reform increases average incomes for the poorest decile of households by EUR 33 per year.
Scenario D, the extended solidarity benefit reform, would affect the incomes of more households across the income distribution. The poorest decile of households are better off, on average, by EUR 21 per year, while households in deciles 2 to 5 would experience, on average, income losses of EUR 13 per year.
Average annual income change from solidarity benefit reform scenarios, by equivalised income decile, 2024
Note: Deciles 6 to 10 shown together. OECD equivalisation scales. Effect on household income per individual shown. Estimates do not include behavioural effects.
Source: OECD analysis using Euromod.
Household with children are the main beneficiaries from proposed reforms
Households with children tend to benefit most from Scenario B (poverty reduction) and Scenario C (work incentives) reflecting the fact that households receiving solidarity benefits have more children on average than the rest of the population (Figure 2.10). Across the population, the annual income effect for children from Scenario B, higher benefit levels, is twice that of any other age‑group (EUR 20 for children aged 0‑14 compared EUR 10 for those aged 15‑24 and EUR 8 for those aged 25‑49).
As explained above, there are some small income losses for pensioners under Scenarios A, B, and C as a result of the closing of the Social Pension benefit. which means that a small number of pensioners with other income sources would lose out from the reforms. The majority of existing recipients of the Social Pension would not see their income change as they are already entitled to more support under the Pensioner Minium Income Benefit. A small number of pensioners, around 8 000 or 0.3% of all pensioners, would see lower incomes, on average by EUR 289 per year for those affected and by EUR 2 per year across all pensioners. This is because the means-test for the Pensioner Minimum Income Benefit is slightly stricter than for the Social Pension which is discontinued under the reforms. As a result, only households with other income sources or financial assets would receive less support.
Again, Scenario D (extended harmonisation of solidarity benefits) has a different profile of impacts across age groups compared to the other scenarios. Adults and children in working-age households would see an income loss, on average, because of the change to how earnings affect solidarity benefit entitlement. The new, smoother, deduction of earnings from benefit entitlement compared to the all-or-nothing income limit for Social Unemployment Benefit, means some households with low-levels of earnings are worse‑off (Figure 2.16 below illustrates this in more detail). Pensioners aged 65 to 79 would see slightly higher incomes on average from Scenario D (extended solidarity benefit) because the reforms increase the incomes of households with a mix of working-age and pensioner individuals compared to the current system. This is because the system under Scenario D would not deduct pensioner support from working-age support, for example for a couple where one person is aged over 65 and the other is aged under 65.
Average annual income change from solidarity benefit reform scenarios, by age group
Note: Effect on household income per individual shown. Estimates do not include behavioural effects.
Source: OECD analysis using Euromod.
2.2.3. Effectiveness of solidarity benefits
The main objectives of a minimum income benefits and other related support are to reduce the extent of poverty while encouraging self-sufficiency with strong work incentives. To maximise the efficiency of these benefit, support should be well-targeted towards low-income households.
Reforms would directly enhance the poverty-reducing effectiveness of benefits
Portugal has one of the least effective benefit systems in the EU in terms of reducing poverty levels. Social benefits reduce the at-risk-of-poverty rate (at 60% of median income) in Portugal by just 2 percentage points, the lowest poverty impact in the EU. The effect on the depth of poverty is similarly low, reducing the average income gap to the poverty threshold by 5 percentage points, less than half the EU-average effect (as shown in Chapter 1, Figure 1.2). The proposed reforms outlined above would have an immediate and direct impact on poverty by changing the level of support available and could be expected to reduce poverty still further in the longer-term if a simplified benefit system is successful in increasing take‑up of support.
The direct, immediate, impact on poverty of the proposed reform scenarios is anticipated to be relatively minor. In terms of immediate effects, increasing the level of support – Scenario B – and reducing the earnings deduction rate – Scenario C – would be most effective at reducing at-risk-of-poverty rates (income below 40% of median income). Figure 2.11 shows that Scenario B and Scenario C would reduce the rate of poverty in Portugal by 0.10 and 0.06 percentage points respectively (equivalent to 10 000 and 6 000 fewer individuals in poverty). This change is relatively small compared to the poverty data which shows the rate of poverty at 40% of median incomes at 6.8% in 2023. Scenario B and Scenario C would also reduce poverty at the 50% of median incomes threshold, although to a lesser extent, by 0.09 and 0.02 percentage points. On the other hand, Scenario A and Scenario D would not reduce these measures of poverty because the level of benefit support would remain approximately the same for non-working households on the lowest incomes. All scenarios would cause a very small increase in poverty at the 60% of median income definition. Scenario A, with the largest increase in poverty of 0.02 percentage points, represents 2 000 more individuals with income below the 60% poverty threshold. This small increase is a result of the small number of pensioners who lose income as a result of reforming the Social Pension as discussed above.
Immediate effect of reform scenarios on at-risk-of-poverty rates by poverty threshold level
Note: Estimates do not include behavioural effects.
Source: OECD analysis using Euromod.
Across the EU, Portugal has the seventh lowest level of minimum income benefit support, although total income support for workless households with children is typical among OECD countries (OECD, 2024[3]). These low levels of existing support mean that beneficiary households are relatively far from the 60%-at-risk-of-poverty threshold (see
Figure 2.7 above), and the changes proposed under the reform scenarios are insufficient to raise incomes above this higher threshold. This explains why the reforms only reduce poverty at these lower income thresholds.
The poverty gap, however, which measures the median distance between household income and the poverty line, is reduced by three of the four reform scenarios. The at-risk-of-poverty gap is a measure of the average depth of poverty and is calculated as the difference between the median equivalised income of people below the 60%-at-risk-of-poverty threshold, expressed as a percentage of the at-risk-of-poverty threshold (Eurostat, 2025[4]). It is an additional measure of low incomes alongside the rates of poverty discussed above which are calculated as the percentage of people with incomes below the at-risk-of-poverty threshold.
Figure 2.12 shows the effect of the proposed reforms on the poverty gap. Scenario B – increasing basic rates of support – has the greatest impact on the poverty gap and would reduce it by 0.35 percentage points. Scenario C – reducing the earnings deduction rate – has a smaller impact, reducing the poverty gap by 0.26 percentage points. Scenario A – simplification – has no direct effect on the poverty gap, while Scenario D – extended simplification – would have the small effect of reducing the poverty gap by 0.05 percentage points. Scenario D is more effective at reducing the poverty gap compared to Scenario A because it provides more support to mixed-aged households and some households receiving supplements as mentioned above in Section 2.2.1.
Immediate effect of reform scenarios on the at-risk-of-poverty gap, as a percentage of 60%-at-risk-of-poverty threshold, 2023
Note: Estimates do not include behavioural effects. The at-risk-of-poverty gap measures the median income of people with income below the at-risk-of-poverty threshold, as a percentage of the at-risk-of-poverty threshold. A fall in the poverty gap therefore reflects a rising median income for people in poverty.
Source: OECD analysis using Euromod.
The proposed reforms would not significantly change poverty rates in Portugal when compared internationally. The reforms only provided limited additional support to households receiving solidarity benefits, and their immediate effects on incomes are relatively small. As a result, the overall poverty rates for Portugal do not change significantly when compared alongside other countries. Figure 2.13 shows that while the poverty gap under Scenarios B, C and D would fall, the rate would not change significantly compared to other countries: the poverty gap in Portugal would remain the 7th highest in the EU at 25%.
At-risk-of-poverty gap in the EU by country with Portugal variants shown for different benefit reform scenarios, as a percentage of 60%-at-risk-of-poverty threshold, 2023
Note: For the purposes of comparison, the impact of the proposed reforms has been modelled in 2024 but used to show the relative change to the poverty gap in 2023. Estimates do not include behavioural effects. The at-risk-of-poverty gap measures the median income of people with income below the at-risk-of-poverty threshold, as a percentage of the at-risk-of-poverty threshold. A fall in the poverty gap therefore reflects a rising median income for people in poverty.
Source: OECD analysis using Euromod and Eurostat, relative risk of poverty gap (ilc_li11).
Increases in benefit take‑up would have a greater impact on reducing poverty
In addition to immediate poverty impact, the proposed reforms aim to have a greater longer-term impact on reducing poverty by improving take‑up of solidarity benefits. This is because higher take‑up would increase coverage of solidarity benefits in Portugal (as discussed in Section 2.2.1).
In addition to the direct poverty-reducing effects of the reforms, higher take‑up would mean that the reforms have a more substantial impact on reducing rates of severe poverty. Take‑up of social benefits may be inhibited by stigma associated with claiming benefits, lack of information about available benefits or entitlement rules, and an unwillingness to fulfil activation requirements (Immervoll and Knotz, 2018[5]; Bargain, Immervoll and Viitamäki, 2010[6]; Eurofound, 2015[2]). Automatically informing eligible recipients of their entitlements (Tempelman and Houkes‐Hommes, 2015[7]), and using trigger points to provide benefit information can help reduce take‑up (Finn and Goodship, 2014[8]).
Figure 2.14 shows that with increase in take‑up, Scenario A would reduce the at-risk-of-poverty rate (40% of median income) from 6.8% by 0.3 percentage points. The other scenarios would reduce levels of deep poverty by 0.4 percentage points. The simulations underpinning these impacts assume a take‑up rate of 74% (an increase of 3 percentage points as compared to current take‑up levels of 71%). This increase would raise the take‑up rate of benefits in Portugal to be in line with the highest take‑up levels for comparable benefits across the EU (Eurofound, 2015[2]).
Effect of reforms scenarios on poverty rates by poverty threshold level, assuming a three percentage point increase in take‑up of solidarity benefits
Note: OECD equivalisation scales. Effect on household income per individual shown.
Source: OECD analysis using Euromod.
Benefit support would be better targeted under proposed reforms
The proposed solidarity benefit reforms would improve the targeting of benefit support towards the poorest households. However, solidarity benefit spending would still be dwarfed by spending on contributory benefits. As a result, the overall benefit system would remain relatively untargeted compared to other European countries.
Reform scenarios marginally increase the share of solidarity support that goes towards the poorest households (putting aside take‑up effects which remain uncertain). Half of existing solidarity benefit spending (48%) is targeted towards the poorest 10% of households under the current system (Figure 2.15). Reform scenarios would marginally increase this share by between 0 and 3 percentage points by providing more support to lower income households (as shown in Section 2.2.2 above). Solidarity support received by richer households persists under the reforms as the more generous support for pensioners remains in place and the benefit household definition continues to shield pensioner solidarity support from earnings deductions from working-age household members. This makes it possible for a household with a high income from working-age adults to still receive solidarity benefits for pensioners living in the same household.
Portugal already has a well-targeted social benefit system compared to other EU countries. As of 2023, 45% of all social benefit spending is received by the poorest quintile of households in Portugal, the third highest percentage in the EU (as shown in Chapter 1, Figure 1.11).14
Overall benefit spending in Portugal is poorly targeted, however. This is because the effect of social benefits is very small in comparison to contributory benefit. Portugal has a very large contributory benefit system: contributory benefit spending is four‑times higher than social benefit spending, the third highest ratio in the EU). In addition, the contributory benefit spending is especially poorly targeted by household income: 43% of non-pensioner contributory benefit income in Portugal is received by the richest fifth of working-age households, the second highest rate in the EU (as shown in Chapter 1, Figure 1.14).
Percentage of total solidarity benefit spending entitlement by equivalised income decile for baseline system and proposed reform scenarios
Note: Solidarity spending includes all solidarity benefits. Estimates do not include behavioural effects. Analysis assumes full take‑up of benefits to show effect of targeting rather than take‑up effect.
Source: OECD analysis using Euromod.
The reforms to solidarity benefits set out in this report are not enough, by themselves, to have a substantial impact on the targeting of the overall benefit system in Portugal and the poorest decile would continue to receive a relatively small proportion of benefit spending. The share of all benefit spending (including contributory benefits) targeted at the poorest tenth of households would remain unchanged at 6.5% of benefit spending.15 This share of spending is around two‑thirds that of average spending across the income distribution (which by definition is 10% of spending in each income decile). The limited impact on targeting of the simulated scenarios results from i) the large proportion of spending directed to contributory benefits; ii) the relatively limited fiscal space for reforms, which restricts their magnitude and iii) the low levels of take‑up, which limits the reach of solidarity benefits among those eligible.
Work incentives are dramatically improved by simplification reforms
The proposed reforms will improve work incentives for solidarity benefit households through two channels. First, all the proposed reforms will smooth work incentives for all benefits that are included in the scenarios. Second, Scenario C – reducing the earnings deduction rate – directly improves work incentives by making it financially more beneficial to work more hours while receiving solidarity benefits.
The smooth work incentives of the proposed solidarity benefit (all scenarios) are in contrast to many existing solidarity benefits, including Social Unemployment Benefit, Social Adoption Benefit, Social Pregnancy Risk Benefit, and other social benefits relating to newborn children. For these existing benefits, individuals are entitled to the same amount of solidarity support up until the household income reaches a fixed threshold, after which the individual is not entitled to any support. The all-or-nothing support as a result of the fixed-income thresholds of existing benefits means that households can be worse off when working more hours, effectively disincentivising work.
Under current benefits individuals can be discouraged from working full-time as they are better-off working part-time instead. Figure 2.16 shows the example of a single adult entitled to social unemployment benefit who is paid an hourly rate equivalent to the minimum wage. While they work fewer than 19 hours per week, they maintain their full social unemployment benefit (Figure 2.16Panel A). However, once they work more hours than this, they lose all their entitlement. As a result, their overall income is lower when working between 19 and 38 hours (Panel B). In this case, the single adult is worse‑off working full-time at the minimum wage compared to working 18 hours per week.
All the proposed reform scenarios remove these fixed-income‑threshold benefits under the single solidarity benefit meaning that all solidarity benefit households would always be financially better-off working more hours. This is because all support would be subject to a single smooth earnings deduction rate. Figure 2.16 shows how the reformed benefit entitlement would compare to the current benefit for a single adult entitled to social unemployment benefit and the unemployment supplement under extended solidarity benefits (Scenario D). While the exact income threshold for existing benefits varies according to the size of the household, the effects, and the impact of the proposed reforms are similar.
Solidarity benefit income and total household income for a single adult entitled to Social Unemployment Benefit under existing system and reformed simplified solidarity benefits, 2025
Note: Income includes solidarity benefit and earnings net of social contributions and income tax. Single adult shown is entitled to Social Unemployment Benefit, but values are the same for any single adult entitled to Social Adoption Benefit or any other solidarity benefit with a fixed income entitlement threshold.
Source: OECD calculations.
In addition to smoothing work incentives, Scenario C would further improve work incentives by lowering the earnings deduction rate. This reform would unambiguously incentivise solidarity benefit recipients to work more hours. Longer-term behavioural effects could also increase the income of those households that work more hours and generate a saving to government spending on benefits (which could partially offset the direct costs of this reform).
A lower earnings deduction rate increases incomes for in-work solidarity benefit households and lowers the participation tax rates they face. Figure 2.17 illustrates the effect of a lower earnings deduction rate on household income (Panel A) and participation tax rates (Panel B) for an example single‑parent, two‑child household. It compares Reform Scenario A, with a deduction rate of 80%, with Reform Scenario C, with a deduction rate of 70%. Panel A shows that a single parent working at the minimum wage, either part-time or full-time, would gain from the reform. Households working more hours while still being entitled to solidarity benefit would gain the most. Panel B shows how the higher income that comes from working more hours is reflected as a lower participation tax rate. Participation tax rates show the percentage of gross earnings that is, in effect, lost through offsetting lower benefits and social contributions. For instance, under Reform Scenario A, a single parent with two children working full-time at the minimum wage has a disposable income (including 221 of Child Benefit) of EUR 1 124 per month from a gross wage (before taxes and social contributions) of EUR 1 015.16 This compares to an income of EUR 881 from benefits when not working. Therefore, the participation tax rate is the difference between the gross wage and the change in income (EUR 769 minus EUR 243) divided by the gross wage (EUR 769), or 76%. Under Reform Scenario C, with the lower benefit deduction rate of 70%, the participation tax rate for full-time work at the minimum wage for a single parent with two children is reduced from 76% to 73%.
Scenario C would have a similar impact on work incentives for other households, with effects extending further up the earnings distribution for larger households. This is because solidarity support extends further up the earnings distribution for larger households, such that the proposed 10 percentage point in the earnings deduction rate would affect participation tax rates for earners above the minimum wage. In addition, households with children would continue to have higher participation tax rates than those without as a result of the means-testing of Child Benefit.
Income (Panel A) and participation tax rates (Panel B) for a single parent with two children entitled to single ordinary solidarity benefit under reformed system (Scenario A), and reformed system with a reduced earnings deduction rate (Scenario C), 2025
Note: Income includes solidarity benefit, child benefits and earnings net of social contributions and income tax. Children aged 4 and 6.
Source: OECD calculations.
At 96%, Portugal currently has the highest PTR in the EU for a two‑adult, three‑child household where one adult is working full-time at the minimum wage. Changes to the earnings deduction rate under Scenario C would bring financial work incentives in Portugal closer in line with average work incentives for social assistance benefits across the EU. Implementing reform Scenario C would reduce the participation tax rate (PTR) for an example two‑adult, three‑child family from the extremely high level of 96% down to 87%. This remains high but is significantly closer to the EU average rate of 70% (Figure 2.18). Alongside the high earnings deduction rate applied to the Minimum Income Benefit (80%), Portugal’s high PTR for low-earnings families is driven also by the addition deduction of social insurance contributions (11% of gross earnings), and the reduced Child Benefit level for those with some earned income.17
Participation tax rates for a couple with three children entitled to social assistance benefits, working full-time at the minimum wage, with and without proposed Portuguese reform, EU, 2024
Note: Portugal reform shows participation tax rate under Scenario C which reduces the earnings deduction rate for solidarity benefits from 80% to 70%. Children are aged 4 and 6. Participation tax rate compares situation of one‑adult working full-time to both adults not working. Household assumed to receive full entitlement of social assistance benefits, child benefits and in-work benefits. Contributory benefits and childcare benefits not included. Taxes and social contributions included in the calculation. Countries without a single minimum wage not shown and not included in the EU average figure.
Source: OECD tax-benefit model. Model version 2.7.0.
In addition to the immediate effect of the reforms on work incentives, there is likely to be some longer-term behavioural effects as households respond to the new work incentives by changing how many hours they work. By smoothing the marginal effective tax rates, all the scenario’s outlined above would reduce the distortions of solidarity support on work incentives. The magnitude of the behavioural effect of such smoothing is, however, uncertain. While, for some individuals working part-time, the reform strengthens the incentive to work full-time, for others it will reduce the incentive to work part-time compared to not working at all.
Reducing the earnings deduction rate (Scenario C), however, would likely have an unambiguously positive impact on labour supply. This is because the reform would improve both the incentive to take‑up work (the extensive margin of labour supply), and the incentive to work more hours (the intensive margin of labour supply) for all benefit recipients at every point of the earnings distribution.
2.2.4. Fiscal impacts
The proposed reforms set out in this report all have immediate fiscal consequences as a result of changing benefit entitlements. Based on administrative data, the government currently spends EUR 937 million per year across all solidarity benefits, of which EUR 427 million is spent on working-age solidarity benefits which are included in the reforms under Scenarios A, B and C (all solidarity benefits are included under Scenario D).
Of the four reform scenarios covered in this report, Scenario A and Scenario D result in a reduction in government spending on benefits before any changes in benefit take‑up, and Scenario B and C result in an increase in government spending (as set out in Table 2.7). An increase in take‑up of solidarity benefits of 3 percentage points would also result in additional spending. None of the scenarios have any direct impacts on other social transfer spending, taxes or social contributions as solidarity benefits do not affect these other fiscal areas.
Scenario A, simplified solidarity support, would result in a direct saving of EUR 6 million per year as a result of changes to benefit entitlement (a reduction in spending on all solidarity benefits of 0.7%). This saving is mostly due to the lower entitlement for pensioners resulting from changes to the Social Pension.
Scenario B, which increases the basic level of support, is the costliest reform option and would cost EUR 41 million per year (4.5% of solidarity benefit spending). The additional cost compared to Scenario A is a direct result of increasing the basic level of support by 10% which benefits all recipients of the new single solidarity benefit.
Scenario C, reducing the earnings deduction rate by 10 percentage points, would cost EUR 15 million per year (1.6% of solidarity benefit spending). The additional costs result from increased support for households with low amounts of earnings. In addition to these immediate fiscal costs, Scenario C may also generate fiscal savings in the longer term. This is because the improved incentive to work would be expected to increase the earnings income of benefit recipients and reduce their entitlement to solidarity support. Such longer-term behavioural impacts are difficulty to quantify and beyond the scope of this report.
Scenario D, the extended solidarity benefit reform, has the greatest savings, reducing government spending on benefits by EUR 40 million per year (1.9% of solidarity benefit spending). The additional saving compared to Scenario A is a result of the smooth earnings deduction compared to the all-or-nothing entitlement under the current system for Social Unemployment Benefit.
The fiscal costs of the proposed reforms are relatively small compared to the overall level of government spending in Portugal (general government spending was EUR 113 billion in 2023). The most expensive reform (Scenario B) represents an increase of just 0.04%. More generous support would strengthen the poverty reduction capacity and impact on targeting of such reforms.
In addition to the immediate, direct costs of the reforms, Table 2.7 also shows the potential additional costs that would result from an increase in take‑up of solidarity benefits. Any increase in take‑up because of the proposed reform options would be very uncertain. The additional cost for government of higher take‑up would be EUR 42 to 43 million across all four options (an increase in solidarity benefit spending of around 4.5%).
Estimated costs of reform scenarios, millions of euros
Note: Figure assume no behavioural effects other than increased take‑up. Cost to government shown, negative figures mean reduced spending. Figures in parentheses show percentage change in overall spending on solidarity benefits, which is assumed to be EUR 1 792 million based on administrative data for 2023.
Source: OECD analysis using Euromod.
2.2.5. Wider benefit design impacts
In addition to the impacts of changes to benefit entitlement rules, the proposed simplification of solidarity benefits also creates opportunities for other policy and operational changes which can have additional positive impacts.
A simplified solidarity benefit would facilitate further reforms to the conditionality, or work-search conditions, applied to benefit recipients. Under current solidarity benefits, not all working-age individuals in households in receipt of Social Unemployment Benefit and other solidarity benefits are required to search for work as a condition of receiving benefits. Combining solidarity support together creates an opportunity to apply work search conditions to all work-capable individuals in a household regardless of the supplements being claimed. For example, under the current system, neither member of a couple receiving Social Parental Benefit are required to search for work, and the partner of an adult receiving Social Unemployment benefit is also exempt from work search requirements. However, under a single system, while a mother of a newborn child or a disabled person would not be required to search for work, their partner would, by default, be required to undertake work search activities, and partners of recipients of Social Unemployment Benefit would also be required to look for work.
Work-search rules could be further tightened to discourage claimants in larger households from opting out of solidarity benefits. While current minimum income benefit rules mean that work-search conditions apply, in theory, to all adults living in a household, individual adults can, in effect, opt out of the benefit and the work search conditions. For example, in a two‑adult household one adult can opt out of claiming: while the household receives less support, only one adult is required to look for work. Under a simplified benefit, a more restrictive work-search regime would ensure that all working-age adults in the household who are capable of work are required to complete work-search activities for the household to receive any support.
A simpler benefit system creates opportunities for processing efficiencies. In combination with digital reforms, the proposed reforms could reduce the operational costs of running the benefit system. In Portugal administration costs for social protection amounted to EUR 912 million in 2022, representing 1.52% of actual social protection benefit expenditures.18 While lower than the EU average of 2.95%, there is scope to move towards countries with the lowest administration cost share of around 1% (as in Cyprus, Greece and Malta). An example of one area of administration under which simplification could reduce costs is in overpayments. A simplified system can facilitate administration and digitisation and hence tackle overpayments, reducing the need for administrative agents to engage with benefit recipients multiple times.
Simplification of solidarity benefits could also work in tandem with improvements to the digital infrastructure. A more streamlined digital claim management system, as is being developed under a separate RRP project (European Commission, 2023[9]), could calculate benefit entitlement on a timelier and more accurate basis. This could, in combination with the simplified solidarity benefit reforms, also reduce overpayments and could generate other efficiency savings.
References
[6] Bargain, O., H. Immervoll and H. Viitamäki (2010), “No claim, no pain. Measuring the non-take-up of social assistance using register data”, The Journal of Economic Inequality, Vol. 10/3, pp. 375-395, https://doi.org/10.1007/s10888-010-9158-8.
[2] Eurofound (2015), Access to social benefits: Reducing non-take-up, Publications Office of the European, https://www.eurofound.europa.eu/en/publications/2015/access-social-benefits-reducing-non-take.
[9] European Commission (2023), Digital transition of Social Security services, https://commission.europa.eu/projects/digital-transition-social-security-services_en.
[4] Eurostat (2025), Glossary:At-risk-of-poverty gap, https://ec.europa.eu/eurostat/statistics-explained/index.php?title=Glossary:At-risk-of-poverty_gap.
[8] Finn, D. and J. Goodship (2014), Take-up of benefits and poverty: an evidence and policy review, https://www.researchgate.net/publication/272786560_Take-up_of_benefits_and_poverty_an_evidence_and_policy_review.
[5] Immervoll, H. and C. Knotz (2018), “How demanding are activation requirements for jobseekers”, OECD Social, Employment and Migration Working Papers, No. 215, OECD Publishing, Paris, https://doi.org/10.1787/2bdfecca-en.
[3] OECD (2024), Assessment of the existing social benefit system in Portugal (unpublished).
[1] OECD (2022), TaxBEN: The OECD tax-benefit simulation model, methodology, user-guide and policy applications, OECD Publishing, Paris, https://www.oecd.org/social/benefits-and-wages/OECD-TaxBEN-methodology-and-manual.pdf.
[7] Tempelman, C. and A. Houkes‐Hommes (2015), “What Stops Dutch Households from Taking Up Much Needed Benefits?”, Review of Income and Wealth, Vol. 62/4, pp. 685-705, https://doi.org/10.1111/roiw.12197.
The definition of a household benefit unit (also called a tax unit or benefit household) has a significant effect on how much support the benefit system provides to larger, more complex, households beyond the headline benefit levels. Benefit unit definitions can be applied to the level of benefit support and separately to the means-testing of the support, and the definition can vary for different benefits a household may be receiving. The choice of which benefit unit definitions are used can also affect work incentives. This annex considers the effect of different benefit unit definitions in more detail. It looks at the general case of different definitions as well as the specific impacts on the single solidarity reforms described in this report.
Different household benefit unit definitions
A benefit unit definition for a given benefit is concerned with which people living in a household are relevant when calculating entitlement to a benefit. The most straightforward household definitions are at two extremes. The broadest definition includes all people living in the same household, the narrowest definition considers every individual as a separate benefit unit. In between these extremes are definitions that include dependent children as part of the same benefit unit as their parents, spouses and partners as the same benefit unit, or all persons below pension age as the same benefit unit.
It should be apparent that the larger and more complex the household situation the more scope there is for multiple benefit unit definitions. Simple households with one or two adults are more straightforward and less likely to be classified differently under different benefit unit definitions.
Typically, contributory benefits consider individuals separately whereas means-tested benefits consider a broader definition of a benefit unit. Different countries apply different benefit unit definitions for minimum income benefits.
Confusingly, the benefit unit definition can also vary within the rules for a benefit. For instance, a benefit may be payable to an individual adult at a fixed level. However, their entitlement to that benefit may depend on the household income. For instance, the social unemployment benefit in Portugal is payable to individual adults but depends on the income of the household. It is worth noting that it is nonsensical to operate the reverse, where a benefit where support is calculated across a broad benefit unit definition, but the means-test operates at a narrower, say individual, level.
Further complication arises around what income is relevant for these cases where multiple individuals in the same household can claim the same (or similar benefits), but the means-test is operated at a household level. For instance, the social unemployment benefit considers other social unemployment benefit received by other adults as relevant income for the means-test. In a household with two social unemployment claimants then, the first recipient will receive the full amount, but the second recipient will receive a lower amount.
The different ways in which benefit unit definitions can be applied to benefits can make for a very complicated system of support, especially for larger households. A more consistent approach to benefit unit definitions across social benefits makes a simpler and easier to understand system. In turn this is likely to improve work incentives as financial implications of working are clearer.
Effect on support levels
Broader benefit unit definitions typically mean a household receives less support in total. This is for two reasons.
First, means-tested benefits usually provide a higher level of support for the first adult in a household, and a lower level of support for subsequent adults and children. This recognises the economies of scale that larger households enjoy, such as benefiting from sharing space, heating bills, and meals.
Second, a broader benefit unit definition for the means-test includes the income of more persons in the household. If these other persons have other income sources, then that will reduce the level of support available. If fewer people are included, then the amount of benefit entitlement is likely to be higher.
Effect on work incentives
Having different benefit unit definitions also affects work incentives. The overall effect of broader or narrower definitions is ambiguous as it can improve work incentives for some individuals and households while reducing the incentive to work for others. This is due to two competing effects.
First, broader benefit unit definitions (especially regarding the level of support) typically decrease the level of support available to the whole household (as discussed above). This means that a household has a lower amount of benefit to lose if an individual increases their earnings. Therefore, with a broader benefit unit definition, individuals are likely to have a greater incentive to work.
Second, a broader benefit unit (especially when regarding means-testing) means that any additional individual earnings will affect a greater share of all the benefits received by the household. If the benefit unit is defined at the individual level, then individual earnings do not affect benefit receipt of others in the household. Therefore, a broader benefit unit means a greater amount of benefit that could be lost as earnings increase and individuals are likely to have a reduced incentive to work.
Which work incentive effect is strongest depends on the household situation. In household where more adults are already working, it is more likely that the first effect will be stronger, and a broader benefit unit definition will strengthen work incentives. In cases where few or no adults are working, it is more likely that the second effect will be stronger, and a broader benefit unit definition will weaken work incentives.
Given the majority of households receiving minimum income benefits are not working it is likely that a broader benefit unit definition creates weaker work incentives.
It is worth noting that a benefit cannot “cherry pick” the rules to maximise work incentives. As mentioned above it would be nonsensical to have a broad benefit unit definition for the benefit level, with a narrower benefit unit definition for means-testing support. The reverse however is possible. A benefit can minimise work incentives by having a benefit level calculated individually and by having a means-test that includes the income of the entire household.
Interaction effects
As different benefit rules can have different benefit unit definitions, it is possible to have many different scenarios where households may be disentitled to one benefit because of income from another benefit. The scale of this complexity increases with the number of different benefits available and is especially true when income from some benefits is included in the means-test for others. While these cross-benefit means-testing rules prevent households from claiming substantially large amounts of benefits in total, they further increase the complexity of the benefit system.
Whole household and separate pensioner household definition for the working-age simplified solidarity benefit (Scenarios A, B and C)
This section of the annex considers two different household benefit unit definitions for the working-age single solidarity benefit set out in Section 2.1 above. These are:
The whole household benefit unit approach (Approach A) which is the definition applied under Reform Scenarios A, B and C. This approach includes pensioners and their income when calculating the single solidarity benefit.
The working-age household benefit unit approach (Approach B) is an alternate approach. This approach excludes pensioners and their income when calculating the single solidarity benefit. Pensioners are then classified as separate benefit units.
These different approaches only affect households with a mix of working-age adults and pensioners. In a household with only pensioners, then they will be entitled to the more generous benefit and have nothing to gain from the single solidarity benefit. In the case of a household with no pensioners, then there is no pensioner income to consider.
Annex Table 2.A.1 illustrates the effect of the different household rules for an example household with working-age adults and pensioners. The example household has a single pensioner receiving Pensioner Social Assistance Benefit (CSI), a workless working-age couple, and two children. In household benefit unit approaches the pensioner receives a monthly CSI amount of EUR 550.67. Under the whole household definition (Approach A), the pensioner is included in the calculation of the maximum entitlement to the single solidarity benefit, whereas they are not under the working-age household definition (Approach B). This means that the household is notionally entitled to EUR 806.65 of single solidarity benefit under Approach A compared to the lower amount of EUR 640.58 under Approach B. However, Approach A also fully deducts the pensioner’s CSI income from the single solidarity benefit, whereas Approach B does not. The net effect is that under the whole household approach, this multi-generation household is worse‑off by EUR 384.59 compared to the working-age household benefit unit approach (being entitled to EUR 255.98 of single solidarity benefit compared to 640.58).
Income for a workless household with one pensioner, a working-age couple, and two children, under proposed working-age single solidarity benefit with different household benefit unit definitions, 2024
Source: OECD calculations.
A comparison of the equivalised household income from the two approaches shows that the effective amount of support per person under the whole household approach is identical to the single adult rate of the single solidarity benefit (EUR 237.25), whereas income under the working-age household approach is EUR 113 higher (EUR 350.37). In both cases however, the equivalised income is significantly below the CSI income level for a single adult (EUR 550.67).
The variation in support between the two household benefit unit definitions will vary depending on the household. In all cases however, mixed-aged households will be worse‑off under the broader whole household approach.
Under both approaches, pensioners are, in effect, significantly worse‑off than they would be if living alone, although this does not account for housing costs. In practice, because pensioners still receive the full CSI payment, it will depend on the how the household as a whole decide to spend their income.
The whole household approach has stronger immediate work incentives because the amount of support is lower. However, the whole household approach may slightly weaken long-term work incentives. Specifically, because all pensioner income is included, an individual is incentivised to work less over their lifetime as this would increase their overall pension income and lead to lower overall benefits in the example household described above.
The whole household approach may also be seen as unfair. If a pensioner has a larger pension, say EUR 900 per month, then the household will not be entitled to any solidarity benefit in the above example.
Whole household and separate pensioner household definition for the extended solidarity benefit (Scenario D)
The above discussion on household benefit units focuses on the main reform in this paper, which looks at reforming most of the working-age solidarity benefits. The more complete reform (Scenario D) proposed in Section 2.1.4 looks at combining CSI as well. The choice of household benefit unit definition also has implications for this more complete reform as well, although as CSI is included in the reform these implications are slightly different.
This section again considers two household benefit units:
The whole household benefit unit approach (Approach A). This approach includes pensioners and their income when calculating the single solidarity benefit.
The separate working-age and pensioner household benefit unit approach (Approach B), which is the definition applied under the reform in Section 2.1.4, is an alternate approach. This approach separates pensioners and their income when calculating the single solidarity benefit.
Annex Table 2.A.2 again considers an example household with one pensioner, two workless working-age adults and two children. In this case, all income is generated by the solidarity benefit so there are no income deductions to consider. Overall, the separate working-age and pensioner household benefit unit (Approach B) is slightly more generous, providing EUR 1 191 of support compared to EUR 1 120 of support under the whole household approach (Approach B). This is because the separate household approach essentially has two adults entitled to the first adult rate which is slightly higher than the second adult rate.
Income for a workless household with one pensioner, a working-age couple, and two children, under proposed complete single solidarity benefit with different household benefit unit definitions, 2024.
Source: OECD calculations.
We can also note that Approach B provides the same support for mixed-aged households under this more complete reform as it is under the working-age solidarity benefit reform in the previous section. This is because they are in fact identical reforms as support for pensioner-only households remains unchanged. However, the whole household approach (Approach A) is more generous when compared to working-age reform as it essentially protects the higher pensioner benefit from being reduced.
When considering this more complete solidarity reform, these different household benefit unit approaches also have impacts beyond workless households. This is because under Approach A, earnings from any adults in the household will reduce all solidarity benefits (including those that pensioners are receiving). Under Approach B (and under the working-age solidarity reform), pensioner solidarity benefits are protected and not affected by earnings from working-age adults in the household.
This explains why the complete solidarity benefit reform leads to better targeted support at low-income households and reduces the overall cost of solidarity benefits to the government (as discussed above in Section 2.2).
The Euromod software (maintained by the European Commission) is used to simulate the effects of the benefit reforms outlined in this paper. It analyses the impact of reforms on household incomes, poverty, and benefit spending by government. This annex sets out more detail on the modelling approach.
All modelling is conducted to show results for the full 2024 year. Where policy rules have changed during the year, the latest policy rules are used in the baseline analysis.
The modelling necessarily relies upon household survey data which is expensive and time‑consuming to collect. The modelling uses the latest available survey data which was collected in 2022. Financial data in the survey is uprated in line with changes in earnings and prices to 2024 values.
The modelling results are calibrated to reflect administrative benefit data. The calibration works to match the total expenditure on solidarity benefits in the baseline scenario by reducing the take‑up of solidarity benefits in the model. The modelling only assumes changes to this take‑up rate when explicitly stated, otherwise the take‑up rate remains the same in all scenarios. Significant changes in benefit claims since 2023 will not be taken into account in the modelling and results.
The microsimulation modelling directly models the new single solidarity benefit and the largest existing solidarity benefits (Minimum Income Benefit, the existing Social Unemployment Benefit, Social Pension and Pensioner Social Assistance Benefit).
Benefits with small numbers of recipients (for example Social Pregnancy Risk Benefit) are not possible to model as the household survey sample size is too small to be able to provide reliable results or the survey lacks the information required to assess a household’s eligibility. Regardless, changes to these benefits will not have a significant impact across the population.
Disability benefits are not directly modelled in the microsimulation model, but this is unlikely to materially affect the results. Disability benefits present a particular modelling challenge as information on disability available in the household survey does not correspond to entitlement criteria for disability benefits. Moreover, the survey does not distinguish between receipt of different disability benefit income, making it impossible to accurately model the interaction of benefit changes across all solidarity benefits. Not modelling disability benefits directly is unlikely to significantly affect the results for two reasons. First, the reforms do not change the overall level of benefit support for households with only disabled persons (in part because disabled persons are unlikely to be working). Second, the number of household claiming disability benefits is included in the calibration of the results as discussed above.
A complete explanation of the modelling steps undertaken using the Euromod software are provided below:
Benefit amount values
Introduce new constants for the following benefit amounts:
Value for first adult in the household,
Value for each additional adult in the household,
Value for each child in the household,
Supplement value for unemployed person,
Supplement value for single pensioner,
Supplement value for a pensioner couple.
Income lists
Introduce new income lists for benefit withdrawal:
Create an income list for household earnings (employment and self-employment income).
Create an income list for other household income (pensions, other benefits and investment incomes).
Change unemployment assistance rules
Make changes to variable bunnc_pt (Unemployment assistance (subsidio social de desemprego)):
Remove the income test from eligibility criteria.
Adjust the calculation to only output a temporary variable which is used in new benefit modelling.
Turn off existing modelling
Disable the modelling of existing benefits:
Turn off policy buncm_pt (Unemployment benefit bonus).
Turn off policy poanc_pt (Social Pension (Pensão social de velhice)).
Turn off policy bsaoa_pt (Solidarity supplement for older persons (Complemento Solidário para Idosos – CSI)).
Turn off policy bsa00_pt (Social insertion income (Rendimento social de inserção ou mínimo garantido)).
New benefit modelling
Introduce a new benefit calculation for calculating the new social benefit:
Define a new tax unit to separate elderly (age greater than 66) and their partners and underage children from the rest of the household. This is achieved by making the household head condition be based on age.
Set up two calculations: the first calculation for elderly people using the elderly household benefit unit definition and the second calculation for households with working age people using the whole household approach. Households with working-age individuals and pensioners will have two calculations.
For each calculation, determine the total maximum benefit value depending on the benefit household composition in line with the benefit amount values above.
Then calculate the income deduction from the maximum benefit amount using income as defined by the income lists above. The reduction considers the income for the relevant household benefit unit for each calculation.
Take‑up calibration
Introduce a take‑up calibration calculation that randomly assigns benefit receipt:
Set up a random seed and a calculation to output a random result.
Set up a calculation to change the benefit amount to nil if the result for the above random calculation is larger than the external take‑up rate.
← 1. The Widow’s Social Pension is only available to the widow of a Social Pension recipient.
← 2. The main income entitlement criteria for the Social Pension are that an individual must not have income more than EUR 204 per month, or a pensioner couple must not have income more than EUR 306 per month. There are no asset criteria for the Social Pension. The pensioner minimum income benefit entitlement income criteria are that an individual must not have income of more than EUR 551, or a pensioner couple must not have income more than EUR 964 per month. While the pensioner minimum income benefit has higher income limits, it also has a wider definition of income. Specifically, a percentage of the value of any assets are counted as income, as are any private transfers of income received. As a result, a pensioner with significant assets, or who receives a large amount of income as a private transfer may be entitled to the Social Pension, but not the pensioner minimum income benefit. Source: Practical Guide 8 002 and Practical Guide 7 009, Social Security Institute.
← 3. Upon the request of DGSS.
← 4. The fixed asset limit threshold for existing solidarity benefits discourages households from saving beyond that limit if they have any concern that they might need to claim solidarity benefits in the future. The suggested approach of taking an imputed value of assets as income does not have the same disincentive. This incentive is particularly important when considering retirement income and benefits given the large share of the pensioner population in receipt of solidarity benefits, and the reasonably high probability of being entitled to support. Incentivising saving encourages households to financially support themselves in the future and therefore reduces future demands on benefit spending.
← 5. For example, a household with a primary residence worth EUR 200 000, and savings of EUR 26 000, would have relevant assets of EUR 10 000 over the EUR 16 000 allowance. 5% of EUR 10 000 is EUR 500, and so the household would have EUR 500 deducted from their monthly single solidarity benefit entitlement.
← 6. Data source: MTSSS.
← 7. Exceptionally, the working-age partner of a pensioner would also be included in the pensioner benefit household.
← 8. Annex A explains how including pensioner support in the single solidarity benefit affects larger more complex households with both working-age and pensioner adults in more detail.
← 9. In practice, very few households would be entitled to multiple supplements.
← 10. Or 120 days if unemployment is the result of a fixed-term contract expiring, termination after a trial period, or the adult is a victim of domestic violence. The conditions are identical to those of the current Social Unemployment Benefit.
← 11. RRP project for Digital transition of Social Security services (European Commission, 2023[9]).
← 12. Source: Eurostat (ilc_iw01), data for 2023.
← 13. Source: OECD analysis using Euromod.
← 14. Social benefit spending includes all means-tested benefits and all non-contributory, non-means-tested benefits. Solidarity benefits are a subset of social benefits.
← 15. Source: OECD analysis using Euromod.
← 16. This assumes the 14 minimum wage payments of EUR 870 are paid averaged over 12 months.
← 17. In Portugal, Child Benefit is paid at a lower rate to households with some – even if relatively low – income. In the majority of OECD countries means-testing of Child Benefit begins at a relatively high-income level. Indeed, if a household with two‑adults and two‑children in which one adult works full-time at the minimum wage were paid child benefit at the same rate as households out of work, the PTR would reduce from 96% to 73%.
← 18. Source: Eurostat, (spr_exp_type).