2. The evolution of minimum income schemes in Europe

Anti-poverty benefits – local and discretionary – were the first (and only) form of public intervention to mitigate social deprivation in most European societies. Dating back to the late nineteenth century, these policies combined a paternalistic approach typical of pre-industrial authoritarian societies with segregating and repressive elements. After all, societies then still mostly viewed social assistance as an instrument of social regulation, aimed at suppressing and preventing cases of deviance rather than fulfilling solidarity and individual support purposes. These benefits were thus occasional, residual, and discretional, involving severe stigmatisation of the beneficiaries. It was with the transition from this kind of poor relief and public assistance scheme to the introduction of the first public compulsory social insurance schemes that the modern welfare state emerged (Alber, Reference Alber, Flora and Heidenheimer1981).

During the Golden Age of welfare state expansion (1945–1975), although remaining in place in most countries, anti-poverty benefits lost most of their functional relevance in a context characterised by economic growth and Keynesian macroeconomic policies that aimed to guarantee full employment. At the same time, the origins of minimum income schemes date back precisely to this period, when last-resort income support schemes dropped the discretionary and ad hoc nature that had characterised them until then to resemble more and more fully fledged enforceable social rights (Ferrera, Reference Ferrera2005). Despite different institutional designs, those schemes shared the vocation to serve as a residual layer of social support underneath universal and/or insurance-based social protection arrangements, providing for the exceptional cases with unusual needs that in a full-employment economy fell through the net of more comprehensive programmes and remained poor (Clegg, Reference Clegg2013). The main functions of these schemes were thus income support and the prevention of extreme marginality. In 1948, the United Kingdom was the first country to introduce a scheme of income support intended to provide sufficient resources to meet the basic life course needs of the worst-off in society; it was later followed by Sweden (1956), Germany (1961), Denmark (1974), Belgium (1974), and Ireland (1975). Some countries opted instead for categorical schemes, directed to specific target groups, primarily the elderly poor (cfr. Madama, Reference Madama2015). Italy followed this route, introducing the social pension in 1969, along with France (1956) and Belgium (1969) and later Portugal (1980) and Spain (1988).

In the late 1980s, a new phase emerged, characterised by the diffusion of safety nets in all European countries and by the emergence of the activation paradigm. Following the introduction of the Revenu Minimum dInsertion in France, all the countries that had originally chosen categorical benefits finally introduced fully fledged MISs. This was the case in Luxembourg (1989), Spain at the subnational level (1989–1995), and Portugal (1997). Eastern and Central European countries, facing the collapse of the Soviet Union and the slow process of EU enlargement to the East, also gradually introduced minimum income schemes (Rat, Reference Rat, Cerami and Vanhuysse2009). This diffusion of means-tested benefits, aiming to guarantee a minimum amount of resources to those who have insufficient means of subsistence, was concluded only very recently in all EU countries, with the adoption of an MIS in February 2017 in Greece and the introduction of the Inclusion Income in Italy in 2018.

In this phase, in most countries anti-poverty programmes deeply transformed from the safety nets of the Golden Age, leading scholars to speak of a third generation of social assistance programmes (Kazepov, Reference Kazepov2010). As part of a broader shift of many European welfare states from transfer-oriented to active-service-oriented welfare (Hemerijk, Reference Hemerijk2012), in fact, the function and scope of MIS broadened to include social inclusion and labour market integration. Under the generally accepted assumptions that passive benefits generate disincentives to seek and accept work and that it is only possible to reinstitute full citizenship through labour market integration of the able-bodied (cfr. Weishaupt, Reference Weishaupt, Marx and Nelson2012), governments made access to minimum income benefits conditional on job search activities and on programmes that, as their ultimate goal, aimed for labour market (re-)inclusion (Lodemel and Trickey, Reference Ldemel and Trickey2001; Moreira, Reference Moreira2008). As a general rule, they strengthened the conditional ties and duties for beneficiaries, and social services and active labour market programmes became part of the minimum income package to break the vicious circle of inactivity, obsolete skills and competences, and demotivation.

It is possible to associate this apparent convergence towards selective universalism (Ferrera, Reference Ferrera and Gilbert2000) and active inclusion partially with the fact that, in the post-industrial scenario, the salience of social assistance benefits has increased. Major transformations in labour market structures and policies – specifically the flexibilization of employment, the spread of atypical jobs, and/or retrenchment in unemployment protection – combined with novel socio-cultural patterns – in particular mounting family instability and more fragile family–kin solidarity ties – have made social assistance a key social protection institution for ordinary European citizens (Marx, Reference Marx2007; Natili, Reference Natili2019; Rosanvallon, Reference Rosanvallon1995). At the same time, the institutional design of the very last tier of social protection regimes remains largely differentiated across Europe (Bahle et al., Reference Bahle, Hubl and Pfeifer2011; Frazer and Marlier, Reference Frazer and Marlier2016; Immervol, Reference Immervoll2010; Marx and Nelson, Reference Van Mechelen, Marchal, Marx and Nelson2012) and the country-specific configuration of MISs provides citizens with very different benefits and programmes to countervail poverty. The next paragraphs aim to provide information precisely regarding this dimension – that is, the different structures of MISs in Europe.

3. The dimension of variation of minimum income schemes in Europe

3.1. The institutional role of minimum income schemes

To understand the key features of minimum income schemes in a country, it is essential to know which principles of distribution dominate the overall welfare state context (Bahle et al., Reference Bahle, Hubl and Pfeifer2011) and, in particular, the main institutional features of the income maintenance system for the working-age population. A more inclusive and universal unemployment compensation system in fact reduces the relevance of MIS, which conversely becomes a central social policy institution when unemployment benefits are less inclusive and comprehensive.

The institutional configuration of unemployment benefits differs quite extensively across Europe, although signs of greater harmonisation have been apparent in the last two decades (for a review, see Clasen and Clegg, Reference Clasen and Clegg2011; Picot, Reference Picot2012). They are never formally universalistic but always based on a combination of contributory and means-tested access (Schmid and Reissert, Reference Schmid, Reissert and Schmid1996). At the risk of some oversimplification, here (Table 1), I adopt the basic, standard classification of unemployment benefits based on three pillars or levels (Vesan, Reference Vesan and Ferrera2012), each resulting from specific political struggles at different historical moments (Alber, Reference Alber, Flora and Heidenheimer1981). The first pillar consists of the standard unemployment insurance, financed by employers and employees contributions, so entitlement is conditional on past employment and contribution records. The benefit levels are earnings related, and the durations are limited. General revenues, conversely, finance unemployment assistance – the second pillar; entitlement is usually based only on being unemployed – in some cases, it is also necessary to have an expired possibility to access first-tier unemployment benefits – and always depends on a means test. The benefit amounts are generally a flat rate. These first two tiers are sometimes called dedicated unemployment benefits. Social assistance benefits – that is, minimum income schemes – constitute the third, residual, pillar.

Table 1. The role of minimum income schemes in income maintenance systems.

Source: Authors elaboration from Picot (Reference Picot2012) and Sacchi (Reference Sacchi2013).

The financing, governance, and institutional features of these different benefit tiers show great cross-national variation in Europe. Historically, the most typical unemployment compensation system envisaged the presence of all three pillars. In this type of system, MIS were to play a residual role, serving as a last-resort safety net underneath universal and/or insurance-based social protection arrangements, providing for the exceptional cases with unusual needs that in a full-employment economy fell through the net of more comprehensive programmes and remained poor (Clegg, Reference Clegg2013). This type of income maintenance system is still in place in many relevant European countries, such as Austria, the Czech Republic, Finland, France, Portugal, Spain, and Sweden. In these countries, MISs still play a residual anti-poverty role, albeit increasingly relevant due to the changing nature of the labour market (Marx, Reference Marx2007).

In other countries, MISs are the only existent income protection scheme for the working-age population beyond a (usually) very inclusive, generous, and (once) long-lasting unemployment insurance scheme. This is the case of Belgium, Denmark, Iceland, Luxembourg, and the Netherlands. In these countries, MIS still have mainly an anti-poverty function – even though their relevance depends on the degree of inclusiveness of the first-tier unemployment benefit. In the last decades, in fact, changes in the duration and in the eligibility condition of unemployment benefits have contributed to increasing the relevance of social assistance means-tested benefits in many of these countries (Clasen and Clegg, Reference Clasen and Clegg2011).

Finally, a different combination of unemployment benefits and social assistance schemes exists in Ireland, the United KingdomFootnote ³ , and, since the famous 2005 Hartz IV reform merging unemployment assistance and social assistance (Fleckenstein, Reference Fleckenstein2008), Germany. Here, below a strict – unable to protect all the unemployed and especially atypical workers – unemployment insurance scheme exists a comprehensive minimum income scheme for the able-bodied working-age population that is complemented by a series of categorical and fragmented social assistance benefits for special categories of people. It is fair to say that it is precisely in this third type of configuration that MISs assume the most important role in the income maintenance system, as they protect also many (long-term) unemployed due to the low inclusiveness of higher level unemployment benefits.

3.2. Generosity

In Europe, the amount of minimum income benefits generally coincides with the difference between a poverty threshold that a law or administrative decree establishes and the beneficiaries estimated income. However, the principle for determining the level of this threshold is not as straightforward as it is, for example for unemployment benefits. In Recommendation (C(2008) 5737) adopted in October 2008 on the active inclusion of people excluded from the labour market, the European Commission declared that adequate income support should be sufficient to lead a life that is compatible with human dignity. Beyond this general (and vague) principle, there is no agreement regarding the definition of an adequate standard, and every country sets a different system to define the minimum income level.

Frazer and Marlier (Reference Frazer and Marlier2016) distinguished between countries where there is a mechanism for establishing the level of payments that is based on a (set of) clearly specified indicator(s) and countries – the majority – where such mechanisms do not exist. Among the former, some countries define their minimum income benefits as a percentage of the social benefits or of the minimum wage. In Portugal, until 2012, the level of the benefit was equal to the social pension. In the Netherlands – as in most Spanish autonomous communities – conversely the amount of the Participatiewet corresponds to a percentage of the minimum wage: single individuals receive 50 of the minimum wage, single parents receive 70, and couples receive a transfer that equals 100 of the minimum wage.

However, in most countries, there is no clear mechanism to establish the benefit amount, and the minimum level is set arbitrarily (politically and/or administratively) every year to cover the basic costs deemed necessary to conduct a decent life, taking into consideration food, clothing, housing expenditures, and so on. Having established the basic threshold level for a single person, the benefit amount is parametrised according to different factors, among which the most common is the household composition. In many countries – especially northern and German-speaking continental countries – the scale of equivalence also takes childrens age into consideration. Moreover, in those cases, the benefit amount may increase significantly to cover particular individual and household needs and expenses. These are particularly generous in Germany, Finland, and Sweden, where it is possible to receive additional transfers to cover expenditures for rent, heating, electricity, health care and medicines prescribed by a physician, school expenses for children, and transportation costs from and to workplaces (Angelin et al., Reference Angelin, Johansson, Koch and Panican2013; Petzold, Reference Petzold2013). Unfortunately, those benefits are very difficult to calculate precisely, since they are often fixed at the local level, so they create within-country differences, they are modulated on specific households features, and they are usually of a discretionary nature (Marchal, Marx, & Van Mechelen, Reference Marchal, Marx and Van Mechelen2014). For these reasons, I do not consider them in my comparative description, despite being aware that this might lower the generosity of MISs in some countries, especially Germany and Sweden.

Table 2 shows the generosity of minimum income schemes in most European countries in 2015. Following common practice (see, for instance, Bahle et al., Reference Bahle, Hubl and Pfeifer2011; Marchal and Van Mechelen, Reference Marchal and Van Mechelen2017; Nelson, Reference Nelson2013), I assess the adequacy of income support through the benefit amount in relation to the relative poverty threshold – namely 60 of the median equivalized income among the adult population. This type of indicator indicates how much minimum income benefits move close households income to the poverty line.

Table 2. Generosity of minimum income benefits in Europe, 2015.

Source: Author's elaboration based on Frazer and Marlier (2016) and the Eurostat online database.

3.3. Coverage

The ability of a minimum income scheme to intercept all the people living in poverty depends on a broad range of issues. As the previous paragraphs underlined, in some countries, a fairly inclusive and protective unemployment benefits system has left a residual role to MIS; in others, reforms and retrenchment in upper-tier unemployment benefits have made social assistance benefits an important social protection institution for vulnerable middle-class citizens (Clegg, Reference Clegg2013). Furthermore, countries have different regulations concerning other types of non-contributory benefits – i.e. disability, child, in-work and/or housing benefits, etc. – that clearly have an impact on the overall number of beneficiaries of minimum income schemes, so that comparative analysis based on number of recipients should be considered cautiously.

However, the total number of beneficiaries also depends on eligibility conditions, which vary extensively across Europe, with important consequences for poor citizens possibilities to access social benefits (Figari et al., Reference Figari, Matsaganis and Sutherland2013; Granaglia and Bolzoni, Reference Granaglia and Bolzoni2016). The most common types of eligibility conditions relate to a lack of financial resources, not having assets above a certain limit, age (e.g. 18 or 25), nationality/citizenship, and/or residence. In many – but not all – countries, to access minimum income benefits, individuals need to have exhausted their rights to any other (social) benefit (cfr. Frazer and Marlier, Reference Frazer and Marlier2016). Beyond these formal requirements, peculiar administrative requirements might (or might not) create barriers that discourage would-be beneficiaries from advancing a request to access last-resort safety nets. Studies have reported transaction costs, such as difficulties in completing the application form and/or a procedure that individuals consider to be too complex and invasive, as reasons for the low take-up level of minimum income benefits (Eurofound, 2015). Other, have documented that sometimes exist administrative practices to discourage applications from specific groups of would-be beneficiaries, such as migrants (Natili, Reference Natili2016). Bureaucratic procedures thus affect significantly the total number of social assistance recipients (Eurofound, 2015; Matsaganis et al., Reference Matsaganis, Paulus and Sutherland2008).

Table 3 reveals the number of beneficiaries in 2015 of means-tested minimum income schemes in most European countries, collected by national statistical offices. These numbers provide useful information regarding the diffusion and relevance of minimum income benefits in a country. However, to obtain a more analytically useful dimension of coverage, in Table 4, I depict the number of beneficiaries as a percentage of the people aged less than 65 living in severe poverty, that is, with an income below 40 of the median average income, and not 60, as it is usually gauged. This type of outcome indicator estimates the ability of last-resort safety nets to intercept people in need, although it does not provide information regarding which factors precisely explain a particularly high or low level of coverage. As mentioned, several factors might in fact contribute to different degrees of inclusiveness of MISs, such as a very low income threshold, behavioural requirements that are too strict, the presence of administrative requirements that discourage potential applicants, a low take-up level, and so on.

Table 3. Beneficiaries of MIS programmes in Europe, 2015.

The figures refer to governments official estimates.

The figures refer to authors estimates based on available funds.

Source: Authors elaboration from the national statistical offices.

Table 4. Coverage of MISs in European countries, 2015.

Source: Authors elaboration from the national statistical offices and the Eurostat online database.

Accordingly, I am able to capture three different types of minimum income schemes in relation to their degree of inclusiveness. The first type protects only a small portion of the people in severe poverty – and for this reason I include it here in the group with a low degree of inclusiveness. The second type includes anti-poverty MISs that are able to intercept (most of) the population in severe poverty: this is the case of Denmark, the Netherlands, Estonia, and Austria. Finally, an inclusive MIS is able to overcome the barrier of severe poverty to protect people in a situation of economic vulnerability as well. It is interesting that this dimension does not entirely overlap with the institutional distinction above: although countries in which the MIS plays a central role in their income maintenance system (GermanyFootnote ⁴ and Ireland) have a very inclusive MIS – as expected – this is also the case in some countries in which the MIS has a last-resort safety net function, such as FranceFootnote ⁵ , Finland, and Sweden.

3.4. Expenditure

Finally, a standard dimension to analyse social programmes and minimum income schemes is the overall expenditure level (Bahle et al., Reference Bahle, Hubl and Pfeifer2011). In Table 5, I differentiate among European minimum income schemes considering their level of expenditure as a percentage of the overall welfare expenditure. Clearly, this type of indicator is strongly influenced by the role of MISs in the overall income maintenance system for the working-age population. However, it is interesting that these dimensions do not entirely overlap. In Italy, the absence of a second-pillar unemployment compensation system should increase the relevance of MISs, which for the moment in the prevision of the government is still extremely residual. Conversely, the expenditure levels are (comparatively) quite high in France and Finland, considering that, in those countries, they still serve as residual safety nets below fairly comprehensive unemployment benefits systems. Less surprising is the high level of expenditure for MISs in Ireland and Germany, where – as already mentioned – means-tested minimum income schemes have a central role in the income maintenance system.

Table 5. Expenditure on MISs in European countries, 2015.

The figures refer to government official estimates regarding minimum income expenditure.

Expenditure for the RSA Activit not included in the analysis.

Source: Authors elaboration from the national statistical offices and the Eurostat online database.

4. Active inclusion services from a comparative perspective

Social assistance and minimum income benefits have deeply transformed in the last decades, so, examining aspects such as risk coverage and eligibility, the benefits structure, financing mechanisms, and organisational arrangements in place is not sufficient to capture their main properties. Nowadays, in fact, minimum income schemes in Europe are complex social policy benefits that combine income support with a block of services and programmes designed to help recipients to return to the labour market. In addition, since the late 1980s – and increasingly afterwards – access to MISs has been conditional on recipients willingness to activate themselves in the labour market (Lodemel and Trickey, Reference Ldemel and Trickey2001; Moreira, Reference Moreira2008).

Schemes can – and actually do – realise this activating function very differently, as there is little agreement regarding the precise meaning of activation. Broadly defined, it is possible to conceive activation as a policy of combining negative and positive incentives with the ultimate purpose of helping minimum income recipients to become self-sufficient through paid employment (cfr. Moreira, Reference Moreira2008). Thus, it involves a number of policy instruments, such as legislation, financial incentives, and social or labour market services (Bonoli, Reference Bonoli, Clasen and Clegg2011). To make sense of the broad constellation of different measures and instruments, different authors have created typologies of active inclusion services (Barbier and Ludwig-Mayerhofer, Reference Barbier and Ludwig-Mayerhofer2004; Bonoli, Reference Bonoli, Clasen and Clegg2011; Lodemel and Moreira, Reference Lodemel and Moreira2014; Marchal and Van Mechelen, Reference Marchal and Van Mechelen2017; Moreira, Reference Moreira2008). Despite relevant differences, these authors have in common an inclination to distinguish between two different approaches to activation (Marchal and Van Mechelen, Reference Marchal and Van Mechelen2017). The first one is narrower and tends to focus exclusively on incentivizing behaviours that increase labour market integration. It includes measures that offer both financial and non-financial stimuli intended to lower the reservation wage, such as low benefit levels, strict availability requirements, or time limits. The second approach tends to focus on the personal development of individuals through social investment-oriented measures. Theoretically, it aims to favour the improvement of individual capacities rather than the inducement of behavioural change (Marchal and Van Mechelen, Reference Marchal and Van Mechelen2017). Building on these works – in particular those of Moreira (Reference Moreira2008) and Weishaupt (Reference Weishaupt, Marx and Nelson2012) – in the following I introduce a new typology to distinguish between different worlds of activation services, by looking at the degree of conditionality and at the types of programmes associated with MISs.

As to the first dimension, all European countries condition access to MIS on beneficiaries willingness to activate themselves. The basic idea is that, to bring more people into employment, it is necessary to impose stringent behavioural rules on MIS beneficiaries. Good indicators of this type of approach are the presence of more stringent assessments regarding job availability, greater recourse to sanctions, and a broader definition of a suitable job offer. An example of strong conditionality requirements is Germany, where, to avoid losing the entitlements, beneficiaries: a) have to perform all medical and/or psychological examinations agreed with the assigned personal assistant; b) during working days, have to be traceable by their personal assistant to the address that they provided and be available if requested to attend the job centre; c) have to respect all the tasks included in the integration agreement and in particular take up all measures for labour market integration and never cause its break-up; d) have to accept reasonable employment; and e) have to behave appropriately to avoid preventing the conclusion of an employment contract. In the case of a breach, the scheme will cut the benefits in terms of the standard rate by 30. The second breach will cause a cut of 60. After the third breach, the recipient will lose the benefits in terms of ALG II totally (Petzold, Reference Petzold2013). The sanctions are even stronger for beneficiaries under the age of 25 years.

Many European countries have introduced such stringent conditionality requirements – especially in the north and in German-speaking continental countries. In Table 6, I report the results of the comparative research that Frazer and Marlier (Reference Frazer and Marlier2016) coordinated, which distinguished between three levels of conditionality associated with MISs: the first one is very strict, the second is limited, whereas the third is characterised by the complete absence of behavioural requirements. It is noticeable that in Europe, in 2015, no countries attached no strings to minimum income benefits, although experimentation in this direction has been launched in Finland and in some cities in the Netherlands.

Table 6. Conditionality of MISs in Europe.

Source: Frazer and Marlier, Reference Frazer and Marlier2016.

The second dimension focuses on the type of programmes associated with MIS, in order to assess whether these are exclusively labour-market related programmes or rather MIS engage beneficiaries in a multi-level and multi-dimensional strategy that intends to reverse their situation of need. In the latter sense, the search for a job has the same priority as recovering from health problems, finding suitable accommodation, and so on (Guibentif and Bouget, Reference Guibentif and Bouget1997, pp. 14–18).

In this particular dimension, particularly interesting is the research that Marchal and Van Mechelen (Reference Marchal and Van Mechelen2017) carried out, which built an indicator to distinguish between different active inclusion instruments available to minimum income beneficiaries. In Table 7, I report some of the results of their research, in particular on two dimensions, which they called Enabling and Service, to capture whether programmes associated with social assistance benefits have only a work-related scope or instead a wider scope. The former (service) assesses whether the beneficiaries of MISs have real access to quality social services through an index that takes into consideration the combination of the range of services available to MIS recipients and their burden of childcare costs. The latter (enabling) instead evaluates whether beneficiaries have access (also) to services that should foster their human capital development. The idea is that the better countries perform in these index, the more their active inclusion strategy have a broader scope than aimed at simply kick-starting beneficiaries back to work.

Table 7. Activation profile of minimum income schemes.

Source: Authors elaboration from Marchal and Van Mechelen (Reference Marchal and Van Mechelen2017).

Combining features in terms of sanctions and conditionality with the types of programmes associated with MISs allows me to provide a basic descriptive table regarding active inclusion regimes in Europe (Table 8). What emerges is a complex picture, as four types of active inclusion profile become apparent. In Bulgaria and Latvia, limited conditionality and the presence of few services associated with MISs seem to imply that the social assistance programmes are still mainly Passive. A very strictly work-oriented activation model combined with a strong conditionality is typical of a Workfare type. This type of active inclusion regime has a strong focus on work requirements, sanctions, time limits and exclusively labour market related services (Lodemel and Trickey, Reference Ldemel and Trickey2001), and it is typical of Estonia, the United Kingdom, the Netherlands, Portugal, and Italy. The presence of very strict conditionality combined with services that aim to empower recipients is typical of Luxembourg and Germany, which seem to have adopted a Paternalistic type of activation profile, as the individual agency of beneficiaries is not fully respected but inclusion is not conceived exclusively in terms of labour market integration. Finally, the availability of a wide ranges of services combined with limited conditionality is typical of an Enabling regime. The countries in the sample that seems to have taken more steps towards an enabling activation regime are Austria, Finland, Belgium, Denmark, and Sweden.

Table 8. Inclusion regimes of MISs in Europe.

5. Minimum income schemes in Europe: a typology

In Europe, minimum income schemes (MISs) differ not exclusively in terms of generosity, duration, eligibility requirements, conditionality, and overall expenditure but also in terms of territorial configuration (Table 9) (Hlsch and Kraus, Reference Hlsch and Kraus2004; Kazepov and Barberis, Reference Kazepov, Barberis, Marx and Nelson2012).

Table 9. Territorial configuration of MISs in Europe.

In some cases, countries finance and regulate MISs at the national level, whereas sub-national governments role is limited to the implementation of active inclusion services. This is what I call here a centralised model, and it is typical of Anglo-Saxon (Ireland and United Kingdom) countries and France.

In the Nordic countries – but also in Germany and the Netherlands – the central state establishes the minimum standards that it must guarantee throughout the country, but subnational governments are particularly involved in regulating and organising the provision of minimum income schemes. This implies that local governments cannot legally deny benefits and services to those fulfilling eligibility criteria (Minas and Overbye, Reference Minas, Overbye and Kazepov2010) but that, at the same time, they have the responsibility to manage and (at least partially) finance the scheme and enjoy a high degree of autonomy in implementing the benefits (Kazepov and Barberis, Reference Kazepov, Barberis, Marx and Nelson2012). As for territorial differentiation, in this model, there are differences in the provision of benefits and services beyond the minimum standard. This is why I refer to the model as an integrated decentralised model. Finally, in some countries, sub-national governments have assumed a crucial role in the anti-poverty field in the absence of a national legislative framework setting minimum standards, guaranteeing coordination between different instruments, substituting regional governments in the case of non-compliance, and facilitating the assurance of the diffusion of best practices. This is the case in Austria (apart from a break between 2010 and 2016), Switzerland, Spain, and, until very recently, Italy. As the Italian and the Spanish case outline, the uncoordinated decentralised model allows the emergence of strong territorial inequalities (Kazepov and Barberis, Reference Kazepov, Barberis, Marx and Nelson2012; Natili, Reference Natili2016).

This brief reconstruction has confirmed that minimum income schemes are complex social policy measures, combining monetary benefits and a broad number of services, which vary extensively across Europe and can be studied from different angles. They have different roles in the income maintenance system, so it is necessary to know which social programmes exist above before trying to compare their effectiveness. They should countervail poverty and material deprivation, so it is necessary to assess their adequacy – in terms of their ability to guarantee a decent standard of living to poor individuals – and to verify the presence of formal and/or informal conditions that might impede the access of a relevant part of the population in need. Furthermore, nowadays, one of their main goals is to help beneficiaries to participate fully in the society and – when possible – to help in finding an appropriate job. Schemes, however, realise this differently, and, although sometimes individual agency is not respected and the only type of services that they provide to beneficiaries are labour market-related, in other cases they adopt a multidimensional approach that should help individuals to solve their issues in different domains of their life. Finally, the definition of who is responsible for minimum income benefits also differs quite extensively across Europe. This is not only important in terms of territorial equality – as a more decentralised system may create less egalitarian systems, in which discretionary powers become particularly manifest (Nelson, Reference Nelson2013) – but also possibly conditions their ability to achieve the goal of relieving poverty (Hlsch and Kraus, Reference Hlsch and Kraus2004).

All these dimensions have important distributive and normative consequences. Moreover, combining these different dimensions, country-specific configurations of minimum income schemes emerge, which are very revealing about the normative foundations behind them and the notion of solidarity they pursue, as addressing in a non-neutral way key concerns about who are the poor, who deserve to be assisted, in what forms and by whom (cfr. Madama, Reference Madama2015). Here – after having underlined in the previous paragraph also differences in the governance of MIS – we construct a typology based on two particular dimensions: their ability to protect poor individual and the type of active inclusion services attached to MIS. The ability to protect from poverty depend on many different things, but in particular on their generosity and on their availability in case of need. While the former dimension is rather straightforward, benefit availability as we have seen is more complex and depend on different factors, such as: availability of other type of social policy benefits – i.e. the role of MIS in the institutional configuration of income maintenance benefits – type of eligibility requirements, and administrative features. Here, we consider benefit available when the degree of inclusiveness is high, or when it is medium but the role of MIS in the income maintenance system is residual.

Combining these elements, I construct my typology of minimum income benefits (Table 10). In several countries, minimum income schemes are Inadequate, in the sense that they are not able to provide a minimum degree of protection to poor individuals while their active inclusion services seem weakly developed. This is the case in Bulgaria, Croatia, Czech Republic, Latvia, and Italy (as for 2018).

Table 10. Minimum Income Schemes in Europe: a typology.

When a fairly low level of protection is combined with a very strong workfare approach, I call MIS Sanctionatory. In those cases, MIS are mainly instruments to make poor people change their behaviour and participate to the labour market. This is the situation in Spain, Portugal, United Kingdom, Estonia and Lithuania, where – to be fair – MIS is slightly more protective that in the Inadequate model. The Protective model usually combine a fair level of generosity and inclusiveness with strong activation requirement, usually limited to job searching activities. Poor individuals are thus protected from the main vagaries of life, but there is less respect of their human agency and a general tendency to give greater importance to labour markets needs over poor individual choices and desires. Finally, in the Enabling model, income support is generous, people in poverty have the possibility to access to those benefits, and they have access to a broad range of active inclusion services – which are not exclusively aimed to make people return to the labour market no matter in what conditions. This model is typical of Northern countries, Belgium and France.