Immigration in a Mediterranean Welfare State - Dipartimento di ...

Journal of Comparative Policy Analysis, Vol. 6, No. 2, 111 – 129, August 2004

Immigration in a Mediterranean Welfare State: The Italian Experience in Comparative Perspective GIUSEPPE SCIORTINO Dipartimento di sociologia e ricerca sociale, Universita` degli studi di Trento, 38100 Trento, Italy

ABSTRACT There is a large body of literature that explores the use of welfare programs by immigrants. It is often argued that welfare programs function as a magnet for flows of unwanted migration. Very little research, however, has been carried out regarding the other side of the coin: the ways in which international migration provides crucial resources for managing the relationships between households, labor markets and the state. Welfare regimes interact often with migration regimes. Using Italy as an empirical case, the paper argues that the Italian migratory system takes place within the context of a structural tension between the welfare system – largely designed according to a male breadwinner model – and the social consequences of women’s rising labor participation rate. Such tension triggers specific migratory systems that supply foreign household workers participating in the labor market with statuses that escape the logic of Baumol’s cost disease. It is argued that the same strain, and likely very similar solutions, can be found in other western European welfare regimes of a conservative type.

Migration and Welfare Immigration is currently a controversial issue in most Western European societies. European governments have tried to curb immigration for more than 30 years, since the recruitment bans of the early 1970s. Since then, the assumption that further flows of unskilled foreign labor are both politically redundant and economically unnecessary has been hegemonic in the European policy networks. Over the last decade, there have been great political efforts to restrain, within the constraints of a liberal polity, the arrival of relatives of previous migrants, asylum seekers and refugees. Such policies have been somewhat successful. Especially if the extraordinary geopolitical disturbances that have marked the last decade are taken into account, the volume of current inflows is far from dramatic (OECD 2003). Still, all Western European countries have to deal with considerable levels of unwanted migration. A variety of channels of entry are still used by migrants with the necessary connections, while sizeable flows of visa overstayers and clandestines Giuseppe Sciortino is a sociologist working at the Universita’ di Trento. His main research interests are international migration, ethnic relation and social theory. Correspondence Address: Giuseppe Sciortino, Dipartimento di sociologia e ricerca sociale, Universita` degli studi di Trento, Via Verdi, 26, 38100 Trento, Italy. Email: [email protected] ISSN 1387-6988 Print/1572-5448 Online/04/020111-18 # 2004 Taylor & Francis Ltd DOI: 10.1080/1387698042000273442

112

G. Sciortino

are recorded in every European territory. The irregular employment of foreign workers is moreover recorded to a growing degree in many economic sectors (Tapinos 2000). High levels of unemployment among resident foreigners do not deter the arrival of new migrants, often insisting upon the same urban areas of settlement. Having denied the existence of a structural demand for unskilled foreign labor, such migratory dynamics may appear puzzling. Why should migrants make a considerable investment and face considerable difficulties (up to, increasingly, risking their lives) to enter a future of marginality and poverty? A popular answer is that such flows are pulled more by the welfare system than by the labor market. According to this view, generous welfare systems act as a magnet for unwanted migratory flows. As the human capital of current migrants is lower than necessary to successfully compete on the labor market, any new intake is likely to become soon a further burden on the welfare state (Levine 1999, Roodenburg et al. 2003). The issue of welfare dependency has also become a crucial dimension of the European attitude towards migration. According to the year 2000 Euro-barometer data, 52 per cent of European citizens think that migrants abuse the welfare system (Thalhammer et al. 2000). Many researchers have tried to assess the validity of the welfare magnet thesis. Research on welfare dependency is not easy, as data is scarce and the analyses have to take into account a variety of national peculiarities, both in the migratory history and in the configuration of the welfare system (Bird et al. 1999). Still, there is a remarkable convergence in the findings. A study of the United Kingdom has found that the foreign-born population’s contribution to government revenues is 10 per cent higher than they receive in government expenditure (Glover et al. 2001). Similar results have also been found for Germany, where immigrants appear to have contributed more to the economy than they have received in transfer benefits (Spencer 1994). German research has also shown that, once controlled for significant variables, there are no significant differences in welfare dependency between natives and foreign households (Riphahn 1998). Studies in Sweden and Denmark have traced the causes of high welfare dependency on migration policy itself, as those countries authorize entries nearly always only as refugees and asylum seekers, whose access to the labor market is typically restricted (Hansen and Lofstrom 1999). A recent comparative study has found that migration may place – although in a typically moderate way – pressure on some kind of welfare systems, while failing to do so on others. The same review tends to confirm that in Western European countries, migrants tend to assimilate out of welfare, as welfare dependency decreases with the duration of stay (Brucker et al. 2001). These findings are surely important from the political point of view. They also contribute to a more realistic background for policy elaboration. It may be observed, however, that they suffer also from a serious weakness in analytical terms: they take the policy framework for granted, unduly restricting the study of the relationships between the migratory system and welfare regimes to the issue of welfare dependency. Migrants, in other words, are seen only as consumers of welfare provisions or as taxpayers. In this article, it will be argued, on the contrary, that migration plays a much wider role in the functioning of European welfare regimes. It will be argued that the configuration of many Western European welfare regimes is actually a main structural factor behind the enduring demand for unskilled foreign

Immigration in a Mediterranean Welfare State

113

labor. And it is surprising that very little research has been carried out studying migration as a main mechanism in the functioning of welfare regimes. In the following pages, the argument will be developed in three stages, using Italy as the main case study. First, the Italian welfare system will be compared with those of the other European countries. The comparative analysis reveals some structural similarities both with the welfare regimes of Mediterranean countries and with some European continental ones. Consequently, the results identified in the Italian household may have a more generalized meaning. In the second section, it will be shown how the functioning of this kind of welfare regime produces a structural demand for labor in the personal and household services sectors. The demand is also bound to increase under the pressure of aging populations and rising rates of activity for women. It will be shown how this demand for household services is usually matched by sustained flows of foreign labor. The final section will deal with the consequences of these flows of household workers for the immigration policies of Western European states. It will be argued that the functioning of welfare regimes push states towards two options: either sizeable flows of circular irregular migration are tolerated through a liberal visa regime operating with contiguous sending countries, or a set of restrictive filtering devices has to be put in place. The Italian Welfare State: A Comparative View The comparative study of welfare states is now a major academic industry. The search for an adequate framework for the analysis of welfare states (and an explanation of their evolution) has informed some of the most significant social science studies of the previous two decades (Castles 1993, Esping-Andersen 1990). Research has identified significant differences in the structure of European welfare states. It has also triggered a lively debate on the ‘‘types’’ of existing welfare arrangements and on the items that should be selected to classify them. If we look at welfare as strictly defined (as a set of entitlements and provision provided by the state or guaranteed by it), it is possible to identify some features of the Italian system that stand out in international comparison (see Table 1). The Italian social expenditure as a percentage of gross domestic product (GDP) is lower than the EU average.1 The main difference, however, is in the structure of such expenditure: Italy has a much larger expenditure for pensions (both old age and survivor). A second important feature is the very limited use of means-tested benefits. The percentage of benefits that are payable only after means testing in Italy is less than half the European average. A third difference refers to the structure of salaries produced by the interaction of labor laws and collective bargaining: a large section of the employed labor force are sheltered from most employment-related risks and are paid according to family wage levels, accompanied by a lack of income support for those who have never worked and the long-term unemployed. The other European Mediterranean states have similar traits in their labor law and welfare and social programs. It has consequently been claimed that there is a ‘‘Mediterranean’’ cluster of welfare states with a specific and distinctive identity (Ferrera 1996, Leibfried 1991). Such classification, while some important features are shared by most Southern European welfare states, has the disadvantage of being

114

G. Sciortino

Table 1. Public social expenditure as a percent of GDP (1998) Type of public social expenditure

Italy

Germany

France

Sweden

UK

Old age cash benefits Disability cash benefits Occupational injury and disease Sickness benefits Services for the elderly and disabled Survivor benefits Family cash benefits Family services Active labor market services Unemployment Health Housing benefits Other contingencies All Social Expenditure

12.84 0.99 0.45 0.71 0.17

10.46 1.05 0.34 0.32 0.75

10.59 0.87 0.24 0.51 0.66

7.46 2.10 0.32 1.13 3.71

9.77 2.64 0.32 0.14 0.81

2.60 0.58 0.30 0.67 0.71 5.51 0.01 0.00 25.54

0.49 1.93 0.80 1.26 1.32 7.80 0.18 0.61 27.31

1.59 1.46 1.23 1.30 1.32 7.27 0.92 0.40 28.36

0.69 1.63 1.68 1.96 1.93 6.64 0.81 0.93 30.99

1.01 1.73 0.49 0.31 0.32 5.62 1.61 0.21 24.98

Source: Own elaboration on data from OECD (2001a).

focused mainly on similarities among welfare programs. Such differences, moreover, are explained nearly always only in reference to the structure of national labor markets.2 The evidence for such clustering, furthermore, is far from definitive (Esping-Andersen 1999). A conceptual alternative is to classify welfare states according to the form of their welfare regime, focusing on the nexus of relationships among states, households and labor markets. Esping-Andersen (1990, 1999) has proposed an influential classification of these regimes in three clusters (liberal, social-democratic or conservative). His work confirms the important similarities among Mediterranean welfare systems, but it shows them as variations of a much larger cluster, including Germany, Austria and – with significant differences in the familism score – France and Belgium. All these countries, as a matter of fact, have the following features: .

.

. .

families are entrusted with the ultimate responsibility for the welfare of their members (including legal responsibilities for their adult offspring and for their elderly members) as well as being seen as the best potential caregivers; the approach to employment management is passive, but it is matched by strong labor market regulations targeted to protect already employed (usually adult male) householders and to guarantee incomes sufficient to maintain a family; welfare programs are based more on monetary transfers to households rather than the provision of services; provisions are arranged mainly through compulsory social insurance programs (with a widespread presence of status segmentation and corporatist traits).

Esping-Andersen’s approach has the theoretical advantage of seeing welfare structures as embedded in a matrix of structural relationships among households, the state and the economy. It is precisely within this framework that the


115

relationships between welfare structures and migratory processes may be investigated in full. The Italian welfare regime is clearly characterized by the key role mandated to families for most welfare matters. Families’ responsibilities for their members are not only legally mandated (both in reference to parents and progeny) but most of the administrative procedures are actually centered on the expectation that this will happen. Families are also the main buffer institution. It is within families that the resources available to labor market insiders (the employed, the self-employed, the retired) are shared with the outsiders and the semi-insiders: the pensions of the aged father may finance unemployed adult children or sustain the study of nonworking nephews, the tenure of the husband may shelter from many risks the fragmented or shadow career of his wife, the level of protections guaranteed to survivors makes the lack of retirement benefits in the shadow economy less frightening. The centrality of the male breadwinner model in the structure of employment protection is derived largely from a vision of the centrality of the family and of the wider responsibilities the primary breadwinner assumes ‘‘on behalf’’ of his household. Such a relationship between the welfare system and households may be seen clearly in the channels used by the welfare state to provide its services. In general, conservative welfare regimes are centered on compulsory social insurance and are therefore service-lean and transfer-centered (Eurostat 2000). Not surprisingly, Italy has a larger percentage of non-means -tested cash provisions (75% of the social benefits delivered, against an EU average of 63%) and a corresponding very low percentage of social benefits delivered in kind (23% against a European average of 31%). The few areas (such as housing and family services) where there are a larger proportion of provisions in kind are also the areas where Italian public expenditure is more limited and residual. Italian households have a fairly high likelihood of being recipients of some kind of public monetary transfer and they enjoy strong labor market protection for at least one member. At the same time – with the exception of health care – transfer and protections are embedded in a context where few services are available on a generalized basis. Italian households must, indeed, provide a very large share of personal services to their members (see Table 2). Within such a regime, households actually operate as ‘‘general contractors’’, that is, as integrators of a large share of personal services, partly self-produced, partly acquired through public bodies and partly bought on the market. The functioning of the welfare regime is consequently contingent upon the smooth combination of these inputs. Welfare Regimes and the Demand for Immigrant Labor The above description has stressed how the Italian welfare regime – as the other conservative-type ones – works in transferring, directly and indirectly, money and guarantees to the (usually male) head of the household, who will subsequently select the most appropriate channel (a kind of make or buy alternative) for the provision of essentials to household members and to the household as a whole. Such a solution has some advantages, as the limited number of jobless households and the limited numbers of people in institutions demonstrate.3 But such a solution also, has its own

116

G. Sciortino

Table 2. Children in formal day care and elderly in institutions Country Austria Belgium Denmark Finland France Germany Ireland Italy Netherlands Norway Portugal Spain Sweden United Kingdom

% of children (aged 0–3) in daycare institutions

% aged 65 or more in institutions

4 30 64 22 29 8 38 6 6 40 3 to 5 5 48 34

4.9 6.4 7 5.3 to 7.6 6.5 6.8 5 3.9 8.8 6.6 – 2.9 8.7 5.1

Source: OECD (2001b).

problems. The main problem is that purchasing personal and household services is not easy. As early as the 1960s, it was observed that productivity gains in service industries lag behind those in manufacturing: most personal services cannot raise productivity much (Baumol 1967, Baumol et al. 1985). This has a very serious implication for their development, known in the literature as Baumol cost disease: the provision of personal services oscillates between the risk of a sharp reduction in their supply (if the wages in the sector follow its relative productivity) and the risk of a sharp reduction in the demand for them (if their wages follow the trend established by the most productive sectors and services, labor costs will price personal services out of the market). The problem appears even bigger in welfare states centered on a compulsory social insurance program. Here, the cost of services for a household must take into account not only the salary of the service-provider, but also a sizeable amount of statutory contribution and employment-related benefits. This is precisely the case in Western Europe, including Italy, where the relative cost of labor-intensive services is high. In the social-democratic type of welfare regimes, the service productivity differentials are absorbed by the state through the direct provision of the services. Where it is not, as in the conservative type, such productivity differentials will operate as a strong incentive for households to self-produce them. As such selfproduction is mostly considered women’s work, the implication is a structurally low female participation rate in the labor market. Another, more indirect, consequence is delayed and low fertility, as the opportunity cost of having children becomes higher. Both are very well-known features of the Italian context.4 Self-production, however, is an increasingly difficult strategy. Albeit slowly, the number of dual career households increases. Lower rates of fertility imply smaller households. The aging of the population increases the number of households where self-production is increasingly difficult. Rising living standards raise social


117

expectations. As a result, there is in Italy, as in the other Mediterranean countries, a large demand for household services. As most of such activity is undocumented and underreported, official statistics are largely unreliable. Such shortcomings notwithstanding, the comparative data summarized in Table 3 shows that the market for household services is definitively larger in the Mediterranean countries. Interestingly, however, the same holds true for all the transfer-based welfare states. In a closed system, the demand for personal services would quickly encounter Baumol’s cost disease. Wages would outpace productivity, thereby pricing the services out of the market. Such an outcome, in turn, would severely strain the relationship between the male breadwinner model, embedded in the welfare regime, and the reality of increasing female participation in the labor market. Such tension may be managed only if a way is found to maintain a comparatively cheap and flexible provision of household services. This is precisely where immigration comes in. Historically, the structural demand for personal services has often been managed through some kind of migratory system. Traditionally, the system has involved rural-urban migration and, subsequently, migration from the south and the north-east to the more developed urban areas of the industrial triangle and the bureaucratic conurbation of Rome. For most of the twentieth century generic domestic service – as well as a variety of more specialized jobs like cook, laundry-maker and nanny – was the Italian female equivalent of the assembly line.5 Contracts ranged from a full-time, in-house servant (with as little as half a day free time in a week) to hourly paid specialized work, sometimes done for a pool of different employers. Domestic work was considered a suitable occupation for women (for its symbolic links with family values), whose honorable reputation during migration was protected precisely by their being ‘‘entrusted’’ to the middle-class households they were working in.6 It was, moreover, one of the very few job opportunities open to Italian women, thus guaranteeing a steady supply of potential workers.

Table 3. Employment in selected service sectors as a percentage of working age population (1997) Country Sweden Denmark Belgium Netherlands Greece United Kingdom Germany Ireland Italy Spain France Portugal EU-15

Health and social work Personal and household 13.6 13.0 6.2 9.5 2.5 7.8 5.7 3.0 3.0 2.7 6.3 3.1 5.7

Source: European Commission (1998).

0.5 0.8 0.9 1.0 1.2 1.3 1.4 1.7 1.7 1.8 2.0 3.2 1.5

Total social services 26.0 27.3 19.5 22.3 12.5 21.4 17.8 15.2 13.5 11.8 20.4 16.9 21.4

118

G. Sciortino

The mid-1970s marked the beginning of a substantial shift. From one side, the demand for domestic labor was strengthened by the growing number of middle-class women entering the labor force as well as from rising standards in the quality of life. If servants had been traditionally a key element of middle-class status, their presence was now also functional to allow middle-class labor market participation without challenging the traditional division of labor within the household. At the same time, however, the pool of domestic workers shrank as women had new opportunities for unskilled work opening up in manufacturing and public services. Simultaneously, stronger working-class households started to withdraw their female members from the unskilled household labor market. Such changes shifted the balance of bargaining power: the remaining domestic workers opted more and more for hourly paid jobs rather than for the traditional live-in variety. The price for these services increased, while the supply decreased. Such a change, accompanied by the introduction of a national contract in 1969, contributed to make Italian household workers less cheap and less flexible as resources for households’ needs. The end of the 1960s and the 1970s witnessed several attempts to reform the Italian welfare system in a more social-democratic direction, including the expansion of public social services directly provided by the state. Such attempts, however, have largely failed (Ferrera 1984). As the welfare state did not absorb the productivity differentials, the setting was characterized by both a sharp increase in the demand for personal services and a sharp decrease in their supply. The result was a powerful pull factor for the recruitment of foreign domestic workers. Since the early 1960s, many women – mostly from the former Italian colonies in the horn of Africa and from the Catholic strongholds of the Philippines and Cape Verde and from Latin America – migrated to Italy to work as live-in maids (Andall 1998). Their presence was openly recognized in the Italian press, much earlier than many other immigration flows starting at the time (Colombo and Sciortino 2004). In the first half of the 1970s, the number of resident foreigners increased by 26 per cent whereas the number of permits for domestic workers nearly doubled. Such active recruitment – managed mostly by the Catholic Church through its mission network – is actually one of the structural roots of the current Italian migratory system, an early sign of the transition of Italy from sending to receiving country. 7 Once started, the flow of domestic workers reached quickly a self-perpetuating stage. Workers already active in the local household service market acted as triggers for the establishment of further migratory chains, using their own reputation as reliable workers as insurance for the reliability of the new workers. Not by chance, women were the majority of the foreign residents in the genesis years between 1970 and 1980; domestic workers’ countries of origin played (the significant undercounting produced by endemic irregularity notwithstanding) a significant role in the structuring of the non-European flows (see Table 4). The increase in foreign domestic work was both demand-induced and policyconstructed. In 1972, apparently with the intention to protect native domestic workers, the Ministry of Labor restricted foreign domestic work to live-in contracts, thus denying foreign women the more appealing option of hourly paid work. The same administrative order, moreover, locked in employment and residence. Foreign workers were put in a very vulnerable position: a change of employer would have entailed returning to the sending country for at least three years (Andall 1998). Such


119

Table 4. Foreign immigrants in Italy – selected nationalities (1970–1990) Country

1970

1975

1980

1985

1990

EU-15 43.8 44.5 39.7 37.7 20.9 Albania 0.1 0.1 0.2 0.2 0.2 Former – Yugoslavia 4.4 4.0 4.1 3.3 3.8 Poland 1.0 0.9 1.1 1.9 2.2 Egypt 0.6 0.6 1.1 1.6 2.5 Morocco 0.1 0.2 0.3 0.6 10.0 Tunisia 0.2 0.3 0.5 1.0 5.3 Cape Verde – – 0.8 0.8 0.6 Nigeria 0.1 0.1 0.6 0.9 0.9 Senegal 0.0 0.0 0.0 0.1 3.2 Ethiopia 0.3 1.3 1.7 1.7 1.5 Mauritius 0.0 0.0 0.1 0.1 0.7 Somalia 0.3 0.4 0.4 0.4 1.2 People’s Republic of China 0.1 0.1 0.2 0.4 2.4 Philippines 0.2 0.4 1.4 1.8 4.4 Bangladesh 0.0 0.0 0.0 0.0 0.6 India 0.7 0.9 1.1 1.3 1.4 Iran 1.2 1.2 3.1 3.1 1.9 Sri Lanka 0.1 0.1 0.3 0.6 1.6 Lebanon 0.5 0.5 0.7 0.9 0.7 Turkey 0.6 0.6 0.6 0.6 0.6 Argentina 1.4 1.3 1.2 1.2 1.6 Brazil 1.0 0.9 0.9 1.1 1.8 Chile 0.2 0.2 0.5 0.6 0.5 Columbia 0.4 0.4 0.4 0.5 0.7 Peru 0.3 0.3 0.3 0.3 0.7 Others 42.5 40.9 38.7 37.1 27.8 Total 146,989 185,715 272,163 422,904 781,158

% women (1990) – 14% 37% 56% 14% 10% 9% 86% 43% 3% 66% 47% 61% 40% 67% 3% 43% 35% 31% 24% 32% 52% 68% 54% 71% 64% – 40%

Source: Own elaboration on the basis of professors A. Birindelli and P. Farina archives.

regulations, valid until 1986, were one of the factors producing the widespread undocumented structure of earlier flows. They contributed, however, also to a high flexibility of the labor of foreign domestic workers, thus making it an attractive substitute for other sources of personal services. The market for household services is still a major pull factor for many migratory flows having Italy as a destination (Ambrosini 2001). As matter of fact, it is a major pillar of the Italian welfare regime. Foreign workers account for 46 per cent of all registered domestic workers in Italy. Such statistics, however, are far from covering the whole segment of domestic work, which is mostly undocumented or underreported. A more accurate picture is provided by the Fondazione Ismu Lombardy Survey, which covered both documented and undocumented foreign residents (Blangiardo 2002).8 The Ismu survey data further documents the structural relevance of foreign labor in personal and household services. While the distribution of foreign workers in agriculture and manufacturing is similar to those of natives, migrants are highly overrepresented in the subsector of personal and household services. Currently, 22 per cent of foreigners work within a household. Not surprisingly,

120

G. Sciortino

Table 5. Employment by economic sector of natives and foreigners in Lombardy Economic sector Agriculture Manufacturing Construction Services of whom: Services to production Distributive services Social Services Personal services

Overall labor force (%)

Foreign workers (%)

of whom irregular (%)

2.1 32.6 7.9 57.5

1.2 30.8 11.2 56.8

17.1 10.5 35.2

13.0

9.4

29.7

19.4 17.1 7.9

10.1 3.3 33.5

23.4 5.6 34.2

Source: Own elaboration on the basis of Istat, Labor Force survey 2001 and Fondazione Cariplo-Ismu, Lombardy survey 2001.

the personal services sector is also – with construction – one where it is easier to find undocumented migrants (see Table 5). The role of foreign domestic work in the maintenance of the Italian welfare regime is duly acknowledged in Italian political debate. Throughout the 1990s, nearly half of the scarce legal admissions of foreign workers were granted to domestic workers. They also enjoyed a fair degree of tolerance in most of the amnesty programs that have marked the decade. And it is worth noting that the right-wing government in power since May 2001 has accompanied the enactment of a new repressive immigration bill with a new amnesty targeted only at domestic and personal care workers. Although the parliament subsequently extended the proposed amnesty to all kinds of wage-earning employment, the household sector has still accounted for 321,121 applications out of 702,156. An analysis of the migrants regularized in 2002 confirms the importance of foreign workers for managing the tensions embedded in the welfare regime. Of the women regularized, 83 per cent are employed by a household and 38 per cent are employed in some kind of caring tasks, usually for an employer aged 65 or more. Those employed in household services, moreover, appear different from other regularized immigrants in terms of gender and national origins (see Table 6). The migratory systems involved in this market, in other words, maintain a distinct dynamic. Furthermore, the declared salaries of the regularized workers are distinctly lower in the household service sector, where the average income is little more than half of the average income of regularized immigrants in agriculture and manufacturing.9 Household workers consequently experience a huge wage differential, thus neutralizing the tendencies embedded in Baumol cost disease. Should the large demand for foreign household workers be considered an Italian peculiarity? Comparative data is not reliable enough to allow a definitive answer. The available data, however, does show that foreign workers are overrepresented in the personal service sector in several European countries, in a way that apparently reflects the structure of their welfare regimes. All conservative-type welfare states see foreigners overrepresented in the household services sector (see Table 7). We may conclude that immigration is indeed connected with the structure of welfare states. Such connection, however, is not only a matter of welfare expenditure for and on


121

Table 6. Immigrants regularized in 2002–2003 by gender and sector of employment (main nationalities)

Regularized immigrants

% of women

% of women in generic household

20.4 15.2 7.7 7.7 5.2 5.1 4.9 4.4 2.5 26.9 100.0 702,156

44.8 85.1 18.4 12.5 65.1 37.6 78.4 71.7 64.0 27.4 45.7 321,000

46.3 43.0 51.7 50.2 48.9 22.9 45.2 43.8 53.6 52.3 45.8 147,000

Romania Ukraine Albania Morocco Ecuador China Poland Moldavia Peru Others General % Total

% of women not % of women in household employed in household work services care work 35.8 47.6 24.3 19.7 40.8 2.7 42.6 48.1 39.8 27.9 37.7 121,000

17.9 9.4 24.0 30.1 10.4 74.4 12.1 8.1 6.5 19.8 16.5 53,000

Source: Ministry of Labor and Social Welfare, Rome, data released in December 2003. Data is provisional as many applications are still under review.

Table 7. Employment of foreigners by service sector, 1999–2000 average Country Austria Belgium Denmark France Germany Greece Italy Netherlands Portugal Spain Sweden United Kingdom

Health and community services

Households

11.3 12.4 26.8 8.7 12.3 4.2 6.7 12.4 10.3 8.1 23.1 20.2

0.8 0.8 7.1 0.6 19.6 10.9 0.2 6.8 18.0 – 1.6

Note: The numbers in bold indicate that foreigners are overrepresented. Source: OECD (2002).

migrants. It is above all a matter of the kind of labor demand welfare regimes produce, and of the segment of such demand that can be satisfied nearly always only with foreign labor. Managing the Migratory Consequences of Welfare Regimes The previous section has argued that each of the various types of welfare regime has significantly different impacts on the structures of immigration systems. Conserva-

122

G. Sciortino

tive-type welfare states require the availability of a large pool of household services, as the rising women participation rate and the aging of the population makes it increasingly difficult for the household to self-produce the necessary services. The market for this kind of services, however, face Baumol’s cost disease: their price oscillates between levels too high for most of the customers or too low to meet the reserve wages of natives. Immigration may then be seen as a way to diminish the severity or the prevention of Baumol’s cost disease. Immigration, in other words, functions as an alternative to the direct provision of these services by the state. Migrants, especially in the initial phases of their trajectories, are more concerned with economic incentives than with status considerations (Piore 1979). Their reserve wages, moreover, are remarkably lower than the natives’. The availability of a pool of foreign labor may consequently keep Baumol’s cost disease in check. The problem, however, is how to manage these migratory flows in a way that does not produce large-scale permanent settlement or, on the contrary, the creation of a structural segment of helots. In other words, the problem for Western European states is that they cannot push the logic of Baumol’s cost disease to its limits, introducing a distinct and segregated salary scheme for foreign domestic workers or completely denying welfare rights to resident foreigners. Comparative works show that an important feature of the contemporary Western European migratory context is resident foreigners’ ability to access key welfare provisions. In most European countries, legally resident foreigners actually enjoy access to welfare programs nearly exactly on the same terms as citizens.10 An equally important element is the principle of equal pay for equal work, regardless of nationality. Such a principle is embedded in all ramifications of immigration policies of Western European states. Such elements sharply differentiate the Western European migratory situation from those, among others, of Gulf countries or of the newly industrialized countries of SouthEast Asia (Sciortino 2000). These constraints are, moreover, at the core of the embedded liberal regime that define Western European societies, as well as having a more generalized significance for the whole compact of the various labor laws (Hollifield 1992). The existence of these constraints produces endemic strains with the economic mechanisms triggering the large-scale migration of foreign household workers. What the welfare regimes need from foreign migrants is different from what the welfare states say should be given to them. If migration is a way to manage the tensions between wages and productivity in the service sector, the ‘‘integration’’ of migrants would cause a constant renewal of Baumol’s cost disease. After a while, foreign workers start moving to sectors where wages are higher and conditions of work more appealing. As a matter of fact, legal domestic workers tend to move from live-in domestic work to hourly paid work for a pool of employers (Andall 2000, Salazar Parrenas 2001). In the medium term, they may end up leaving the household service sector altogether. Without further immigration, the domestic services sector will then again totter between lack of supply and skyrocketing costs. If the migration policy of a country is relatively liberal, such tension may well be kept in check by the migratory dynamics itself. Workers will experience the participation in the household service sector as a transitional phase, leaving it as soon as better conditions are identified and enough linguistic skills have been acquired, while new entrants will fill the gap. Such a solution would be, however,


123

completely at odds with the basic tenets of Western European migration policies that are designed to prevent any further settlement of foreign labor on Western European territory. Within these constraints, a solution is to allow a sizeable inflow of irregular migrants that will be able to provide the needed services at affordable prices without acquiring residence rights and consequently any future access to welfare resources. The movement of workers out of the domestic service sector, however, is contingent on the availability of a legal title for residence since legal residence status is a prerequisite for leaving the service sector or for climbing up its ladder. Undocumented workers, moreover, have a strong incentive to be domestic workers: households are often able to provide accommodation and they are much less likely to be raided by labor inspectors. This is precisely what is happening across Europe, where irregular migratory systems provide in most areas the bulk of household workers (Morokvasic 1994, Slimane 1995, Baganha 1998, Anderson, 2000, Ambrosini 2001). The consequences of irregular migration are, however, quite different according to the geopolitical situation of the receiving state. Where the source of foreign labor is territorially contiguous, a liberal visa policy will be enough to guarantee that irregular migration systems will be mostly circular, arranged around a succession of short stays. A strict policing of the boundaries between regularity and irregularity – matched by a benign neglect of ‘‘working tourists’’ – will be enough to prevent permanent settlement on a large scale (Morawska 2001). The situation is however different where the sources of labor are not contiguous and a tight implementation of visa policies makes circular migration unlikely. Here a sizeable segment of long-term irregular migrants – many of them are employed in households and the personalservice sector for lengthy periods – will progressively build-up. The presence of a structural segment of unregistered long-term workers raises serious worries for the receiving state. In this case, some path out of the irregularity has to be created at regular intervals. Such a process, however, has to be managed in a way that allows for the enduring role played by migrants in the functioning of the welfare regime. The form such concern takes is the introduction of filtering mechanisms, that restrict the implications of regular residence for migrants. Italy started developing an immigration policy virtually from scratch in the early 1980s; the first National Immigration Act was passed in 1986. Starting from scratch, however, does not imply having a free hand. In designing its immigration policy, Italy was constrained by a variety of international conventions as well as by its previous tradition as a sending country (Sciortino 1999). The result was an early and straightforward introduction of the principle of equal treatment and equal access to welfare measures for citizens and resident foreigners alike. The subsequent legislative evolution has introduced remarkable restrictive changes in the Italian control policy, but it has also shown a remarkable continuity as far as equal treatment is concerned. One example will suffice: the right-wing government that has ruled the country since May 2001 rushed to introduce several restrictive changes in the Italian immigration control policy. At the same time, however, it has been fairly careful in avoiding dramatic changes in existing integration policies. The section of the immigration code related to integration measures has in fact survived basically unaltered, although its implementation has considerably slowed down.11

124

G. Sciortino

Such early inclusion of foreigners in the welfare programs, however, does not imply their large presence within the system. Legal migrants’ use of the welfare programs in Italy is actually more limited and uncertain than the increase in the overall foreign population would lead us to expect (Commissione per le politiche d’integrazione degli immigrati in Italia, 2001). Such a phenomenon is surely connected with their stage in the life cycle, as most immigrants are currently young and employed. It is, however, also connected to a very sharp selective process, accompanied by a systematic use of administrative delays. The structure of the immigration regime categorizes migrants into a variety of residential statuses, with very different levels of rights and entitlements. Furthermore, there is evidence that the overall structure of such a migratory regime keeps migrants as long as possible in the categories where fewer rights are enshrined and where the protection of such rights is less stable. The most excluded group is that of undocumented migrants. As in most Western European countries, this group is entitled only to emergency medical care and to a limited degree of education for their children. A second group consists of asylum seekers, refugees and temporarily protected persons. The members of this group, whose size is quite limited, are entitled to special assistance programs of very limited duration and amount. As a matter of fact, social assistance for the group is basically left to non-governmental organizations (NGOs) and to a handful of local authorities. A third, and by far the largest, group consists of immigrants with a residence permit. Residence permits must be renewed at regular times. Even longterm residents must frequently undergo a general check, since the permits granted are usually valid for one or two years. Changes in the necessary requirements, given police officers’ discretionary power and immigration offices’ organization, guarantee that such renewal is far from being a simple process. As a matter of fact, the regular status of most immigrants is often contingent upon the willingness of police authorities and administrative staff to handle their situations in a way that keeps unbroken their legal permanence (and the rights potentially associated to such length of residence).12 The holders of a residence permit are entitled to access all the employment-related insurance programs on the same conditions of natives, as well – when and if employed – as most services provided in kind. Employment related programs, however, do not cover household work, and the same applies to most legal protections against employers. The eligibility for most services in kind, moreover, is defined in a way that put household workers systematically at disadvantage in comparison to other groups of immigrants (Zontini 2002). In general, given the frequency of their irregular employment, either as a consequence of their being in the shadow economy or as a strategy to avoid taxation, their access to employment-related programs is often illusory. A fourth group is constituted by immigrants with a residence card. Such a title, introduced only in 1998, is granted after five years of regular sojourn with an income sufficient to sustain both the immigrant and their dependents. Residence cardholders are fully included in the welfare system and are eligible for non-contributory cash benefits and income support programs. The case of the residence card is a particularly clear example of the filtering function of these statuses. When the residence card was introduced by the center-left government in 1998, it was clearly targeted to the creation of stable denizen status for long-term residents.13 In the bill, the requirements for the granting


125

of such a card were quite straightforward: five years of lawful residence, no criminal record and evidence of being lawfully employed. When the immigration bill was passed, around 65 per cent of the regular foreign residents had been living in Italy for more than the required five years. The residence cardholders were expected to number in the hundreds of thousands. This, however, has not happened. From one side, the whole process received a very low administrative priority, with territorially contiguous offices adopting different procedures and requesting different – and additional – paperwork (Fondazione Cariplo-Ismu, 2000). At the state level, the Home Ministry released an administrative interpretation of the norm that allowed only a tiny percentage of the foreign residents to apply for the card. Although such a restrictive administrative interpretation was overturned in the summer of 2002, the number of residence cardholders is currently little more than 20,000, out of a resident foreign population of 1.5 million. Such restrictive management of migrants’ statuses is even more evident in naturalization laws. In 1992, the houses reformed the citizenship law in a way that severely restricted the eligibility for naturalization on the grounds of prolonged residence. Such reform, approved by a large parliamentary majority without visible opposition, may be taken as a proof of a strong, though not outwardly very visible, exclusionary view in the Italian polity. Before the 1992 citizenship law, the Italian state allowed foreign residents to apply for naturalization after five years and, in certain circumstances, after three. The reformed 1992 law requires 10 years of uninterrupted residence before an application for Italian citizenship can be made. The law, moreover, makes explicit the introduction of an ethnocultural principle in the definition of the boundaries of the Italian polity. According to the new law, remote descendants of Italian emigrants are embraced within such a polity, while the naturalization of individuals with no Italian ancestry is heavily discouraged. The emphasis on ius sanguinis is increased enormously: the right to citizenship is extended to any grandson or granddaughter of an Italian citizen (provided they have lived in Italy for three years), whereas ius soli is applied only to children born in Italy from unknown parents. The strategy seems to have been quite successful. Since the 1992 law, the number of newly naturalized foreigners is quite limited (Pastore 2002). The analysis of the Italian migratory regime reveals an embedded preference for discretionality, as well as a systematic filtering designed to keep migrants within the less protected statutes for as long as possible. There is consequently no evidence that such filtering is directly aimed at restricting foreigners’ access to welfare state services or to keep them within the boundaries of certain economic sectors. At the same time, it would be difficult to deny that such generalized preference has, as a by-product, two important consequences: it severely reduces migrants’ access to the welfare system and it provides a structural incentive for them to work in personal and household services. The actual configuration of the migration regime, in other words, does produce a context whereby the demand for foreign labor originated by the welfare regime can be met. Conclusions The article has presented an introductory exploration of the ways in which welfare regimes interact with migration regimes. Using Italy as its main reference point, it

126

G. Sciortino

has been argued that such interaction cannot be understood by looking only to the welfare provisions immigrants may expect to enjoy. The theory of the welfare magnet should be seen in a very different way: it should be argued that different kinds of welfare regimes produce different demands for foreign labor. Welfare states’ drawing power, however, is not only a matter of the kind of welfare provisions immigrants may expect to enjoy. Migration is also a resource for managing the relationships between households, labor markets and the state. In this vein, the study has shown that the Italian migratory system can hardly be understood without taking into account the structural tension between a welfare system largely designed according to a male-breadwinner logic and Italian women’s rising labor participation. Such a tension triggers specific migratory systems, providing household workers at conditions able to escape the logic of Baumol’s cost disease. It is likely that the same tension operated in other conservative-type welfare regimes. A second, and related, argument has been developed in reference to the basis of integration policies in Western European states. This essay has contended that an adequate analysis of the integration policies enacted by those states cannot be performed using undifferentiated notions of either welfare or immigrants. It is known that different programs within a welfare system may have widely different access rates according to their intrinsic logic. An equally important point, however, is that each program has its own conditions of eligibility. Such eligibility, however, is contingent on the kind of legal statuses available to the resident foreigner. Focusing on eligibility conditions, it is possible to understand the ways in which Western European states deal with the large demand for foreign household workers, trying at the same time to prevent their settlement. Notes I wish to thank Paolo Barbieri, Tomas Hammar and Dietrich Thra¨nhardt for suggestions and criticisms of an earlier version of this essay. The same gratitude applies to the anonymous reviewers. 1. If calculated per head of population in purchasing power parity, the difference is 7 4 per cent (Eurostat 2000). 2. According to such a perspective, such similar features would derive – or at least be interconnected to – a sharply polarized labor structure: a large section of the population (public employees, employees in medium and large firms, and some sectors of agricultural work) are supposedly sheltered from any short- or medium-term risk while enjoying generous incomes, whereas other sections (employees of small firms, the long-term unemployed, the young and those who have never worked, and the employed in the shadow economy) are basically excluded from protection or only marginally included through political patronage. 3. Besides the high percentage of people aged 65 or more residing outside institutions, it is worth noting that, in 1999, in Italy, there were only 4.3 institutionalized disabled for every 10,000 individuals (Istituto nazionale di statistica 2002). 4. Although the number of females in the Italian labor market is growing, the increase is unable to match the more sustained growth of Western Europe as a whole. While the Italian female workforce was 8 percentage points below the EU-15 average in the 1960s, it is today nearly 13 percentage points behind (Eurostat 2000). 5. For a history of the Italian domestic service sector, see Sarti (2004). 6. Such a familistic vision did in fact help in keeping the availability of domestic workers highly flexible: a law enacted in 1958 still failed to limit the length of the working day and to remove the traditional exemption of domestic work from a whole range of protective employment norms (Andall 1998).


127

7. The early history of the Italian migratory system is poorly documented and little researched. Certainly, however, the flow of domestic workers is a major segment of the earlier flows. Not the only one, though. Migration from China has been recorded as early as the 1930s, while local migratory systems have never stopped linking Sicily with North Africa and the Italian north-east with the territories of the former Yugoslav federation. A further link was established by Africans and Asians moving to Italy in the aftermath of the immigration restrictions of the early 1970s. Much later, the collapse of the Soviet bloc reopened Italy to migratory systems rooted in Eastern Europe and the Balkans (Colombo and Sciortino 2004). 8. Lombardy, besides being one of Italy’s economic engines, has a large metropolitan area (Milan) and a very large female workforce. In short, it is a region where the identified structural tension should be felt with particular intensity. 9. The real wage differential is however likely to be lower. Household workers, in the frequent case of live-in contracts, receive part of their salary in kind, through the provision of lodging and food. 10. For an attempt to identify the impact of such a situation on the meaning of citizenship in the contemporary political arena, see Soysal (1994). 11. For an analysis of the main changes introduced by the right-wing government to the Italian immigration policy, see Fondazione Ismu (2003). The principle of equal treatment seems also to be popular (or at least not so unpopular) among the Italian population. Opinion surveys document how Italian public opinion’s fairly negative vision of immigration goes hand in hand with a fairly liberal attitude towards the rights of legal foreign residents (Ispo 2000, Diamanti 2001). 12. There is a good deal of evidence, as documented also by quite frequent judicial reviews, that a residence permit does not provide security and stability in itself. Many immigrants have difficulties in documenting an acceptable amount of work activities when they renew their permits, while a variety of minor infractions of immigration rules may turn out to be fatal for the holding of such status. Furthermore, there is evidence of an astonishing degree of difference in the ways in which even territorially close offices apply the procedures and interpret the norms. 13. On the notion of denizens, see Hammar (1990).

References Ambrosini, M., 2001, La Fatica Di Integrarsi. Immigrati e Lavoro in Italia (Bologna: Il Mulino). Andall, J., 1998, Catholic and State Construction of Domestic Workers: The Case of Cape Verdeans Women in Rome in The 1970s, in: K. Koser and H. Lutz (Eds) The New Migration in Europe (London: Palgrave), pp. 124–141. Andall, J., 2000, Gender, Migration and Domestic Service: The Politics of Black Women in Italy (Aldershot: Ashgate). Anderson, B., 2000, Doing the Dirty Work (London: Zed Books). Anderson, P. 1997, In a Twilight World: Undocumented Migrants in the United Kingdom, available at: http://www.geocities.com/jrsuk/twilight.htm (accessed 4 June 2003). Baganha, M. I., 1998, Immigrant Involvement in the Informal Economy: The Portuguese Case. Journal of Ethnic and Migration Studies, 24 (2), 367–386. Baumol, W., 1967, The Macroeconomics of Unbalanced Growth. American Economic Review, 57, 415–26. Baumol, W., Batey Blackman, S. and Wolff, E., 1985, Unbalanced Growth Revisited: Asymptotic Stagnancy and New Evidence. American Economic Review, 75, 806–817. Bird, E. J., Kayser, H. and Frink, F. R., 1999, The Immigrant Welfare Effect. Take Up or Eligibility? IZA discussion paper 66, Bonn: IZA. Blangiardo, G., 2002, L’immigrazione straniera in Lombardia: La seconda indagine regionale (Milan: Fondazione Ismu). Brucker, H., Epstein, G., McCormick, B., Saint-Paul, G., Venturini, A. and Zimmerman, K, 2001, Managing Migration in the European Welfare State (Milano: Fondazione Debenedetti). Castles, F. E., 1993, Families of Nations: Pattern of Public Policy in Western Democracies (Aldershot: Ashgate). Colombo, A. and Sciortino, G., 2004, Italian Immigration: The Origins, Nature and Evolution of Italian Migratory Systems. Journal of Italian Modern Studies, 9 (1), 49–70.

128

G. Sciortino

Commissione per le politiche d’integrazione degli immigrati in Italia, 2001, Secondo Rapporto Sull’Integrazione Degli Immigrati in Italia. G. Zincone ed. Bologna: Il Mulino. Diamanti, I., 2001, Immigrazione e Cittadinanza in Europa (Venice: Quaderni fondazione Nord-Est). Esping-Andersen, G., 1990, The Three Worlds of Welfare Capitalism (Cambridge: Polity Press). Esping-Andersen, G., 1999, Social Foundations of Postindustrial Economies (Oxford: Oxford University Press). European Commission, 1998, Employment rates report 1998. European Commission, DG Employment and Social Affairs, Brussels. Eurostat, 2000, European Social Statistics: Social Protection 1980–1998 (Brussels: Eurostat). Ferrera, A., 1984, Il Welfare State in Italia (Bologna: Il Mulino). Ferrera, M., 1996, Il Modello Sud-Europeo Di Welfare State. Rivista Italiana Di Scienza Politica, 26(1), 67–101. Fondazione Cariplo-Ismu, 2000, Sesto Rapporto Sulle Migrazioni 2000 (Milan: FrancoAngeli). Fondazione Ismu, 2003, Nono rapporto sulle migrazioni (Milan: FrancoAngeli). Glover, S. G. C., Loizillon, A., Portes, J., Price, R., Spencer, S., Srinivasan, V. and Willin, S. C., 2001, Migration: An Economic and Social Analysis. Occasional paper No. 67, London: Home Office, RDS. Hammar, T., 1990, Democracy and the Nation-State: Aliens, Denizens and Citizens in a World of Intenational Migrations (Aldershot: Avebury). Hansen, G. H., and Lofstrom, M., 1999, Immigrants and Welfare Participation. IZA report 100. Bonn: IZA. Hollifield, J., 1992, Immigrants, Markets and States: The Political Economy of Postwar Europe (Cambridge, MA: Harvard University Press). Ispo, 2000, L’Atteggiamento Degli Italiani Nei Confronti Degli Immigrati (Rome: Commissione per le politiche d’integrazione degli immigrati). Istituto nazionale di statistica, 2002, La Situazione Del Paese Nel 2001 (Rome: Istat). Leibfried, S., 1991, Towards a European Welfare State?, in: G. Room (Ed) European Development in Social Policy (Bristol: Bristol University Press). Levine, P., 1999, The Welfare Economics of Immigration Control. Journal of Population Economics, 12(1), 23–43. Morawska, E., 2001, Gappy immigration control, resourceful migrants and pendel communities, in: V. Guiraudon and C. Joppke (Eds) Controlling a new migration world (Routledge: London), pp. 173–199. Morokvasic, M., 1994, Pendeln statt auswandern. Das Beispiel der Polen, in: M. Morokvasic and R. Hedwig (Eds) Wanderungsraum Europa. Menschen und Grenzen in Bewegung (Berlin: BVS Verlag), pp. 116–141. OECD, 2001a, Social expenditure database, Available at http://www.oecd.org. OECD, 2001b, Society at Glance: OECD Social Indicators (Paris: OECD). OECD, 2002, Trends in International Migration 2001 (Paris: OECD). OECD, 2003, Trends in International Migration: Annual Report 2002 (Paris: OECD). Pastore, F., 2002, La Communita’ Sbilanciata. Diritto della cittadinanza e politiche migratorie nell’Italia post-unitaria, CeSPI Working Paper, 6/2022, Rome. Piore, M., 1979, Birds of Passage: Migrant Labor and Industrial Societies (New York: Cambridge University Press). Riphahn, R. T., 1998, Immigrant Participation in the German Welfare Program. Finanzarchiv, 55, 163– 185. Roodenburg, H., Euwals, R. and ter Rele, H., 2003, Immigration and the Dutch Economy (The Hague: CPB Netherlands Bureau for Economic Policy Analysis). Salazar Parrenas, R., 2001, Servants of Globalization: Women, Migration and Domestic Work (Stanford: Stanford University Press). Sarti, R., 2004, ‘‘Noi abbiamo visto tante citta`, abbiamo un’altra cultura.’’ Migrazioni, identita` di genere e servizio domestico in italia: uno sguardo di lungo periodo. Polis, XVII, 114–47. Sciortino, G., 1999, Planning in the Dark: The Evolution of the Italian System of Immigration Controls, in: G. Brochmann and T. Hammar (Eds) Mechanisms of Immigration Controls (Oxford: Berg), pp. 233– 260. Sciortino, G., 2000, L’ambizione della frontiera: Le politiche di controllo migratorio in Europa (Milan: FrancoAngeli).


129

Sciortino, G. and Colombo, A., 2004, The flow and the flood: immigrants in the Italian newspaper discourse. Journal of Italian Modern Studies, 9, 94–113. Slimane, L., 1995, L’immigration clandestine de main d’oeuvre dans la re´gion bruxelloise (Brussels: Bruylant). Soysal, J., 1994, Limits of Citizenship: Migrants and Postnational Membership in Europe (Chicago, IL: University of Chicago Press). Spencer, S., 1994, Immigration As an Economic Asset: The German Experience (Stoke-on-Trent: Trentham Books). Tapinos, G., 2000, Irregular Migration: Economic and Political Issue, in: OECD, Combating the Illegal Employment of Foreign Workers (Paris: OECD). Thalhammer, E., Zucha, V., Enzenhofer, E., Salfinger, B. and Ogris, G., 2000, Attitudes Towards Minority Groups in the European Union: A Special Analysis of the Eurobarometer Year 2000 Survey (Vienna: European Monitoring Center on Racism and Xenophobia). Zontini, E., 2002, Female Domestic Labour Migrants and Local Policies in Bologna: The Story of a Filippino Woman, in R. Grillo and J. Pratt (Eds) The Politics of Recognizing Difference (Aldershot: Ashgate), pp. 159–176.