Abstract

The first generation of post-World War Two (WWII) migrants to Western Europe have joined the ranks of the European elderly. Typically, they have accumulated limited assets because of lower paid employment and earned limited rights within public pension schemes that in most European

countries relate pensions to years of residence or contribution. In some respect they share the problems of those of the European-born population, mainly women, who started work late, have interrupted work histories, or have low life-time earnings. This paper argues that changing the rules for those sections of the European Union (EU) population who are reliant on public pensions will help to reduce the risk of poverty in old age among some of the most underprivileged sections of the EU elderly population. Among these policies are: delinking length of residency from pension rights or giving credit for the years that a person lived outside the country, financial support to those who are short of contribution years, allowing people to take their pension abroad without a penalty, and improving the rights and entitlement of the surviving spouse (usually women).

Introduction

The ageing population of the European Union (EU) has provided new challenges for policy makers that go well beyond the preoccupation of orthodox economics with the solvency of public pension schemes. It has opened the debate on population policy--'how could the EU replenish its rapidly ageing population?' To answer this question the EU has to look beyond the European continent where the population is ageing as well. There is nothing new in this, the reconstruction of Europe after the Second World War required similar re-thinking about population movement and led to organised immigration of labour from different parts of the world, notably from the Caribbean, the Indian sub-continent, north Africa, and Turkey to Europe. These migrants have now joined the ranks of the elderly in the EU and form the majority of the non-EU ethnic population. This paper will build on the literature on ageing in Europe that has been conducted under the EU's Programme of Actions for the Elderly and the EU Observatory On Ageing and Older People and will also draw on the wider literature on the economics of ageing. (2) The research agenda of the Observatory provides a useful starting point for setting up our comparative study of ageing among migrant communities. The Observatory covered the following topics for its country studies: '1--living standards and way of life; 2--employment and the labour market; 3--health and social care; 4 --the social integration of older people in both formal and informal settings'. (3)

How do current pension and welfare schemes in EU countries provide for elderly migrants? Are ethnic elders treated the same as the rest of the old population or do they suffer from what has been referred to in the literature as 'double jeopardy' thesis--paying the penalty of age as well as race. Answering such questions would help us to search for policies that should shape future plans for providing old-age welfare for recent migrants to the EU. In dealing with the above issues this paper will also try/seek to answer the question: 'Do ethnic minorities face a more uncertain economic future than others when they age in the EU?' by looking at the entitlement of the present generations of ethnic minorities to pensions and other social security provisions.

Population ageing in the European Union

As a consequence of declining fertility and rising life expectancy, one in five people in Europe were over age 60 in 1999, a ratio that is going to increase to one in four by 2020 and one in three by 2040 (European Commission, 2000). One of the consequences of population ageing is the rise in the elderly dependency ratio. In 1999, for every person of aged 60 or older there were 2.6 people of working age (20-59) in the EU. With the fall in birth rates this ratio is going to decline in the next 20 years. These general trends conceal an important fact that the ageing population is not a homogenous mass but is differentiated by social class, gender, ethnic background, and race; factors that shape the social and physical ageing of individuals and largely determine their well being in the old age. Such differences have to be taken into account if social and economic policies are to be successful in their objective of providing reasonable living standards for the mass of the elderly population in the 21st century.

Ethnicity and Ageing

Who are the ethnic populations of Europe? Answering this question is not easy. Defining ethnicity is fraught with difficulty. Ethnicity could be read as demarcation lines between different groups. From the point of view of the majority group it is the 'otherness' that becomes the defining characteristic of an ethnic minority. The 'white' majority population may perceive all non-whites as one ethnic group. For example it is common in the US to refer to non-whites as 'ethnics', whilst in Australia those whose first spoken language is not English are referred to as 'ethnic'. In these examples colour and language of the majority are used as demarcation lines which indirectly assign the title ethnic to the 'other' groups.

The categories of colour and language as well as religion, culture, country of origin and national background have been used to provide a direct definition of an ethnic grouping. This approach could be refined further by relying on self-definition rather than on an outsider's view of who is a member of an ethnic group. Though this is a much more democratic way of defining a group as ethnic, it still has the problem of masking socio-economic differences within ethnic groups, differences which reflect and determine the present and future standard of living of individuals. Self-definition could also be approached from 'above' or 'below'. In the former case the upper echelon within an ethnic group could define who belongs to that group and thus use this as a means of social control, whereas in the latter case groups and individuals could locate themselves within an ethnic grouping. Recent British population censuses of 1991 and 2001 used the latter approach to identify different ethnic groups that in published results are consolidated under broad categories. (4) At EU level there is no uniform system for recording the number of ethnic people living in Europe.

According to the latest Organization for Economic Cooperation and Development (OECD) data on international migration, in 1998 the foreign-born and foreign population (5) in the European Economic Area (EEA) totaled 20 million (6) (OECD, 2001). They constituted 5 per cent of the EEA population. It is important to point out that the foreign-born and foreign population include migrants from developed (e.g. European, US) as well as from developing countries. The proportion of European migrants varies greatly from country to country but the largest groups live in the countries with the smallest populations. (7) The great majority of the foreign-born and foreign populations originate from outside Europe, and mainly come from developing countries. Women constitute approximately half of these foreign born populations, mainly on the account of family reunion and family linked migration. Men dominate the flow of new migrants into the EEA, and are then followed by their families (OECD, 2001, box 2: p. 25).

These migrants moved to Europe in two distinct phases (1945 to early-1970s and mid-1970s to the present) that followed the economic cycle in Europe in general and the EU in particular. The first phase started in the immediate post-WWII years when migrants were drawn from the peripheral and economically depressed areas of Europe such as Italy, Spain, and Portugal as well as the less developed countries with colonial and cultural links to Europe such as India, Pakistan, the Caribbean islands, and the north African countries of Algeria, Morocco, Tunisia, and Turkey. Many of these migrants have now reached the end of their working lives and have joined the ranks of the pensioners in their countries of residence. The second phase started in the late 1970s and continues to the present (Castles and Miller, 1993). It is because of this second phase that the ethnic population in Europe has a younger age structure than the indigenous national population (European Commission, 2000b). These second generation migrants and new migrants face similar racial and other prejudices as previous generations of migrants when it comes to job discrimination. A recent study of migrant and ethnic minorities in four European countries (Belgium, Germany, Netherlands, and Spain) found that ethnic minorities were severely discriminated at various selection stages and within their sectors of employment. The most severe cases and rates of discrimination were observed in the service sector, particularly in occupations that have high visibility such as waiters and sales assistants (Zegers de Beiji, 1999). These are disturbing results that reveal much about the attitudes of employers. More on this later.

In assessing the welfare of migrants in old age we first take a brief look at the economic position of the first generation with regard to two key determinants of pension entitlements and assets.

Eligibility for and Entitlement to Social Insurance

Social insurance and social protection (mainly consisting of health, old-age/survivors and unemployment benefits) vary greatly across the EU countries due to differences in the national history of social security system as well as in social needs and demographic structure. In most countries old-age/survivor benefits represent the largest component (about 50 per cent) of total social expenditure, followed closely by health expenditure.

To qualify for social insurance in an EU country a person has to satisfy two broad conditions of eligibility and entitlement. To be eligible for social insurance in old age a person has to be gainfully employed before retirement in most countries except in Denmark, Lichtenstein (non-EU), Norway (non-EU), Sweden and the Netherlands where the entire population is included in a universal system. (8) Entitlement to social insurance requires residency and/or contributions to the system before retirement (European Commission, 2001). Besides eligibility and entitlement, the EU and EEA countries also differ in the ways they calculate their benefit payment. This determines the amount of benefit paid out and is related in most cases to the number of years of residency and/or of making contributions. In all countries there are also a variety of occupational pension schemes that can provide a far larger old age pension than that provided through state social security arrangements.

These general principles as well as the finer details of pension law of any country would discriminate against the ethnic/migrant population to the extent that they differ from the indigenous population with regard to: period of residence and contribution, sector of employment, legal status, and access to occupational pensions. Obviously some of these differences also exist within the indigenous population, and especially between men and women, with the latter often building up less entitlements to pensions because of a greater likelihood of interrupted work histories, participation in part-time work, and the receipt of lower levels of pay.

An important rule for social insurance that could affect migrants is whether they are required to be resident to claim and to receive their pension, the two being distinct procedures. (9) This is an important distinction for those migrants who may well want to retire to their country of origin. While retiring within the EU does not pose a problem with either the claim or receipt procedure, it is retirement to outside the EU that is a problem. Information is not available across all EU or EEA member countries on residency as a precondition for 'claiming and receiving' a pension. Only four countries (Belgium, Denmark, Netherlands, and the United Kingdom (UK)) have reported to the European Commission that they do not require residency as a precondition for receiving a pension. However, the same countries, except the UK, would require a person to be resident in order to claim a pension. The UK is the only country where residency is not required for claiming or receiving a pension, but British pensioners living outside the UK do not receive the annual inflation-linked increase to their pensions (European Commission, 2001a).

Another pension rule that could affect immigrants adversely is whether the pension paid is related to years of residency (or work), to earnings or to contributions to the pension scheme. All countries apply a residency or work rule for receiving a pension--you should have lived or worked in the country before qualifying to receive a pension. In only four countries, Denmark, Ireland, Netherlands, and Austria, is the minimum pension not related to earnings or contributions (European Commission, 2001a).

In countries where early retirement is based on years of work rather than age, immigrant workers could also be adversely affected. In Luxemburg for example, 40 years of work are required to qualify for early retirement. Greece, on the other hand, has the lowest number of qualifying years (15) for early retirement (European Commission, 2001a).

In sum, immigrants' entitlement (both in terms of amount and flexibility) to any public pension system would be lower than that of the rest of the population because of immigrants' shorter working life, and likely lower earnings and contributions. When we extend this discussion to survivors, mostly relating to women who live longer than men, the rules for the non-immigrant population normally apply, with a married spouse having most rights. In all cases, the surviving spouse will receive benefits according to the entitlement of the deceased (European Commission, 2001b, table VII). (10)

Residency/Nationality Status

Since the legal status of an immigrant could have some bearings on his or her eligibility for, and entitlement to, state pensions, it is instructive to briefly survey the residency status of immigrants and their rights to naturalisation across the EU.

Naturalisation procedures in most countries operates according to principles of ius sanguinis and/or ius soli--the former is based on descent from a national of the country while the latter is based on birth in the territory of the country. These principles are rooted in the way states and nations have been formed and have come to define themselves throughout history. The folk or ethnic model of the nation-state of e.g. Germany and Austria has adopted the principle of ius sanguinis, though in recent years Germany has allowed migrants and their German-born children to apply for German citizenship. Before this change, citizens of the former Soviet Union (FSU) who were of German descent and had lived in the FSU and other eastern European countries did qualify for German citizenship (and indeed many migrated to Germany after the collapse of the Soviet Union) but children of other migrants born in Germany did not. States that have been formed through incorporation of diverse groups in a single territory such as the UK and the Netherlands or through immigration such as the United States (USA) and Australia follow the principle of ius soli--anybody born in the country is a citizen, and descendents would also have the right of citizenship as well. In general the citizenship rules of the countries following the ius soli principle are more liberal than those of countries that put the emphasis on lineage as the prime condition for citizenship. The ius soli countries usually allow long-term residents to apply for citizenship along the principle of ius domicili.

An important aspect of naturalisation that is of usual concern to immigrants is whether their host country would allow dual citizenship. Maintaining original citizenship is important to immigrants not only for psychological feelings of belonging to a land and culture that they have left behind, but also for practical reasons of being able to go back 'home' without the need to obtain visas, and more importantly not to lose some of their civil rights such as the right to vote or own property. These are important considerations when it comes to the decision to retire to or spend time in the 'home' country. Rules on dual nationality vary across EU countries. Austria, Denmark, Finland, Germany, Luxembourg, and Sweden legally require renunciation of the former nationality at the time of naturalisation while France and the UK allow dual citizenship (Guimezanes, 1995). These rules are subject to regular reviews and changes. The general trend is towards the acceptance or tolerance of dual citizenship (Castles and Miller, 1998). In the author's view, dual citizenship is an acknowledgment of dual or multiple identities in a world where mobility and migration have become an important and integral part of international economy and culture.

The nationality status of the second-generation migrants who are born in the host country is somewhat more secure than that of their parents. In countries following ius soli principle (e.g. the UK and Ireland), children of migrants born in the host country are entitled to citizenship provided that at least one of the parents were a citizen or permanent resident. In France, Netherlands, Belgium, and Italy a combination of ius soli and ius domicili is applied: a child born to an immigrant is entitled to citizenship after being resident for a certain period having fulfilled certain conditions (such as a language test in the Netherlands).

Savings and Accumulation of Assets

The duration of the work period and income during one's working life are important determinants of financial security in old age that are related to the accumulation of assets (including housing) and the build-up of various forms of financial savings as well as 'rights' under private pension funds and state pension schemes. Not only has income to be large enough to accommodate consumption and saving, it has to be regular to allow the build-up of assets. Intergenerational transfers (such as inheritance) also contribute to the long-term accumulation of assets.

The majority of the first generation post-WWII migrants score poorly on the above criteria for financial security in old age because of their inferior position in the labour market. They were mostly employed in declining industries (e.g. construction, engineering, textiles, and domestic services), worked in low skilled and low paid jobs with little prospect of promotion, and had suffered very poor working conditions with grave consequences for their long term health, especially in old age. They were also more at risk of unemployment--they were the first to be fired in the case of any downturn in business. Once they were laid off, they remained unemployed for longer period than their indigenous counterparts. Initially what was behind this low status was the fact that migrants were recruited into low skilled and low status jobs (Castles and Kosack, 1985). It has also been argued that this type of migration attracted migrants who wanted to earn and save money for a few years and then return home. Such migrants would not necessarily see any value in investing in language skills and education to improve their lot. However, the more important factors were a combination of legal and social discrimination practices that not only kept migrants in such jobs but also prevented access to better jobs for qualified migrants (Castles and Kosack, 1985).

Could these migrants accumulate savings for their old age? For the majority this would be possible only if they had sent some money home where the real value of their savings would be higher. Their savings would be too meagre in the host country, a fate that they share with other low paid workers. All they could do is build up claims to the state pension. In countries where entitlement is related to years of contribution (e.g. UK) migrants and those who start work late would be at a disadvantage. Where pension is related to years of residence during one's working life (15-65, as in the Netherlands) once again migrants would lose out.

When it comes to housing assets, migrants are in no better a position. Most of the migrants were expected to fill specific gaps in the labour market and were considered as temporary workers. Immigration permits were job--and employer-specific and often immigration rules expected the employers to provide housing for the migrants for the duration of employment. In reality only German employers housed a large section of the single migrants, albeit often in substandard accommodation. In other countries with large numbers of migrants the private market provided most of the housing. The low income of migrants and their low status meant that they lived mainly in the most marginal urban areas. Where public housing was available, rules such as minimum residency and minimum income disadvantaged migrants, as did the usual practices of racial and ethnic discrimination. Castles and Kosack (1985) conclude that: 'the low socio-economic position of immigrants in Western Europe finds one of its most obvious expressions in the sphere of housing, where immigrants are part of the lowest stratum of society' (p. 314). In short immigrants age without accumulating much in the way of housing assets or in the case of public housing tend to have the worst types of housing compared with the indigenous population.

A transient and temporary status means that migrants view the question of savings and investment in assets such as housing more in the context of their countries of origin--saving and remitting money--rather than in relation to their needs in the destination country. The low earnings of migrants mean that they would have low savings with far less purchasing power at destination than at origin. This is another major reason for the remittance of savings.

Once migrants' families are reunited and children are born and start their education in the new country the parameters of migrants' lifetime decisions change. Perhaps within 10 years of their arrival the plan to go home becomes but a distant dream for most migrants. Psychological costs of this realisation aside, the practical costs are also large since migrants would be behind the indigenous population by about a quarter of a working life (around 10 years) in terms of building of accumulated assets and entitlements to pensions, in part because of the likely decision to remit their savings rather than save in a pension. This would apply mainly to low-income and low-status migrants who, in practice, form the majority of migrants.

The final source of long-term savings to be considered is inherited assets that are located in the country of origin. If migrants decide to transfer their inherited assets to the host country they face several problems. Non-monetary assets have to be liquidated and the proceeds converted to hard currency. In most developing countries the hard currency value of assets is low compared with similar assets in developed countries (that is the countries of destination) implying that the inherited assets would not significantly increase the overall savings of immigrants.

Non-pension Sources of Income and other Social Support

Private pensions and statutory public pensions are not the only source of income maintenance and social support in old age. Available data on sources of income of the elderly in a selected number of developed and developing countries show that in general the richer the country the higher the percentage of those older than age 60 relying on pensions/welfare for their livelihood (World Bank, 1994). In low-income countries, except in urban areas of China, family and own work are the main sources of income. The data also suggests that as countries get richer formal systems gain in importance in providing support to the elderly. In the high-income countries of the EU, social protection--other than pensions--is directly related to whether the main state pension provides an adequate standard of living in old age.

All EU countries (except Greece) have some kind of a general (social assistance) scheme to guarantee a basic minimum standard of living that, in principle, could supplement individual pension entitlements that fall below this minimum standard of living. In countries where there is a gap between the two, it is usually due to the fact that pensions are related to years of contribution and/or years of residence. In all countries the scheme is targeted at younger age groups. Only in the UK is the scheme also used as a supplement to state pension. In all countries except the UK, claimants must be 'willing to work' to qualify for supplementary benefits. Entitlement to a supplementary income is in most countries a matter of individual right that also takes account of family circumstances and household size (European Commission, 2001b: pp. 516-519). Supplementary income schemes are generally open to legal residents, irrespective of nationality, that entitlement includes refugees, asylum seekers, and stateless persons. Nor are there many restrictions with regard to minimum age. Some countries operate other schemes that are targeted at the older generation. These are in essence non-contributory pension schemes and are in operation in Spain, France, Ireland, Italy, Lichtenstein, Portugal and the UK. These schemes are in the main anti-poverty measures. They do not discriminate against the immigrant community on paper, yet in so far as access to them is based on application they could be discriminatory in practice if the immigrant community does not have the information, language or legal skills to apply for them. Considering the lack of information at this stage on the uptake of these schemes we cannot go any further in evaluating their effectiveness. Other countries rely solely on their main pension system to provide a minimum standard of living standard for the old.

Migrants and Labour Market Discrimination

Our story began with the post-war migration flows and one would imagine that, after nearly 50 years of struggle to counter discrimination and racism, the second generation and new migrants would have had more equitable access to the labour market. Alas the available evidence points in the other direction. According to a recent International Labour Organization (ILO) sponsored study labour market discrimination against non-white people in Belgium, Germany, the Netherlands, and Spain is rife (Zeger de Beijl, 2000).

Objectives of the ILO research programme were: (1) to test discrimination at the entry level for semi-skilled jobs, (2) to measure the occurrence of discrimination against second-generation migrant and minority males in the 20-25 age group. Selected regions were characterised by a high concentration of second generation and minority groups and by relatively high demand for labour in industry and service sectors. The study used a 'situation testing' method to focus on the process of discrimination in real life situations where people make decisions about those who apply to them for positions. A tight procedure was set up to ensure accuracy and reliability of the results. Teams of two job seekers with the same age and skill profiles but one with ethnic/minority racial characteristics and the other with majority population characteristics were trained to apply for semi-skilled positions using exactly the same procedures. They could apply for jobs advertised or make unsolicited approach to employers. It is clear that at the outset the project was keen to highlight the labour supply side factors of language or skill as an explanation for refusing an immigrant applicant. In all countries, regions were chosen that have traditionally had high concentrations of migrants.

In the German region of North Rhine-Westphalia, the researchers observed severe cases of discrimination against second-generation Turkish immigrants who found that one fifth of all vacancies were closed to them because of discriminatory practices. A significant number of discriminations occurred in small service sector enterprises, a very worrying sign given that these enterprises have been at the forefront of job creation in recent years. For the first generation migrants supply-side characteristics, such as German language proficiency, and their age as well as economic restructuring could, to some degree, explain higher than average unemployment. But such supply-side factors are not adequate explanatory factors for the high unemployment among the second-generation migrants who were born and brought up in Germany. The German case study 'provides clear evidence that the unfavourable labour market position of migrants and their disproportionately high level of unemployment (at two and half times the rate of their German counterparts) is, to a considerable extent, attributable to the discrimination they face when they apply for jobs' (Zeger de Beijl, 2000: p. 63). Similar stories are repeated across other countries.

In the Netherlands, discrimination against Moroccan job applicants was widespread with one in three job vacancies closed to them. The researchers state that similar results would have been obtained if Turkish immigrants had been used, a group whose unemployment rate is even higher than that of Moroccans. Indeed when testers from a Surinamese ethnic group were selected, similar results were again obtained. However, the discrimination rates declined to one in fifth when Surinamese testers presented themselves at a higher skill level, indicating that some relationship might well be present between skill/education and lower levels of discrimination.

Spain presented an interesting case. For economic and political reasons, Spain has a long history (centuries rather than decades) of emigration that has been maintained well into the 20th century. As a result more Spanish people live outside Spain than there are foreigners living in Spain. In 1996 there were just over half a million foreigners (half of them from EU countries) living in Spain. With such a history one would expect a less discriminatory approach by Spaniards than in other European countries. Alas, this is not so.

In 1996, Moroccans were the largest group of all foreign residents of Spain (77,000) followed closely by the British (68,400) and Germans (45,900). The majority of EU immigrants were retirees and therefore not in the labour force. Other significant groups of immigrants were from Spanish former colonies of Argentina, Peru, and the Dominican Republic. The largest group of foreign workers (36.5 per cent) was Moroccan. The research focused on the industrial areas of Madrid and Barcelona and the main tourist area of Malaga that have had the largest concentrations of the target group: Moroccan youth between the age of 20 and 25. The testers found that one in three jobs were closed to Moroccan job applicants in Madrid while in Barcelona and Malaga provinces one in two Moroccan applicants were rejected. These results should be taken as representative of discrimination against migrants in general, except perhaps in the case of Spanish speaking Latin American migrants whose phenotypes and culture are closer to the majority population. It is important to point out that Spanish language was not an issue in the selection process, as the testers as well as second generation Moroccans had no language problems.

In Belgium foreigners comprised about 9 per cent of the population in 1996, the majority of whom (60 per cent) were from other EU member states. Migrants from the non-EU countries were mostly Moroccans and Turks, who were mostly employed below their real skill levels and had followed fragmented and insecure career paths, that in turn would affect their claims to job related or contribution related pension schemes. These migrants were also severely affected by the rise in unemployment since the early 1990s, with unemployment rates twice as high as that for the nationals. To some extent the higher rate of unemployment of migrants could be explained by the concentration of migrants in labour intensive old heavy industries, but there has also been a strong element of discrimination against migrants when it came to the employment practices of Belgian firms. Tests were carried out in Brussels, Antwerp and Liege to cover both Flemish and Walloon parts of the country, using testers of Moroccan origin (with Belgian nationality). The results showed that one in three job vacancies were closed to them. It would be safe to assume that similar results would have been obtained if Belgian nationals of Turkish or other non-EU origins had been used. In turn, the situation of migrants without Belgian nationality is likely to be even worse.

What is the upshot of this for the welfare of second-generation migrants as they age? To the extent that public pension systems in the countries concerned link pension entitlements to years worked, contributions, and earnings, anybody with an interrupted and low-paid work history is going to be affected negatively. Whilst in Belgium, Germany, and Spain public pension systems are linked to work years, and contributions, as well as earnings, in the Netherlands only years of work are taken into account. Private pensions systems are ill suited to the needs of low-skilled immigrants on the account of their low pay and interrupted work histories.

Policy Considerations

In all EU countries, entitlement to a full state pension is related to one or a combination of the following: years of residence, years of insurance, employment, and contributions. In addition some countries require residency to claim or receive a pension.

One of the first policy requirements is to help migrants to increase their level of entitlement toward a full pension by supplementing years of residence with a credit for the years that the migrants were outside the country. Besides reducing poverty in old age, this policy adjustment could be viewed as compensation for the cost of upbringing (including education) of migrants, a cost that has not been borne by the host country. In some EU countries (e.g. UK), similar policies of social security credit already exist with regard to people who stay at home to perform care giving activities.

Furthermore, immigrants who do not have full entitlements because of their shorter period of contribution/insurance can be required to pay extra contributions with support from their employers. This may well have a disincentive effect on employing migrant workers and may encourage unregulated employment practices. However it is possible to reduce the cost to both parties by offering tax concessions. This would be in line with existing incentives to increase savings for old age. Such a policy would also help all those who have started their working life late, as well as a large proportion of women and those with an interrupted work history. In fact both equity and pragmatic political considerations should dictate that policy rules are designed in such a way as to be as inclusive as possible by not offering any special privileges to immigrants. The principle of 'what is good for the Goose should be good for the Gander' should be applied in all such matters! This principle makes policy formulation much easier by giving it maximum public support and legitimacy also more easily securing the successful implementation of any policy.

Residency rules on claiming and receiving pensions should be relaxed to facilitate the transfer of pensions abroad. Migrants may wish to return home or elsewhere on retirement for cultural and social reasons as well as for financial reasons to increase the purchasing power of their pensions. According to the available information Belgium, Denmark, the Netherlands, Norway, and Finland require residency to claim a pension, but thereafter the pension can be received abroad. In the UK, residency is not required for claiming or receiving a pension but if the pensioner were to choose to reside outside the UK the pension would no longer be eligible for increases according to adjustments to the consumer price index. The rules on index linking of benefit increases could sensibly be modified to allow for inflation in the country of residence. This should not pose a great problem as British missions abroad could provide such information, which is usually available from central banks, without much problem.

A further important issue that concerns immigrant families is the right of a surviving spouse to claim a pension based on the earned rights of his or her partner. This would affect women more than men given that in most couples men have been the main earner but it is women who have the longer life expectancy. In all the Member countries, the married spouse and, in some cases, the live-in partner, would qualify to claim a survivor's pension. Some countries also allow a divorced partner to claim a pension based on the rights of the deceased. However it is an important current requirement that marriage documentation is in order for such pension claims. In most countries, marriage documentation or evidence of living together is sufficient for making a claim. The policy issue in this area is not so much a need for changes to the law but one of making sure that immigrants are made fully aware of any legal requirements for marriage and partnership. A simple rule stipulating that people should register their partners with pension funds or offices in charge of state pensions could overcome this problem. The amount of benefit going to a surviving spouse varies between 40 to 80 per cent of the deceased's pension, in turn raising the question of whether the receipt of the deceased's pension would then ensure a reasonable living standard. To answer this question of income adequacy in old age goes beyond the scope of this paper and should be the subject of other studies on social security and poverty reduction.

References

Anthias, F. 1992. Ethnicity, Class, Gender and Migration: Greek-Cypriots in Britain. Aldershot: Avebury.

Barresi, C. and Stull, D. (eds.). 1993. Ethnic Elderly and Long Term Care. New York: Springer.

Baubock, R. (ed.). 1994. From Aliens to Citizens. Aldershot: Avery.

Bohning, W. R. 1972. The Migration of Workers in the United Kingdom and European Community. Oxford: OUP.

Castles, S. and Kosack, G. 1985. Immigrant workers and class structure in Western Europe. Oxford : Oxford University Press. (2nd ed.)

Castles, S. and Miller, M. J. 1993. The Age of Migration: International Population Movements in the Modern World. Basingstoke : Macmillan.

Castles, S. and Miller, M. J. 1998. The Age of Migration: International Population Movements in the Modern World. London: Macmillan. (2nd ed., rev. and updated.)

Clark, R., Kreps J. and Spengler J. 1978. "Economics of Ageing: A Survey", Journal of Economic Literature XVI (September): pp. 919-962.

Clark, R. and Spengler J. 1980. The Economics of Individual and Population Ageing. Cambridge: CUP.

Disney, R. 1996. Can We Afford to Grow Older? A Perspective on the Economics of Ageing. Cambridge, Mass.: MIT Press.

European Commission. 2000a. European Social Statistics: Demography. Eurostat.

European Commission. 1999. Social Protection for Dependency in Old Age in the 15 EU Member States and Norway. DG for Employment, Industrial Relations and Social Affairs (series on: Social and Security and Social Integration).

European Commission. 2000b. European Social Statistics: Migration. Eurostat.

European Commission. 2001a. Old Age in Europe. Directorate General for Employment and Social Affairs, Unit EMPl/E.2, June.

European Commission. (2001b). MISSOC (Mutual Information System on Social Protection) Social Protection in the EU Member States and the European Economic Area: Situation on 1 January 2001.

European Commission. 2001c. European Social Statistics: Social Protection Expenditure and Receipts 1980-1990. Eurostat. (2001 edition)

European Commission. 2002. Joint Report on Social Inclusion, DG for Employment and Social Affairs (series on: Social and Security and Social Integration).

Findlay, A. 1996. "Extra Union Migration: The South-North Perspective" in Rees, P., Stillwell, J., Convey, A. and Kupiszewski, M. (eds). 1996. Population Migration in the European Union, Chichester: J.Wiley.

Fenton, S. 2003. Ethnicity. Cambridge: Polity Press (key concepts series).

Guimezanes, N. 1995. "Acquisition of Nationality in OECD Countries", in OECD (ed.) Trends in International Migration: Annual Report. Paris: OECD.

Hurd, M. D. 1990. "Research on the Elderly: Economic Status, Retirement, and Consumption and Saving", Journal of Economic Literature XXVIII (June): pp. 565-637.

Johnson, P. and Falkingham, J. 1992. Ageing and Economic Welfare. London: Sage.

Klusche, W. (ed.). 1996. Die Herausforderung des Alters: Antworten der Sozialen Arbeit in Europa / The Challenge of Age: Social Work Responses in Europe. Report on the 2nd European Social Work Symposium.

Kostakopoulou, T. 2002. "Long term resident third country nationals in the European Union: normative expectations and institutional openings", Journal of Ethnic and Migration Studies, 28 (3): pp. 573-576.

OPCS. 1996. Ethnicity in the 1991 Census. Vol. 1. London: HMSO.

OECD. 2001. Trends in International Migration: SOPEMI 2000 Continuous Reporting System on Migration. Paris: OECD.

Rees, P., Stillwell, J., Convey, A. and Kupiszewski, M. (eds). 1996. Population Migration in the European Union, Chichester: J.Wiley.

Richmond, A. H. 1988. Immigration and Ethnic Conflict. London: Macmillan.

Rogers, A. and Serrow, W. J. (eds). 1998. Elderly Migration: an International Comparative Study. A final report to the National Institute on Aging. University of Colorado at Boulder. May. (Leeds Uni. Sociology K-8 ROG).

Rowland, D. T. 1991. Pioneers Again: Immigrants and Ageing in Australia. Canberra: AGPS.

Walker, A. and Maltby, T. 1997. Ageing in Europe. Buckingham: Open U.P.

Walker, A. Guillemard, A-M., and Alber, J. 1993. Older People in Europe: Social and Economic Policies. Brussels: CEC.

Zeger de Beijl, R. (ed.). 2000. Documenting Discrimination against Migrant Workers in the Labour Market. Geneva: ILO.

Mahmood Messkoub (1)

Economic Division, Leeds University Business School

University of Leeds, UK.

Notes

(1.) The first draft of this paper benefited from the comments of participants at the annual meeting of the Development Studies Association, Glasgow (10-12 September, 2003). All the remaining errors are mine.

(2.) See, for example, Walker and Maltby, 1997; Walker, Guillemard and Alber, 1993; Clark, et al., 1978; Clark and Spengler, 1980; Hurd, 1990; Johnson and Falkingham, 1992; and Disney, 1996.

(3.) Walker and Maltby, 1997: p. 4.

(4.) For further details see OPCS, 1996, Volume One, table 2.2: p. 49. For a concise and very useful theoretical discussion of sociological concept of ethnicity and race see Anthias, 1992, ch. 2. See also Barresi and Stull, 1993, and Rowland, 1991.

(5.) Foreign-born are those whose country of birth is different from their country of current nationality, while foreign population are those who live in a country but are a national of a different country.

(6.) The EEA consists of EU states plus the countries of the European Free Trade Association, i.e. Iceland, Lichtenstein and Norway. Note that Switzerland is neither in the EU nor in the EEA.

(7.) Among the EEA countries five EU members with small populations stand out as having the largest proportions of foreigners with European Union member country nationality: Luxembourg with 90.5%, Ireland with 75.2%, Belgium with 75.1%, Portugal with 53.7% and Sweden with 43.4%. A similar pattern emerges for intra-EU migration flows. What lies behind this pattern are linguistic and cultural affinities (e.g. Germans in Austria, Dutch and French in Belgium) as well as historical ties (e.g. Portuguese and Italians in France, and Italians in Austria) (OECD, 2001).

(8.) In the countries where a non-contributory pension scheme exists (e.g. UK) the level of pension is very low (just over half the contributory related pension) but residency is the only qualifying condition.

(9.) 'Claim' requirement is usually a less frequent procedure than 'receive'.

(10.) Proof of marriage is needed for immigrants who have married outside their country of residence or have not registered their marriage according the law of the country of their residence. A recent court case involving the Department of Social Security (DSS, UK) and a widow from the UK's Sikh community clearly revealed the problems facing ethnic women. Under the headline 'Sikh widow wins DSS court fight for pension,' The Guardian (23 October, 1999) reported that Mrs Kirpal Bath, after the death of her husband in 1994, applied to the DSS to draw her widow's pension. The DSS told her that there ware no record of her marriage and she was not therefore entitled to such a pension. The problem seems to have arisen because Mr and Mrs Kirpal married in a sikh temple whose ceremony was not recognised since, according to the DSS, 'the ceremony did not take place in a registered building', nor was the marriage 'registered in a registry office.' This is despite the fact that for 37 years the couple had been treated as married by the Inland Revenue and the DSS and to whom Mr Kirpal had paid taxes and social security contributions respectively. The judge ruled in favour of Mrs Bath by accepting that the couple were married at the temple even though this fact was not registered at that date. What is so remarkable about this 'unremarkable' news appearing in a national daily is the way in which lack of knowledge about the law on the part of the immigrants in 1956 was combined with a strict interpretation of the rules by bureaucrats, and then thoughtlessly utilized in 1994, to discriminate against a 54 year old widowed ethnic women.

Bibliographic Sketch

Mahmood Messkoub (m.messkoub@leeds.ac.uk) previously taught at Queen Mary (University of London) and the Institute of Social Studies (the Hague) and since 1989 has been a lecturer in the economics division of Leeds University Business School. His current research interests are in the areas of the economics of ageing and migration.