Work and the Diaspora: Locating Irish Workers in the British Labour Market

Irish migrant workers still make a significant contribution to the UK labour force, but this contribution is confined to particular occupation and industry groups. This paper begins with a brief review of the literature on Irish workers employment and an argument is developed that the work of Irish-born people in Britain is still both racialised and gendered. Then, using data from the UK Quarterly Labour Force Survey (QLFS), the work experiences of over one thousand Irish-born people in the UK are explored. The findings suggest that Irish-born men and women still work in the stereotyped occupations of the past. For example, most women work in public administration and health while twenty six per cent of men work in construction. The majority of Irish-born men work in manual skilled or unskilled jobs. The paper concludes that there has been no real qualitative change in the way that Irish-born workers experience employment in the UK.


Introduction
Put simply, we used to export people and import cash. Now, we can do it the other way round. Lemass's brilliant economic plans worked on the basis that Ireland needed access to capital, having already more than enough labour. Thirty years on, access to capital is guaranteed but labour shortages mean capital can't always be put to work. (The Irish Times, 29 November 1999) Ireland's recent economic success, coupled with a public discourse relating to the immigration of asylum seekers and economic migrants to Ireland has led to a real set of migration debates coming back to the fore in diaspora scholarship. At the heart of this debate is the historical tension between Ireland as a country whose citizens migrated to live and work elsewhere and the current needs of capital for a pool of available labour. However, despite the upturn in the Irish economy, many migrant Irish workers still provide their labour elsewhere. This is particularly true of the British economy, with an estimated 550,000 Irish Nationals (Owen 1995;Halpin 1997) still living and/or working in Britain. This paper aims to provide an update on the characteristics of the Irish in Britain, placing particular emphasis on employment and work issues .. Two issues distinguish this paper from those that precede it. First, the analysis and discussion offered is sociological and not exclusively historical or geographical. We acknowledge the difficulty in distinguishing aspects of sociology, history and geography, but the emphasis here is placed on debates that are essentially social in character, such as race and gender. We aim to try and understand the Irish experience of employment as being gendered and racialised, and as the product of longterm social processes. From this we can generate two main research questions: i) Has the Irish experience of work in Britain changed over time?
ii) To what extent is the Irish experience of work in Britain gendered and how does this compare to UK-born workers?
Second, the data is taken largely from the UK Labour Force Survey (LFS), 1987, the UK Quarterly Labour Force Survey (QLFS), 1997 and the Irish Labour Force Survey (Irish LFS), 1997. The LFS remains an under utilised data resource for exploring the Irish in Britain (notable exceptions include Hazelkom 1989and Halpin 1997.Instead, most research on the Irish in Britain has been carried out using the Census of Population. Although the Census is the most commonly used and largest data source, it is not without limitations. For example, it is only undertaken every ten years and, as such, subtle fluctuations in issues of labour market participation are missed. It could be argued that the Census only provides broad indicators of trends and not the specifics of that experience. Despite being smaller, the QLFS with its quarterly survey design and its detailed survey instruments can provide more current information on a wide range of topics. However, given these two aims it would be a mistake to think we are going to provide a grand theorisation of the Irish in Britain and, indeed this has been done successfully elsewhere (Mac Laughlin 1997;Buckley 1997). Instead, what we do provide is a 'statistical portrait' of the employment experiences of the Irish in Britain set against a brief conceptual framework.
The remainder of this paper is structured as follows. First we review the literature, which informs the diaspora debate, placing particular emphasis on work and employment related debates, and attempt to understand how Irish workers come to be located in certain occupations. The paper then moves on to an analysis of Labour Force Survey data. Finally we offer our conclusions.

Understanding Irish Workers in Britain: Evidence from the Literature
The high rates of emigration from Ireland which began in the mid-nineteenth century and persisted during the twentieth century, have meant that despite occasional periods of net gain Ireland has consistently lost population through emigration. Up until the Famine years of the 1840s Ireland, like the rest of Europe, had experienced rapid population growth. The Irish married young, a high proportion married and birth rates were high. As a result the population reached a peak in excess of 8 million. The dramatic decline in population which occurred during and after the 1840s can partly be explained by the catastrophic impact of the Famine which led to high rates of mortality and as a direct result of emigration. However, as Kennedy (1973) has argued, the Famine itself does not suffice as an explanation for the high rate of emigration which has characterised Ireland's population profile since. Indeed migration data suggest that paths of emigration to the US, Canada, Australia and Britain had been established prior to the 1840s and, as McDowell (1962) indicates, as early as 1841 there were already 400,000 Irish born living in British cities. This suggests that other factors operating within and outside Ireland also had a significant impact leading many to emigrate.
Breathnach and Jackson (1991) have argued that economic factors go someway to explaining these high levels of emigration. A lack of employment opportunities within Ireland has served to 'push' migrants out of the country seeking better economic opportunities overseas. Research on twentieth-century migration flows highlights the important role of purely economic factors in pushing men, in particular, to leave with lack of work in Ireland and low pay frequently cited as important push factors (Walter 1980). For women the push factors have differed somewhat, and often focused on an active rejection of rural life as well as a desire to maintain friendship networks by emigrating to join friends already settled overseas and to seek better career opportunities (Walter 1980;King and 0' Connor 1996). Whatever the reasons for leaving Ireland it is evident that the geographic and cultural proximity of the UK and the economic opportunities available have served as pull factors making Britain an important migrant destination where the Irish have long been the main source of unskilled migrant labour (Walter 1986:132).
The rapid extension of English industry could not have taken place if England had not possessed in the numerous and impoverished population of Ireland a reserve at command ... and from the time when it became known in Ireland that the east side of the Irish sea offered steady work and good pay for strong arms, every year brought armies of the Irish hither. (Engels 1987:123) Evidence from the first wave of Irish migration to Britain between 1840 to 1920 seems to support Engels' view with the Irish providing a source of labour in areas of early industrial development. The importance of their contribution during the nineteenth century is well-documented (Lawton 1959;Jackson 1963;Lees 1979;Swift 1992) with Breathnach and Jackson (1991: 3) describing the migrants during this period as 'a corps of unskilled labour' with 'little or no education' who took employment in the construction industry, manufacturing and personal service. Lawton's (1959) analysis of the 1861 census found that the majority of Irish migrants in Liverpool worked in dock-related employment, factories and domestic service. A further example of this process is the way in which Irish men came to be defined as labourers or more historically 'navvies' in the British labour market.
By 1871 anyone curious enough to travel through Britain in search of immigrants and refugees would have no difficulty in locating such groups ... An observer of immigrant minorities in Britain, taking stock in 1871, would have discovered the largest group carne from Ireland. (Holmes 1988: 20-1) Unlike the supposed pluralism of the US, many previous observers of Irish migration to Britain have pointed out that both male and female Irish emigrants were ghettoised into certain employment sectors. Women, for example, were overrepresented in areas of employment that can be regarded as an extension of their domestic role as wife/mother/carer or, in the twentieth century, nursing (King and O'Connor 1996;Walter 1991Walter , 1997. Irish men have also been concentrated in certain employment sectors, particularly construction, which Walter (1997: 61) identifies as an 'ethnic niche'. Jackson (1963) described the employment taken by Irish-born men as being 'heavy labour in the building industry, railway and road construction and maintenance .., jobs which often had little appeal to the British worker' (Jackson 1963: 97-8).
Male workers, driven away from Ireland by famine or other hardships provided their own labour and performed the hard, dirty and physical work required in the development of the British State (Goodwin 1999). Navvies emerged as the Industrial Revolution required men to build canals, docks and railways and, once invented, stayed on to provide labour for the 'great schemes of public work' (Sullivan 1983). In doing so they were characterised as hard working, drinking, rioting itinerant men.

Or as Buckley suggests:
Irish men labour to erect the infrastructure of Britishcities,moving on to vacate them for middle-class professionals -thedoctors, engineers, lawyers, lecturers andbusiness people, some of them Irish themselves -who will use them (Buckley 1997: 105).
The second main wave of migration, between the 1930s and 1970s, saw a large influx of Irish migrants to Britain with the Irish-born population almost doubling from 367,424 in 1931367,424 in to 726,121 in 1961367,424 in (King et at. 1989. During this time cities in the UK offered high levels of employment in a variety of industries and acted as a magnet to immigrants, particularly in the decades after the Second World War (King and O'Connor 1996). This period was characterised by emigration to areas of lighter manufacturing and the destinations changed to cities in the Midlands and the South East, with labour shortages offering a wider range of employment opportunities (Jackson 1963). King and Shuttleworth (1988) and Strachan (1991) indicate that the post-war flow changed the status of the Irish migrant from that of an underclass, taking the unskilled labouring jobs that the British did not want, to a much more integrated position in the British occupational and social structures (Strachan 1991: 22).
In this period the range of occupations available to Irish women also became mere diverse. A recruitment drive by the British government in the 1950s gave Irish women the opportunity to train for jobs previously unavailable, for example in the civil service, banks, social work and transport (Lennon et al. 1988). Walter (1986) suggests that in the 1960s, the arrival of migrants from the New Commonwealth and Pakistan (NCWP) provided a new supply oflabour in Britain, serving to some extent as a 'replacement population'. The resultant demographic change saw the Irish experience moving to reflect more closely general trends in employment. The new stream of Irish migrants headed for areas of Britain which were experiencing economic growth in the post-war period and they benefited from the role played by the already established Irish communities across Britain (Walter 1986).Amongst both male and female migrants the role of chain migration has always been important, with many relying on social networks on arrival. For example, relatives and friends already established in Britain often provided accommodation and information about employment opportunities to new arrivals (King and O'Connor 1996).
The 1970s saw a period of return migration to Ireland, which was experiencing an economic boom. This prosperity turned out to be short-lived and in the 1980s migration resumed, to destinations mainly in the South East, particularly London. The social profile of emigrants in this 'third wave' of movement has been described as differing from previous flows due to a higher percentage of highly qualified emigrants (Sexton 1987;Shuttleworth 1991;Hanlon 1997;MacLaughlin 1997).
However the flow continued to include the less skilled and Walter's (1991) analysis of the Census data between 1951 and 1981 found that the Irish-born men, in particular, continued to be over-represented in occupational sectors such as general labouring. At the same time the number of Irish-born men employed in professional occupations declined although for men as a whole employment in this sector increased. These findings are supported by King and Shuttleworth (1988) who found that in 1981, 26 per cent of Irish-born men worked in construction and as many as 64 per cent of immigrant men working in the construction industry were Irish. These findings are further supported by evidence from Hazelkom (1992) who also reports that in the 1980s Irish men in London were over-represented in sectors such as construction and metal manufacturing. For Irish-born women the picture was better and although they remained over-represented in certain sectors, overall their occupational patterns had begun to more closely reflect those of the population as a whole.
Data from the 1991 Census suggests that in more recent decades it remains difficult for the Irish-born to achieve upward mobility (Walter 1991). For women the picture is one of continuing concentration in certain traditional occupations, particularly low-level unskilled employment. The percentage employed in the personal service, clerical and health sectors is high, with twice as many Irish-born women than 'white' women as a whole working as health associate professionals (Owen 1995: 19). Irish-born men remained concentrated in four industries of distribution, engineering, transport and communications but with nearly one third working in construction (Owen 1995).
The late twentieth century again provides a slightly different picture. The era has been characterised by Ireland's economic 'miracle' (Sweeney 1998;O'Hearn 1998) and although Ireland was still experiencing net emigration during this economic boom period of the 1990s, net in-migration by the end of the decade had been forecast. Data from the Irish Central Statistics Office shows a positive figure for in-migration for the period 1991-1996, the first time since 1971-1979(CSO 1997. Sweeney (1998: 8) believes that the 1990s immigration flow is based 'on solid economic performance' which he contrasts with the experience of the 1970s return migration.
It has been suggested that the most recent arrivals are more highly educated and highly skilled than their predecessors (Shuttleworth 1991(Shuttleworth , 1997. However, it is argued by authors including Hanlon (1997) andMac Laughlin (1997) that this is in fact due to a change in perception, with recent Irish migration being described as a 'brain drain' of individuals leaving to seek better opportunities overseas. For example, Hanlon (1992: 183) suggests 'pre-1960s migrants are perceived to have been uneducated, unskilled rural labourers and in comparison, today's emigrants are highly educated, highly skilled, career orientated people'. Sweeney's (1998) description of the social profile of Irish emigrants in the 1990s in contrast to the earlier migrants provides a useful illustration of this generalisation. He writes that: These emigrants are far different from the 'Paddies' of the 1950s who looked for 'the start' in English building sites. Many are well educated. While 44 per cent left to find work abroad (most of whom had been out of work for less than a year), another onefifth left jobs in Ireland to take up employment overseas, 10 percent to complete their education. Around half of graduates who emigrate retum, with experience. (Sweeney 1998: 53) Swift (1992) suggests that the temptation to generalise should, however, be avoided because Irish migration streams have never been totally homogenous. Hazelkom (1992) highlights the fact that highly skilled and qualified migrants have always been part of the flow with the mid-nineteenth century stream including doctors and lawyers and other professionals (Hazelkom 1992;Swift 1992). This pattern has persisted throughout the twentieth century (Drudy 1986;Jackson 1986). Indeed, Hanlon (1992) estimates that middle class migration has always accounted for between 10 and 25 per cent of the total stream. Whilst there is no doubt that the migration during the third wave included a significant percentage of graduates and highly skilled emigrants, the more traditional flow consisting of low-skilled emigrants has persisted (Aspinall 2000), and the extent to which skilled migrants used their skills in 'appropriate' occupations is questionable.
From this brief overview it is possible to identify a number of work and employment related trends emerging. First, the accounts seem to characterise Irish workers doing heavy, hard and tedious work, despite some migrants having some level of education. Second, labour migration is a gendered process in which Irish men and women undertake work in traditionally stereotypical masculine or feminine occupations. Finally, and most interestingly in terms of this paper, despite some slight variations, the employment patterns of the Irish in Britain appear to have an underlying consistency in terms of occupation and industry. It appears that over time Irish workers in Britain have been mainly located in a small range of industries such as construction, manufacturing, dock-work, domestic labour,clerical and caring work. The occupational experience of the Irish in Britain has for the most part been in unskilled or semi-skilled, manual or labouring work. Given the fluctuations in the British labour market, how do Irish workers in Britain come to be located in similar occupations historically?
At the broadest level one can say that the experience of Irish workers in Britain has, like that of other ethnic groups, been both racialised and gendered. As discussed earlier, the Irish have been attributed the role of a reserve army of labour (Jackson 1963;Miles 1982, Walter, 1986, Hazelkorn 1991 and their labour acted as a crucial component of the capitalist industrial development of Britain. By virtue of being part of this reserve army Irish migrants, both skilled and unskilled, are subordinated at the level of class and as members of a migrant minority group. Further to this, the experience is also gendered in the sense that Irish men and women do very different jobs that conform to traditional notions of masculinity and femininity. As Buckley (1997: 109) suggests Overviewing Irish workers in Britain, we see then rigorously channelled in to hypertrophied gender-stereotypes, with millions of Irish women intensely engaged in the feeding, cleaning, healing, caring and teaching of Britons and millions of Irish men focussed into clearing, constructing and fabricating the economic landscape of contemporary Britain.
The Irish have also suffered anti-Irish racism (Walter 1984;Castles and Kosack 1985;Jackson 1986) and, in the early periods of migration, anti-Catholicism (Busteed et at. 1992). In the twentieth century anti-Irish racism in Britain was a constantly underlying factor tending to surface at times when political differences between the countries were in the news. For example, during the 1920s, after independence, anti-Irish feelings were strong in Britain and the Irish who had immigrated during that period faced hostility and overt racism. During and after the Second World War anti-Irish sentiments came to the fore again and since the political situation in the North has become more contentious, the Irish in Britain have faced further hostility. Castles and Kosack (1985: 446) argue that 'the Irish are more or less accepted today, but only a few decades ago they were the victims of widespread discrimination' .
However, although the sentiments of Castles and Kosack (1985) may be true to some extent, it could still be argued that aspects of anti-Catholicism still underpins anti-Irish sentiments (Clayton 1996;Hickman 1995). Indeed, logic of this approach is clear with many individuals both now and in the past confusing the terms 'Irish' and 'Catholic' or using them interchangeably (Boyce 1990;Fields 1997). To understand this one has to look back at the past and to understand the processes that have taken place over time. There are two useful works that do this in different contexts. Brewer and Higgins (1998), in a discussion of Northern Ireland, suggest that anti-Catholicism is a social process which achieves the 'production of different rights, opportunities and material rewards between people in a society where religious labels are used to define group boundaries' (Brewer and Higgins 1998: 12). Hickman (1995: 22) provides a similar a argument, namely that both anti-Irish racism and anti-Catholicism have been integral to the formation of the British national identity, are consistent elements of British culture and have shaped government and institutional practices regarding the Irish in Britain. During the nineteenth century this British national identity emerged, the opposite of which was the Irish Catholic migrant (Hickman 1995: 19). 'The Irish Catholic other was significant because it intertwined both a discourse of inferiorized colonial subject, and that of the other against which English nationalism has pitted itself since the sixteenth century: Roman Catholic' (Hickman 1995:21).
According to Hickman (1995) elements of anti-Irish anti-Catholicism exist through time starting from the very point in time when the British and Irish interacted. For example, when talking of the Tudor period she writes: During this Tudor period the image of the Irish which was widely disseminated was of a 'wild' unciviIised and maybe uncivilizable people who practised a regressive way of life, were pagan in their religion, licentious in their sexual mores, treacherous and savagely violent ... (Hickman 1995: 25).
However, moving forward in time to the late nineteenth century Hickman (1995: 21) reports that 'the poverty and type of manual labour which their lives required the Irish in Britain to undertake became proof of their savagery'. These sentiments have emerged over time and can still be found in residual and altered forms in the modem era (Brewer and Higgins 1998).
The works of Brewer and Higgins (1995) and Hickman (1995) provide a useful starting point in helping us understand how anti-Catholicism and anti-Irish sentiments emerge. The value of Hickman's (1995) work is not so much that it considers the Irish as a race per se, but more that it offers a consideration of how the Irish have become defined in particular ways in Britain via long term processes and interactions. The notion of 'process' is the tool through which we can explore and understand change and how a series of interrelationships develop over time. Inthis sense anti-Catholicism develops via the historical interactions between Britain and Ireland, between monarchy, republicanism, Catholicism, Protestantism, Gaelic and Anglo-Saxon world views, and between industrial and rural. From the work of Hickman (1995) the consequence of these developing interrelationships is that early notions of race, Irishness, Britishness and all that they entail get intertwined with other issues and perceptions such as notions of criminality, sobriety or even the relative merits of Irish 'workers' and the role that they can play in a labour market. Elements of these interrelationships go on to inform current practice and identities, with the Irish diasporic identity being constructed by others around particular occupations and industries. However, to what extent is this still the case and what is the situation with Irish employment more recently? The paper moves on to a discussion of the data sets used and the process of identifying the Irish in Britain at the end of the 1990s.

Data Source and Method: Researching Irish Workers in Britain
Statistics cannot properly recreate the variations of individual experience. But this material forms the basic starting point for coming to grips with the diaspora. It also represents an attempt of modem scholarship to provide a fuller description of Irish communities. (O'Day 1996: 196) This paper is based on an analysis of data collected as part of the 1997 QLFS. This quarterly survey, with a random sample of over 63,000 households, is carried out by the Office for National Statistics (ONS) in Great Britain and by the Central Department of Finance and Personnel in Northern Ireland. At the individual level the LFS contains data on a total of 150,000 people (Felstead et al. 2000;HMSO 1997).
To supplement the QLFS two other sources of data were used. First, data on the Irish-born from the 1987 UK LFS allows us to look back over a ten-year period to explore whether the composition and characteristics of the Irish-born workers have changed. Second, selected data derived from the 1997 Irish LFS is also included allowing some comparison between Irish-born workers in Britain and workers in Ireland.
As with recent work using 1991 Census data (see Owen 1995) it was decided to use 'Irish-born' as a category for identifying the Irish in Britain. Using a combination of the QLFS variables CRY and CRYOX, questions designed to identify the country of birth, it was possible to identify a total of 1,362 individuals. However, one of the problems with using 'Irish-born' as a selection category is that the LFS is a UK wide survey and contains respondents from Northern Ireland. Furthermore, as the CRY variable contains values for both the UK and Ireland, the interviewers are advised to check whether it is Northern Ireland or the Irish Republic if respondent answers 'Ireland'. This means that even though a respondent could have been born in Northern Ireland but identify themselves as being born in Ireland, their country of birth would be recorded as UK. Obviously, such an approach raises a range of concerns regarding the interplay between country of birth, ethnicity and cultural identity and the fact that most large-scale surveys cannot deal with this complexity makes any discussion of an 'Irish-born sample' problematic culturally and politically (Owen 1995). As such, there was some unease at using these variables in isolation. The focus here was on Irish-born in the British and not the UK labour market so it was decided that place of residence and place of work would be used to refine the sample. Using the QLFS variable URESMC (region of usual residence), it became evident that 115 of the 1,362 respondents who identified themselves as being Irish-born lived in Northern Ireland. Using the variable REGWK (region of place of work) of the 115, it was found that 35.8 per cent also worked in Northern Ireland. The location of work for those remaining could not be determined or was not recorded. Itwas, therefore, decided to exclude those who identified themselves as Irish-born but who lived in Northern Ireland leaving a workable sample of 1,247.
To allow comparisons a similar identification process took place in the 1987 LFS data set. Using the variables COUNTRY (country of birth) and URESCOMJ (region of residence), and after excluding those Irish-born living in Northern Ireland, this gave an additional sample of 1,482. Table 1 presents the basic characteristics of the 1987 and 1997 LFS sample including marital status, age, year of arrival and ethnicity. In 1987 the total sample numbered 1,482 and of these, 797 or 53.8 per cent, were women and 685 or 46.2 per cent were men. In 1997, of the 1,247 total sample, 556 or 44.5 per cent were male and 691 or 55.5 per cent were female. Similar patterns of a high female to male ratio have been identified from other recent data sets (Owen 1995;Halpin 1997). This reflects Irish migration statistics over time, which have demonstrated that women have outnumbered men during most periods of emigration since the nineteenth century (King and O'Connor 1996).
The majority of both samples were aged between 30 and 60, although the 61 and over category is the largest of these. This category combines the older age categories (e.g. 61 and over, 71 and over, etc.) and as such is likely to include proportionally more people than the 10 year categories used for the other adult 'under 61' age range. However, the size of this category does also reflect the high numbers of Irish emigrants who arrived in Britain during the years immediately after World War II. Many of this group of migrants have remained in the UK since first arriving decades earlier. It appears that there is little chance of return migration for this group because although clearly still Irish, the cities where they settled have become home and perhaps more importantly many now have children and grandchildren in the UK who they would find hard to leave behind (King and Shuttleworth 1988). The size of the oldest age group increased from 29.4 per cent in 1987 to 38.7 per cent in 1997 and the average age increased from 49.6 to 53.5 years old. The Irish-born are on average 10 years older than the population as a whole; in contrast there are very few Irish-born in the younger age cohorts. The 1991 Census data also identifies the Irish as an elderly population with the largest cohorts in the 40 and over categories and very few children (Owen 1995).
As with the Census of population it is possible to determine the year of arrival. Table I illustrates the flow of migration to Britain since the start of the twentieth century. Between 1900 and 1997 it is possible to identify a number of 'peaks and troughs' in Irish migration. The peak migration years according to the LFS data were 1941-70 and 1981-90, with 1951-60 being the single highest period of arrival. This pattern reflects the general trends in Irish migration as discussed earlier, confirming that this population is an ageing one. When the data for 1987 and 1997 is compared it becomes evident that the numbers who arrived in the period between 1900-40 have decreased significantly between the period 1987-97, a pattern which fits with the ageing profile of this population. The percentage of those who arrived in every ten year period has decreased between 1987 and 1997, which in the later periods may be due to return migration in the 1970s and perhaps again in the later 1990s. Although not significant in a statistical sense, it is clear from the data that there is some gendered element to migration, with female migration being slightly higher than that of males in the 1940s and the 1980s but with male migration being much higher in the 1950s when Ireland experienced a huge decline in employment in the agricultural sector (Walter 1991:11).
The variable of marital status shows that the majority of the sample in both 1987 and 1997 were married; 62.6 per cent in 1987 and 56.1 per cent in 1997. The decrease in the percentage of married people is due, in part, to the increased size of the widowed category. Both the separated and single categories also increased over the period. In both 1987 and 1997 there were more single Irish-born men in Britain than there were single Irish-born women. Overall the single category is the second most important, becoming more important over the 100year period. This reflects the historic patterns of migration streams, which have always comprised of a high percentage of young, single adults (Kennedy 1973). The number of widows and widowers is also high for both samples and is higher in 1997, a statistic which again fits with the pattern of the Irish being an ageing population. The most striking figure here is the very high percentage of widows in 1987, 18.2 per cent, which by 1997 drops to 9.8 per cent, reflecting the age of the women.
The variable ETHCEN (ethnic origin, census of population definition) showed the vast majority of the Irish-born respondents were recorded as white in both 1987 and 1997. Indeed, in 1987 there were only eight people recorded in an ethnic group apart from 'white', and in 1997 there were only two. These findings are in line with those presented elsewhere (Owen 1995) that the Irish in Britain are almost exclusively white.
Given the geographical nature of the Census of population, it has been possible for many commentators to identify the patterns of Irish-born migration to the UK on a county and regional basis (Walter 1980(Walter , 1986Strachan 1991;King and O'Connor 1996). Table 2 presents data on the region of usual residence in 1987 and 1997. This shows the largest concentrations of Irish-born for both periods are located in South East England (21.6 per cent and 20.2 per cent), outer London (16.5 per cent and 19.5 per cent) and inner London (14.2 per cent and 12.4 per cent). When 1987 and 1997 are compared the most important regions of London, the South East, the West Midlands and Greater Manchester maintain their position with little change over time. Some regions of Britain show a decreased concentration over the period such as the East Midlands, parts of Yorkshire and Humberside and parts of the North West. One explanation for this maybe the decline of the manufacturing industry; certainly regions which were previously 'magnets' for migrants due to the heavy industry located there, have, in recent decades, experienced a steep decline in employment opportunities. However, this does not help to explain why certain other regions which may also have been expected to decrease in fact showed a small increase, for example, Tyne and Wear, South Yorkshire, Wales and Strathclyde/Clydeside. These findings compare favourably with Census data which identified similar geographic patterns. Owen (1995), for example, found concentrations of the Irishborn in Greater London (36.2 per cent) and in the South East (54.6 per cent) and Halpin (1997: 35) concluded from his analysis of the 1994 LFS data that 'for London the Irish are substantially over-represented'. When residential distributions outside the South East are analysed similarities are also evident. For example, the 1991 census showed the West Midlands as a whole as the next most important region with a concentration of 11.4 per cent (11.7 per cent in the 1997 LFS) followed by Greater Manchester 6.4 per cent (6.3 in the 1997 LFS), the East Midlands 4.7 per cent (4.7 per cent in LFS) and Merseyside 1.9 per cent (2.3 per cent in the LFS). Such similarities in these figures suggest that, despite the relative small sample size of the LFS, it contains a fairly representative sample of the lrishborn in Britain. Britain, 1987Britain, -1997 Location In the remainder of the paper, data is presented on the employment characteristics of the Irish-born in Britain. In order to offer some comparative data and to explore more fully how the Irish fare in the British labour market, the data for Irish-born workers is compared to data for the UK-born population taken from the 1987 and 1997 LFS. Some comparison is also made with the Irish LFS of 1997. Table 3 presents the data on economic activity for Irish-born and UK-born populations in 1987 and 1997 and in Ireland in 1997.

The Irish Experience of Work in Britain: Evidence from the Labour Force Survey
The 1987 LFS data shows 48.3 per cent of the Irish-born in employment, 55.2 per cent of men and 42.1 per cent of women. By 1997 the percentage of the Irishborn in employment had fallen to 45.1 per cent, a figure slightly higher than for the UK-born population (44.7 per cent). The 1997 data shows more Irish-born men than UK-born men in employment (51.3 and 49.5 respectively) and fractionally more Irish-born women (40.2 per cent) than UK-born women (40.1 per cent). Halpin (1997) identified similar patterns with 52.2 per cent of Irish-born men in employment and 42.3 per cent of women. Slightly fewer Irish-born men than UKborn men are employees, 39.9 per cent and 40.7 per cent respectively, whereas slightly more Irish-born women (37.2 per cent) than UK-born women are employees (37.0 per cent).
A difference between the UK and Irish-born groups emerges in the self-employed category which includes a high percentage of Irish-born men, 11.2 per cent in 1987 and 12.6percent in 1997 as compared to only 8.3 per cent of UK-born males in 1997. Irish-born men are much more likely to be self-employed than UK-born men and both UK and Irish-born women. This high rate of male self-employment is also found in the 1991 Census data and as Owen (1995) has explained it may be attributable to the fact that many construction workers are classified as self-employed. Such a definition, however, is problematic as such workers do not have the independence or autonomy of the 'own account self-employed' and are, as such employees with no control over the price of their labour (see Burchell and Rubery 1992).
A comparison of the Irish in Britain data for 1997 and the Irish LFS findings of the same year show a slightly higher percentage in employment in Ireland (49.1 per cent) than in the UK. However, almost equal percentages are employees, 38.4 in the UK and 37.7 per cent in Ireland. Analysis by gender shows that of those in employment 60.8 per cent are males (only 51.3 amongst the Irish in Britain) and 37.7 per cent females (40.2 per cent for Irish-born in Britain).
The self-employment category reveals similarities between the two national data sets. In Ireland there are more self-employed men than women (11.8 per cent and 1.7 per cent) reflecting the pattern for Irish-born in the UK. The number of selfemployed men in Ireland is slightly lower than it is for Irish-born men in the UK in 1997 but higher than the comparable figure for UK-born men in the same year. The most striking figure is the percentage of Irish-born who are classified as being inactive.' This category is defined as including individuals who would like to work but are 'unavailable', for reasons such as long-term sickness or disability. The category also includes those who are not working and do not want work, for example students. This is the most important category numerically, including 41.6 per cent in 1987 and 48.6 per cent in 1997. The inactive category is also important in the Irish LFS, some 45.3 per cent. The comparable figure for the UK-born group in 1997 is only 28.6 per cent. In all three data sets there are more women than men classified as inactive. This high rate of inactivity amongst Irish-born reflects to some extent the elderly profile of the group. Indeed further analysis of the data shows a high proportion of retired individuals amongst the Irish-born, almost twice as high as the UK figure. It could also be attributed to the fact that the Irishborn in Britain suffer poorer health than the UK population as a whole (Greenslade 1997) evidence of which is shown in Table 3. Whilst this is a broad comparison, not based upon other factors such as qualifications held and so on, it does nonetheless provide illuminating data about the health of this ethnic group. Other smaller scale and more detailed analyses which have compared the Irish born with other ethnic groups and accounted for other factors have reached similar conclusions (Balarajan and Bulusu 1990; Raferty et at. 1990). The increase in the percentage inactive between 1987 and 1997 is, again, attributable to the ageing population.
The 1997 LFS data finds the Irish-born more likely to be unemployed than the UK-born with an overall figure of 4.4 per cent as compared to 3.3 per cent. When this is broken down by gender Irish-born men are the worst affected, with an unemployment rate of 5.0 per cent compared to 4.2 per cent for UK-born men and 3.9 per cent for Irish women. On a more positive note, in the period 1987-97 the unemployment figures for Irish men in Britain were halved. The Census data for 1991 showed an unemployment rate for Irish-born men some 50 per cent higher than the white average whereas for Irish-born women the figure was only slightly higher than the average (Owen 1995). Unemployment rates in Ireland in 1997 at 5.6 per cent overall, 7.0 per cent male and 4.4 per cent female, were higher than those for the Irish-born in the UK in 1997. Table 4 shows the industrial sector of employees and self-employed in the 1987 and 1997 LFS and the 1997 Irish LFS and although the data is not fully comparable due to differences in classification, it does suggest some interesting findings.
In 1987 the 'other services' category was the most significant, accounting for 31.2 per cent overall. This category includes the health sector, education and cleaning services amongst others and was the most important sector for women, including almost half of them (49.4 per cent). For men, construction with 29.1 per cent was by far the most significant category in 1987.
In 1997 the most significant category for the Irish-born was 'health and social work' accounting for 17.9 per cent overall. When this is disaggregated by gender the picture is more telling, some 33.3 per cent are female and only 3.2 per cent male. This is the most important category for women, whereas for Irish-born men, construction, the second most important category overall, is by far the most significant industry providing employment for 26.4 per cent in total, only slightly lower than in 1987. Other important categories overall are manufacturing, wholesale and retail, real estate and business services. For women after health the most important category is retail, followed by real estate, manufacturing and education. For the UK-born the most important category overall and specifically for men, is manufacturing. Construction is the third most important sector for this group, accounting for 11.9 per cent, far lower than the Irish figure. For UK-born women health and social work is also the most important sector, accounting for 20.2 per cent, but it remains far more significant for the Irish-born women.
The Irish LFS reveals very different concentrations. Discounting the 'other' category the main industry of employment for Irish women is in the finance sector (24.3 per cent) followed by manufacturing (16.5 per cent). For men this pattern is inverted and the most important category is manufacturing (22.5 per cent) followed by finance (18.7 per cent). Agriculture, forestry and fishery is the third most important (14.8 per cent) and construction ranks fourth with 11.1 percent. The 1997 occupational data presented in Table 5 shows similar distributions of UK and Irish-born by occupation.' For example, comparable numbers of Irish and UK-born work as mangers and administrators, 15.5 per cent and 15.8 per cent respectively. The percentage working as professionals is also similar. The associate professional category reveals the most striking difference, including a much higher percentage of Irish-born women than UK-born women; 19.2 per cent as compared to 11.2 per cent. This classification includes nurses, accounting for 12.7 per cent of this category for Irish-born women, by far the most significant occupation for Irish women within this classification.
The 1987 data for social class presented in Table 5 shows the most important category for the Irish-born overall was towards the bottom of the table. The skilled manual category accounted for over a quarter of the sample. The next most important classification was the intermediate sector but there was little difference between this and the two following categories: partly skilled and non-manual, both at the lower end of the table. The highest category of 'professional' accounted for only 3.1 per cent whilst in contrast the lowest classification of unskilled included three times as many as this (9.9 per cent). Disaggregated by gender the data shows more than a third of Irish-born men in the skilled manual category, followed in importance by intermediate (19.8) and partly skilled (15.8). The unskilled accounted for 9.1 per cent of men. At the other end of the spectrum only 5.1 per cent of men were in the professional category. For women the picture is slightly different. For them the most significant category was partly skilled (27.5 per cent) followed by skilled non-manual (26.6 per cent). There were more unskilled women than men, 10.8 per cent as compared to 9.1 per cent and fewer women than men, less than one per cent, in the professional category. By 1997 the overall picture had improved for the Irish-born with the intermediate class replacing skilled manual as the most significant category for the Irish-born, mirroring the pattern for the UK-born.  Disaggregated by gender this category remained the most significant for Irish-born men and women. In comparison for UK-born men, the skilled manual category was slightly more significant and for UK-born women skilled non-manual is much more important. An examination of the lower end of the table shows that the unskilled category, is more significant amongst the Irish than UK-born. Some 10.2 per cent of Irish-born men, compared to 4.1 per cent UK-born fall in to this category whilst for the Irish women the figure is 8.3 per cent and for UK women the figure is 5.9 per cent. Between 1987 and 1997 the Irish-born became better represented in the higher social classifications of professional and intermediate but at the same time the number in the unskilled category only improved fractionally and indeed for Irish men this increased. Table 6 shows the percentages working either full-time or part-time in 1987and in Ireland in 1997 per cent of the Irish-born were working full-time, 96 per cent of males and 49.6 per cent of females. By 1997 the overall percentage of Irish-born working full-time had increased to 77.7 per cent compared to 72.5 per cent of UK-born. In Ireland a higher percentage, 87.6 per cent were in full time employment.
Although between 1987 and 1997 the number of Irish-born men working fulltime had slightly decreased, the percentage ofIrish-bom women employed on this basis had increased by just over 10 per cent. There are more Irish-born women working full time (60.1 per cent) than UK-born (53.2 per cent) and more Irishborn men (94.7) than UK-born men (90.5).
The Irish LFS shows a high overall figure working full-time; 87.6 per cent. A gender analysis shows 89.1 per cent of males and 76.9 per cent of females working full-time. This figure for women is much higher than is found amongst UK-born or Irish-born women in the UK LFS, perhaps reflecting the recent increase in employment opportunities for women in Ireland (Breathnach and Jackson 1991).
Turning to part-time employment 27.5 per cent of UK-born work part-time and 22.3 per cent Irish-born. Only 5.3 per cent of Irish-born men work parttime whereas for the UK-born this figure is 9.5 per cent. Owen's (1995) analysis of the 1991 Census had a similar finding, revealing that four per cent of Irish-born men worked part time, slightly lower than the average for the population as a whole. Slightly fewer Irish-born women than UK-born work part-time which is surprising given the tendency for Irish-women to move to part-time employment later in life (Walter 1991).
The 1997 Irish LFS shows only a small proportion of women working parttime, however, they are more likely to be in part-time employment than their male counterparts. For Irish men the number in part-time employment is more than double the percentage found in the other datasets. The data for tenure, hours of work and highest qualifications held by employees and self-employed are displayed in Table 7. Taking tenure first, the data shows that the Irish-born stay in the same job for longer than the UK-born population, with an average length of tenure of 105.8 months as compared to 97.5 months. For women this pattern is  LFS 1987, QLFS Summer 1997, ILFS 1997 more marked with Irish-born women holding the same job for an average of 93.5 months as compared to UK-born women at 81.7 months. For men the difference is smaller, although Irish-born men with a tenure of 116.5 months tend to hold their jobs longer than the UK-born men at 111.1 months on average. The 'hours of work' data reveals that the Irish-born tend to work longer hours, 36.0 on average as compared to 35.5 for the UK-born. Irish-born women and men have an average usual working week of just over 1.5 hours longer than the UKborn. Owen (1995) identified the same trend from the 1991 Census. Data from the Irish LFS shows that within Ireland the working week is even longer, 40.8 hours overall, 44.8 for men and 35.0 for women.
Focusing on full-time employment only, the data shows that Irish-born men work a slightly longer week than the UK-born while Irish women work a slightly shorter week than their UK counterparts. Part-time employment shows a bigger differential, with Irish part-time workers having a considerably longer working week, 21.6 hours as compared to 18.6 hours. For Irish-born men in part time work the hours worked are 24 while UK-born men work only 17.4 hours. For women the part time hours are more closely matched but still find the Irish-born women working an average of almost 4 hours more per week. The data on educational qualifications held by employees and self-employed shows that for both Irish and UK-born the most significant group are those who attained secondary level qualifications, such as '0' levels and NVQ level 1. Amongst the Irish-born 37.2 per cent achieved this level and 36 per cent of the UK-born. The next most important category for the Irish-born is those with no qualifications, just under 25 per cent of the sample compared to only 14.9 per cent for the UK. At the top end of the scale the pattern also favours the UK-born with 14.3 per cent holding degrees or equivalent as compared to 11.3 per cent for the Irish-born. An interesting finding from the table is the high percentage of Irish women holding other Higher Education qualifications. Some 15.6 per cent of Irish women hold this level of qualification as compared to only 4.2 per cent of Irish men, 8.0 per cent of UK men and 10.8 per cent of UK women. This is likely to reflect the high number of Irish women who entered teaching and nursing professions in the UK. The Irish fare less well when post-secondary qualifications are considered. Whereas 25.4 per cent of UK-born hold some post secondary qualification the comparable figure for the Irish-born is only 16.8 per cent. In both the Irish and UK-born categories almost twice as many men than women hold post -secondary qualifications. Halpin (1997) found fewer Irish-born than UK-born graduates in the 1994 LFS sample. He also found less than 4 per cent of those in the Irish-born over 40 age group holding degrees which was half the percentage for UK-born (8.2 per cent). However, amongst the younger age group, between 23 and 30, he found a quarter of Irish-born with degrees as compared to only 14.6 per cent of UK-born. The 1997 LFS data found similar patterns. As age increases the likelihood of holding a degree decreases and in the oldest age category only 6.3 per cent have attained this level of education. However, although the percentage holding degrees is higher amongst the younger age groups, showing signs that the Irish migrants are indeed becoming better educated, there are still more 21-30 year olds in the non-qualified category (16.2 per cent) than those with degrees (14.4 per cent). The same is true for 31-40 year olds of whom 20.1 per cent hold no qualifications. In the older age group the percentage holding no qualifications at all is high, at 63.2 per cent.

Conclusion
This paper began with the suggestion that the work experiences of the Irish in Britain are both racialised and gendered, and that this is part of a historical process through which Irish 'otherness' has been constructed and judgements have been made about the'value' of Irish labour. In order to explore more recent experiences, two main research questions emerged. First, has the Irish experience of work in Britain changed over time? Second, to what extent is the Irish experience of work in Britain gendered and how does this compare to UK-born workers? In answer to these questions a number of conclusions can be drawn.
First, the Irish experience of work in Britain has changed slightly overtime, but that such change is underpinned by consistency. For example, the social class data reveals that there has been some upward movement for the Irish-born between 1987 and 1997, particularly evident in the 'intermediate' classification. However, whilst there have been some changes over the 1o-year LFS period the more striking finding is the lack of change since 1987 and indeed since Irish migration on a large scale began. The Irish remain concentrated in certain geographical regions and segregated in certain industrial sectors. Whilst there has been much attention given to the new wave of highly qualified Irish migrants the data persistently reveals a significant flow of unskilled migrants. The resultant patterns mean that the Irish remain over-represented in the lower social class classifications. The data on edu-cational qualifications finds the Irish over represented in the unqualified category and conversely, under-represented at the top end of the scale. Despite some younger migrants holding degree level qualifications, there are still significant numbers who have only secondary qualifications or none at all. Occupational data also confirmed that Irish-born men are persistently over-represented in the construction industry and women in nursing, patterns which have continued throughout the last one hundred and fifty years.
Second, the Irish experience of work is gendered and when comparing Irishborn workers with UK-born workers, it appears the Irish men suffer the worst fate in the UK; they work longer hours, they are more likely to suffer unemployment, they have a higher rate of poor health and are over-represented in the lower social classes. The unemployment data shows that Irish men are more likely to be unemployed than Irish women and than the UK-born population in general. Linked to this finding is the high rate of inactivity amongst the Irish-born. This is perhaps due to the elderly profile of the sample and the higher than average possibility of having a limiting long-term illness (Greenslade, 1997).
For women, the 1997 Irish LFS revealed that they are more likely to be in employment than women in the UK and far more likely to work full-time than those who live and work in Britain. This reflects to some extent the improved employment opportunities for women in Ireland (Breathnach and Jackson 1991) and also the older average age of the population in the UK. Between 1987 and 1997 the number of Irish-born women working full-time has increased but the percentage of women in Ireland who work full-time is far higher. The Irish-born women displayed a tendency to hold the same job for longer, perhaps suggesting that the Irish find it more difficult to move out of particular sectors.
What we have presented here may be nothing new and may not come as a great surprise to anyone involved in Irish diaspora scholarship. However, what is a concern is that these trends of gender stereotypical employment, and the different experiences of employment to a UK-born population still continue. In particular their is a continued need to highlight the experiences of Irish men in Britain as their experience seems to be particularly problematic (see also Aspinall 2000;Buckley 1997). The next large-scale opportunity to explore these debates for the Irish in Britain will be the 2001 Census of Population. Hopefully analysis will reveal that all of the Irish in Britain are faring better than before. Notes 1. We would like to thank SusanWalker for her valuable suggestions, Jim McAuley for his supportand the anonymous referees for theirhelpful comments. 2. Formoreinformation on thiscategory see theLFSUserGuide (1997) Volume 3:Details of LFS variables, p.61. 3. It was difficult to provide comparable data from 1987 and from the IrishLFS 1997 as vastly different categories wereused.