Social networks and job seeking

Recent discussions of the concept of employability have emphasised the value of social capital (in the form of access to networks) in facilitating individuals’ progress in the labour market (McQuaid and Lindsay, Reference McQuaid and Lindsay2005). For Fugate et al. (Reference Fugate, Kinicki and Ashforth2004: 24): ‘Social capital is the goodwill inherent in social networks. [It] contributes an overtly social element to employability, and it confers information and influence to the ‘holder’ via networks.’

There is evidence that social networks can play an important role in transitions to work. Hannan (Reference Hannan1999), using British Household Panel Survey data, demonstrates that that while unemployment duration, the use of formal search facilities and other factors impact upon exits from unemployment, access to social networks significantly affects job entry rates. Levesque and White's (Reference Lévesque and White2001) research with long-term welfare recipients in Canada supports this position – they argue that the range and diversity of social network resources available to the long-term unemployed can be the defining factor determining successful transitions to work, with a broad range of weak social network ties (i.e. acquaintances connected with the world of work rather than close friends or family) ‘validating’ the human capital of job seekers by providing access to references and good-quality vacancy information.

Earlier, it was Granovetter's (Reference Granovetter1973, Reference Granovetter, Marsden and Lin1982) seminal research on the job search strategies deployed by higher-skilled workers that highlighted the potentially crucial role of social networks. Granovetter noted the importance of the ‘strength of ties’ to job seeking, arguing that ‘weak ties’ (acquaintances, colleagues) provide more non-redundant information than do ‘strong ties’ (family and close friends). Research conducted since Granovetter introduced the idea of the importance of the strength of weak ties has similarly concluded that that the range and diversity of job seekers’ networks affect their ability to locate employment (see Lai et al., Reference Lai, Lin and Leung1998 for a review of evidence).

There are counter-arguments to the strength of ties hypothesis. Mouw (Reference Mouw2003: 868) has suggested that no causal link can be identified between high levels of social network resources and job outcomes, instead arguing that the similarity between the status of individuals and their social ties is likely to explain any ‘social capital effect’ – i.e. ‘there is a tendency for similar people to become friends rather than a causal effect of friends’ characteristics on labour market outcomes’. The homophily of social networks has similarly been acknowledged by other studies as representing an important methodological challenge to any research seeking to identify distinct causal relationships between networking and socio-economic status (McPherson et al., Reference McPherson, Smith-Lovin and Cook2001).

Nevertheless, there remains clear evidence that effective social networking can facilitate transitions to (and within) work, and so at least reinforce individuals’ labour market status and employability. People with access to diverse, work-based networks are generally more likely to locate a ‘resource-rich’ contact person, with greater potential to positively impact on job search outcomes (Lai et al., Reference Lai, Lin and Leung1998). Network diversity and the heterogeneity of contacts can therefore be important to job search success, and so to the integration of potentially excluded groups, such as young people (Holzer, Reference Holzer1988), women returners (see Chapple's, Reference Chapple2002, US research) and ethnic minorities (see Ooka and Wellman's, Reference Ooka, Wellman and Fong2006, research in Canada). Furthermore, it is clear that social networks, whatever their impact, are at least considered to be important by many job seekers, and appear to play a crucial role in regions where formal employment services are less advanced or less used by communities (Lindsay et al., Reference Lindsay, Greig and McQuaid2005).

Long-term unemployment, area-based disadvantage and the tertiary sphere of sociability

The manner in which a lack of social network resources (and especially the absence of a diverse range of work-related contacts) can impact on employability also has a spatial dimension. Research undertaken in the US (Wilson, Reference Wilson1996) has pointed to the links between social network structure and poverty, with residents of ‘segregated’, disadvantaged urban areas finding the range and quality of their employment networks limited by poverty and social isolation. As Reingold's (Reference Reingold1999) work in US cities notes, people in such disadvantaged urban areas can struggle to access ‘vertical networks’ that provide access to better jobs – this ‘network poverty’ (Perri 6, 1997) can severely limit individual opportunity.

Furthermore, it has been shown that the experience of long-term unemployment in particular, and the related restricted social contact and sense of exclusion experienced by some people (which can be particularly intense in disadvantaged urban areas), can in itself undermine confidence and reinforce geographical isolation. MacDonald et al.'s (Reference MacDonald, Shildrick, Webster and Simpson2005: 884) work with young people in disadvantaged neighbourhoods in English cities finds that social networks tend to be strong, but tightly knit, based on close personal or family contacts, and limited to the immediate locality – the limited scope and range of social networks can ‘close down’ employment opportunities for young people, with their job search often ‘proximate with the spatial reach of such networks’. Similarly, Hedstrom et al.'s (Reference Hedstrom, Kolm and Aberg2003) research with urban job seekers in Sweden identifies the negative impact of restricted social networks (concentrated in disadvantaged areas) on exits from unemployment.

Gallie et al. (Reference Gallie, Paugam and Jacobs2003) explore a more specific relationship between what they call the ‘tertiary sphere of sociability’ and the experience of unemployment. Using EU-level data, they demonstrate that unemployed people are increasingly less likely to participate in organised social activity, through religious institutions, social/sports clubs or community organisations. Gallie et al. (Reference Gallie, Paugam and Jacobs2003: 28−29) suggest that limited access to the tertiary sphere of sociability is a function of pre-existing job seeker characteristics within specific social settings, rather than being caused by unemployment per se – but they are also clear that the social isolation experienced by some job seekers ‘may seriously accentuate the psychological deprivations arising from unemployment’. Julkunen's (Reference Julkunen2002) research with young people in Scotland and Nordic countries more clearly identifies a relationship between long-term unemployment and a decline in activities within the tertiary sphere of sociability. What is clear is that where long-term unemployed people are excluded from this tertiary sphere of sociability (assuming that long-term unemployment may exacerbate such forms of isolation), there are potential implications for both the further erosion of social networks, and broader socio-psychological and mental health problems.

Profile of the study areas

The research was carried out in Glasgow, Scotland's largest city (and the UK's fourth largest) with a working age population of approximately 370,000. At the time of the research, claimant unemployment was 4.8 per cent, compared to Scottish and UK figures of 3.2 per cent and 2.6 per cent respectively. However, claimant unemployment figures mask a more serious problem of pockets of long-term economic inactivity, with large numbers claiming non-work related benefits – at the time of the research more than 100,000 working-age residents of Glasgow were claiming benefits of some kind. Recent research has suggested that the ‘real level of unemployment’ (non-voluntary inactivity among those willing to work) within the city remains at 14 per cent, well above Scottish and UK averages (Beatty et al., Reference Beatty, Fothergill, Gore and Powell2007).

The research was carried out in two areas of Glasgow: Pollock in the south-west, and Springburn in the north of the city. Most interviewees resided in ‘Social Inclusion Partnership’ (SIP) areas within these localities (SIPs were, until recently, the Scottish Government's main targeted policy to address social exclusion in particularly deprived localities). Both areas reported above-average levels of unemployment and economic inactivity. Household surveys at the time of the research found that in the ‘Greater Pollock SIP’ area and the ‘Springburn and East Balornock SIP’ area (which contained the Springburn study area) only 32 per cent of 17−74 year olds were in work (compared with Scottish and Glasgow averages of 58 per cent and 47 per cent respectively). The same research found that 49 per cent of Greater Pollock and 65 per cent of Springburn adults held no formal qualifications, compared to 41 per cent of Glasgow adults, and 33 per cent of the Scottish population.

Methodology and sample profile

Structured, face-to-face interviews were carried out by the author with 220 claimant unemployed people at Jobcentre Plus offices in the two areas.Footnote ² Interviews were conducted in quiet areas near the front of offices, but well away from Jobcentre Plus staff, to ensure that interviewees felt confident when discussing sensitive issues. The sample accurately reflected the gender balance within the claimant unemployed population in Glasgow (80 per cent men and 20 per cent women). The age profile of the sample also reflected the wider claimant unemployed population reasonably accurately. However, the long-term unemployed were deliberately over-represented, to ensure a usable sample for comparing the barriers faced by these and other job seekers. Approximately one-third of interviewees had been unemployed for more than one year, compared to 16 per cent of Glasgow's unemployed. The gender balance of the long-term unemployed group was identical to that of the overall sample. The long-term unemployed group did, however, have an older age profile (reflecting the concentration of older job seekers among the total long-term unemployed population in the city). In terms of the key ‘individual factors’ affecting job seekers’ employability (McQuaid and Lindsay, Reference McQuaid and Lindsay2005), there is evidence that those participating in this study faced significant barriers to work. For example, 37 per cent of interviewees held no qualifications, rising to 61 per cent among the long-term unemployed. Competing in a Scottish labour market where only 17 per cent of economically active people are similarly unqualified, these job seekers will be disadvantaged.

Social networks and job seeking

A basic analysis of unemployed people's use of job search methods demonstrated that social networking played an important role, but was one search strategy among many. As Table 1 shows, more than 90 per cent of interviewees reported using newspaper advertisements to look for work on a regular (defined here as weekly) basis, while the majority also visited Jobcentres to use computerised ‘Jobpoint’ search databases and/or seek the advice of staff. However, social networking also emerged as an important tool, with 53 per cent using close friends and family and 40 per cent using work-related ties.

Table 1 Percentage of interviewees using selected job search methods (%)

Notes: **Significant difference at 1% level (Chi squared test). *Significant difference at 5% level.

It is clear that the long-term unemployed among the sample were less connected with former colleagues and other (potentially more relevant) work-related network ties. Only 29 per cent of those unemployed for more than one year used such contacts for job seeking on a regular basis, compared to more than 45 per cent of those unemployed for less than a year (a difference significant at the 5 per cent level using chi squared test).Footnote ³ Long-term unemployed people were also slightly less likely to use close family and friends to look for work, but the difference here was not significant. Indeed, among the few significant differences in job seeking between the long-term unemployed group and other interviewees were two potentially crucial aspects of informal search behaviour – the use of work-related social networking and directly approaching employers (in the latter case the difference between the two groups was significant at the 1 per cent level using chi squared test).

If the strength of weak ties thesis has merit, then there may be cause for concern here. The majority of unemployed job seekers were not able to network with work-related personal acquaintances on a regular basis, and long-term unemployed people in particular were excluded from this form of networking. Given the evidence that such work-related ties provide more valuable, ‘non-redundant’ job search information, a lack of access to such networks could reinforce the disadvantage experienced by the long-term unemployed.

Further probing of interviewees’ perceptions of their own social networks confirmed that there were significant differences between the experiences of the long-term unemployed and other job seekers. First, long-term unemployed people were less likely to see networking as an important aspect of the job search process. Those who had been unemployed for less than one year were significantly more likely (5 per cent level using chi squared test) to consider using personal contacts to be ‘very important’ to looking for work (27 per cent of these more recent job seekers, compared with only 15 per cent of the long-term unemployed) – although it is also noticeable that relatively few job seekers in either group appeared to acknowledge the potential importance of networks. Second, the long-term unemployed were significantly more likely (1 per cent level using chi squared test) to report never having found out about a job through personal contacts (42 per cent of the long-term unemployed, compared to 24 per cent of other job seekers).

Finally, as Figure 1 illustrates, long-term unemployed people were less likely to have contacts in the same type of work (i.e. sector and/or occupation) that they were pursuing. Only 24 per cent of long-term unemployed people said that their main job search network ties – the contacts that they most often used for looking for work – were employed in the type of work that they were looking for, compared to 46 per cent of other job seekers (significant at the 1 per cent level using chi squared test). Of greater concern is that more than one-fifth of the long-term unemployed said that their main network contacts were themselves unemployed. In comparison, less than 5 per cent of other job seekers most often used other unemployed people for job search networking (the difference between the two groups was significant at the 1 per cent level using chi squared test).

Figure 1. Employment status of ‘main social network contacts’ used for job search (%).

The experience of long-term unemployment and the tertiary sphere of sociability

The same interviews with job seekers sought to probe the extent to which the experience of unemployment was linked with broader processes of social exclusion. Accordingly, the research involved a series of questions about job seekers’ engagement in regular social activities. Long-term unemployed people were significantly less likely (5 per cent level) to report socialising on at least a weekly basis with friends/family (29 per cent, compared to 46 per cent of those unemployed for less than one year). Furthermore, although only just over one-fifth of all job seekers were involved in regular, organised social activity (what has been termed the tertiary sphere of sociability), there were again significant differences between the long-term unemployed and other job seekers. Only 11 per cent of long-term unemployed people reported being involved in such social activities (for example, attending a sports or social club) on a weekly basis, compared to 26 per cent of other job seekers – a difference significant at 1 per cent level using chi-squared test.

These data, taken together with the findings above on the regular job search activities of interviewees, highlight how long-term unemployment can contribute to a ‘narrowing’ of social life − as Sinfield (Reference Sinfield1981: 92) notes, there is a danger that the long-term unemployed ‘tend to withdraw from society and to expect less’. A decreasing engagement in formal and informal sociability was clearly linked to the experience of unemployment (and the resulting loss of income and lack of daily contact with former colleagues and acquaintances) for many people. Many interviewees talked about how they had been less able to engage in social activity since losing their jobs, with long-term unemployed people particularly likely to raise these issues. Indeed, 46 per cent of all interviewees said that they socialised less as their period of unemployment grew (56 per cent of the long-term unemployed, compared to 42 per cent of other job seekers − a difference significant at 1 per cent level using chi squared test).

Gallie et al. (Reference Gallie, Paugam and Jacobs2003), reporting on European Household Panel Survey data, argue that declining access to the tertiary sphere of sociability is less to do with the experience of unemployment in itself than the initial, relative social isolation of some people in some communities even prior to becoming unemployed. But the (admittedly small) study reported above highlights the need for further research on how long-term unemployment in particular may accentuate the experience and impact of social isolation. This is not to say that being in the category ‘long-term unemployed’ in itself explains why certain job seekers were more likely to fall victim to these forms of social exclusion − rather, long-term unemployed people participating in this research faced multiple additional barriers to work, which were in turn often associated with more limited access to social capital. Compared to other interviewees, the long-term unemployed were statistically significantly more likely to: have few/no qualifications, consider their occupational skills to be less than ‘good’, think their job prospects to be poor and report work histories with little by way of stable employment − all job seeker characteristics that were also significantly associated with more limited access to social networks. The long-term unemployed were also more likely to report health problems.

Finally, it is important to note that those unemployed for one year or more were significantly more likely (1 per cent level using chi squared test) to live alone (51 per cent, compared to 31 per cent of other job seekers). Single occupancy in Glasgow is higher than the Scottish average (41 per cent and 33 per cent of the households respectivelyFootnote ⁴) but these data suggest that long-term unemployed people are particularly likely to be alone – important, given that there is evidence that living alone appears to reduce the chances of exiting unemployment (Gallie et al., Reference Gallie, Paugam and Jacobs2003; Green and Owen, Reference Green and Owen2006). The long-term unemployed were also significantly less likely to have access to basic household resources such as a telephone and their own transport; and they often reported very low household incomes. More than half of the total sample reported a household income of less than £100 per week (53 per cent, rising to 62 per cent of long-term unemployed people). By way of comparison, a survey for Communities Scotland conducted near the time of the research concluded that only 9 per cent of Glasgow households had a net weekly income of less than £100 (Newhaven Research, 2004). It is therefore clear that most unemployed people within the sample, and especially the long-term unemployed, were among Glasgow's poorest residents – the legacy of a disempowering benefits system that provides little by way of disposable income while meeting housing costs through passported, means-tested benefits. As Gallie et al. (Reference Gallie, Paugam and Jacobs2003: 28) note, it may be that ‘unemployment heightens the risk of people falling into poverty, and poverty in turn makes it more difficult for people to return to work’.