How are poverty, ethnicity and social networks related?

Key points

• Eight out of ten people have friends from a different neighbourhood.
• Two out of three people from ethnic minority groups have friends from a different ethnic group. The figure is lowest for White ethnic groups.
• People in poverty are less likely to have mixed social networks.
• Three factors are related to a reduced likelihood of being poor: having a mixed ethnic friendship network, having friends from outside your neighbourhood, and having all friends who are employed.
• Other factors, such as having no qualifications, are stronger predictors of being in poverty than social network composition.
• The benefit of having mixed friendship networks is felt most by ethnic groups with lowest levels of poverty and least by those with highest levels of poverty.
• The number of close friends was a stronger predictor of poverty status than how mixed somebody’s social network was.
• Interventions should be targeted at deprived neighbourhoods and ethnic groups who have mixed networks but are not benefiting from them as much as they could.

Introduction

Building on previous research, and policy interest in social networks, this study aimed to investigate how social networks are associated with poverty for ethnic groups in the UK. The research made use of a large national survey, the UK Household Longitudinal Study, Understanding Society.

This project addressed three questions:

1. Do ethnic and income groups have different types of social networks?
2. Do mixed social networks reduce the risk of being poor?
3. Does where you live matter for the relations between poverty, ethnicity and social networks?

The analyses revealed four key findings:

1. Ethnic and income groups have different types of friendship networks.

At least two in three people from ethnic minority groups in the UK have friends from a different ethnic group. More than four in five people (of all ethnic groups) have friends from a different neighbourhood.

White British and White Irish ethnic groups are less likely than ethnic minority groups to have ethnically mixed social networks. Mixed ethnic group networks are most common for Mixed, Black African and Black Caribbean ethnic groups and least common for White British and White Irish ethnic groups. Having friends from outside their neighbourhood is most common for Black African and Mixed ethnic groups and least common for Pakistani and Bangladeshi groups.

Networks including unemployed friends are most common for Bangladeshi, Pakistani, Black Caribbean and Black African ethnic groups.

People in poverty are less likely than those not in poverty to have ethnically mixed social networks and friends from outside their neighbourhood; and more likely to have networks including unemployed friends.

2. Mixed social networks can reduce poverty, but not as much as other factors

There is a clear message from the results: the probability of being poor and of being very poor is less for individuals with mixed friendship networks than those without mixed friendship networks, for ‘families doing ok’, ‘struggling families’, ‘young solos’ and ‘single parents’ in all ethnic groups. For example, having a mixed friendship network could reduce the likelihood of an individual in a struggling family’ being very poor by a third compared with not having a mixed friendship network.

Taking into account individual and household characteristics associated with poverty, three factors relating to social networks reduced the risk of being poor:

• having a mixed ethnic friendship network;
• having friends from outside your neighbourhood; and
• having all friends who are employed.

However, there are two important caveats to these results:

1. Other factors such as having no qualifications or being separated or divorced, are stronger predictors of being in poverty than social network composition.
2. The benefit of having mixed friendship networks in terms of the degree of reduction in poverty is felt most by ethnic groups with lowest levels of poverty (particularly the White British ethnic group), and least by those with highest levels of poverty (particularly the Pakistani and Bangladeshi ethnic groups).

3. Having two or more close friends is associated with a lower likelihood of poverty

Black African, Pakistani, Bangladeshi and Black Caribbean ethnic groups are most likely to have only one or no close friends; as many as 1 in 5 Black African people reported having only one, or no, close friends.

The number of close friends was a stronger predictor of poverty status than how mixed somebody’s social network was. Having two or more close friends reduces the likelihood of being in poverty compared to having no or one friend. This suggests that social isolation is a particular risk for poverty (or consequence of living in poverty).

4. The level of deprivation of a neighbourhood affects the relationship between mixed social networks and poverty

People living in advantaged neighbourhoods benefit most from having mixed social networks, in terms of likelihood of being in poverty. This is consistent with the finding that the most well off individuals benefit most from having mixed social networks, in terms of potential poverty reduction. Conversely, having mixed friendship networks can be associated with higher risk of poverty (compared with not having mixed friendship networks) for people in the most deprived neighbourhoods. This may be to do with the costs associated with maintaining mixed social networks.

Implications

These results clearly indicate that there is not an overall ‘problem’ of a lack of mixed friendship networks for people from ethnic minority groups. On the whole, people in all ethnic minority groups have more mixed networks in terms of both ethnic group and neighbourhood than the White British majority. Therefore, if attention is to be paid to encouraging more mixing, this could be best directed at the White British, rather than at minority groups. In addition, individuals and households in poverty, regardless of ethnicity, have less mixed networks and could benefit from support to increase the diversity of their contacts.

This research suggests that having mixed social network composition can reduce poverty risk. Strategies concerned with social networks may thus be appropriate. However, social networks cannot be viewed in isolation; broader inequalities, including in education and employment, shape the networks that people have access to and also have a greater independent effect on levels of poverty.

The findings of this report support previous research that suggests that how people use their social networks is as important as what kind of networks they have. Initiatives may best be focused on supporting people to use social networks; increasing their understanding of the potential to draw social, emotional and financial support from their networks, and the ways in which these benefits may be maximised. These initiatives could be targeted at groups who have fairly mixed networks but do not seem to be benefitting from them as much as others. This would include people from Pakistani and Bangladeshi ethnic groups and people in the most deprived neighbourhoods. Such initiatives might include mentoring schemes, toolkits and promotion of the third sector’s role in supporting network development and use. It would be important for such support to be ‘no cost’ to the target groups if it is to be effective and sustainable.

The finding that people from some ethnic minority groups are particularly at risk of having no close friends, and that this in turn may increase their risk of being in poverty, suggests that social isolation should be a priority for policy and practice.

Conclusion

The findings suggest that initiatives on social networks and poverty should have two focuses:

• encouraging greater use of mixed social networks; and
• ensuring that individuals are not socially isolated.

This research suggests that such initiatives can improve individuals’ economic wellbeing. They also suggest that these types of interventions may be most suitably targeted at deprived neighbourhoods, groups who have mixed networks but are not benefiting from them as much as they could (including Bangladeshi, Pakistani and Black Caribbean ethnic groups) and groups who appear particularly socially isolated in terms of number of close friends (namely Black African and Pakistani ethnic groups). Such interventions need to identify and address barriers that discourage nurturing close friendships and diverse social networks; and to minimise the costs of developing and maintaining social networks.

Although this study has revealed mixed social networks to be associated with poverty for ethnic groups in the UK, it has also demonstrated that other factors – education and relationship status for example – are stronger drivers of poverty. This suggests that, in addition to any attention to social networks, continued emphasis is needed on addressing the fundamental dimensions of the prevalent and persistent socio-economic ethnic inequalities in the UK.

Methods

This study used Wave 3 of the UK Household Longitudinal Study, Understanding Society (data collected in 2011-12). Understanding Society allows analysis of our three main themes: poverty, ethnicity and social networks. Social friendship network composition was measured via four indicators: Ethnic mix of network, neighbourhood mix of network, proportion of friends who are unemployed and number of close friends. Poverty was defined as being below the poverty line of 60 per cent of median equivalised gross household income.

The analyses were undertaken for ten ethnic groups: White British, White Irish, Other White, Mixed, Indian, Pakistani, Bangladeshi, Black Caribbean, Black African, Other. Neighbourhood analysis used Middle Super Output Areas and Indices of Multiple Deprivation for England and Wales.

The analyses were conducted in Stata 13 and are presented via cross-tabulations and regression models (single and multilevel binary and ordered logit).

This study has not considered the dual-direction of the relations between poverty and social networks, how social networks are associated with changing poverty status or the type of social network.