Do white voters support welfare policies targeted at ethnic minorities? Experimental evidence from Britain

ABSTRACT

Academics and policymakers have begun discussing ethnically focussed policies to offset the penalties ethnic minorities suffer due to discrimination, but little is known about how such policies would be received by the public. We employ new survey experiments fielded in Britain to examine public reactions to equal opportunity policies targeted at working class, black or Muslim Britons, and redistributive policies aiming to reduce inequality between the rich and the poor or between whites and ethnic minorities. We find widespread support for class- or income-targeted policies among white British respondents, but much lower support when identical policies are ethnically targeted. We find evidence for two mechanisms which help to explain this drop in support. Firstly, political and ideological values such as egalitarianism and support for an active state strongly predict support for class- and income-targeted policy, but have a much weaker effect on support for ethnically targeted policy. Secondly, hostility to immigrants has a strong negative effect on support for ethnically targeted policy, but no effect on support for class- and income-targeted policy. The combination of intense opposition from anti-immigrant voters and indifference from traditionally pro-welfare left wingers results in very little public support for explicitly ethnically targeted policy.

KEYWORDS:

• Race-targeted policies
• public opinion
• anti-immigrant sentiment
• survey experiments

Acknowledgements

We thank Maria Sobolewska, Jens Hainmueller, Dominik Hangartner, Dan Hopkins, Rahsaan Maxwell, Sara Wallace Goodman, Justin Gest, Cara Wong, Paul Sniderman, Elisabeth Ivarsflaten, Scott Blinder and participants at the 2014 International Society of Political Psychology annual meeting for their helpful comments on earlier versions of this paper. All remaining errors and omissions are ours alone.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. One notable exception is Northern Ireland, where affirmative action policies have been used to combat religious discrimination (McCrudden, Ford, and Heath 2004; Muttarak et al. 2013).

2. A detailed report on the sampling strategy employed by YouGov, and the demographic balance of the sample, is provided in the Online Appendix.

3. More detail on variable construction, and correlations between variables, is provided in the Online Appendix.

4. Some of these measures show a modest relationship, reflecting a common left-wing/welfarist orientation. Details and robustness checks on our regression models are provided in the Online Appendix.

5. These group attitudes measures are modestly related to each other, most likely reflecting a degree of common hostile or tolerant orientations towards outgroups. Detailed analysis of correlations, and robustness checks on modelling, are provided in the Online Appendix.

6. This is not the result of collinearity – the effect of social distance is weaker and less robust than that of hostility to immigrants, across a range of specifications. See the Online Appendix for details.

7. See the Online Appendix for details.

8. Simulations undertaken using ‘Clarify’ (King, Tomz, and Wittenburg 2000).

Additional information

Funding

We gratefully acknowledge support for this research from the British Academy [grant number SG102531].