Cost-benefit analysis of extending support to domestic abuse victims with NRPF: A technical report for the Domestic Abuse Commissioner

Kath Scanlon, Bert Provan, Ria Ivandic, Mariña Fernández-Reino, Fanny Blanc and Christine Whitehead

Published 13 December 2022

This report provides the technical underpinning to the report by the Domestic Abuse Commissioner’s Report1 laid before Parliament in December 2022. That report was in response to the Home Office request that she set out ‘the gaps in evidence available to the Home Office on support for Migrant Victims of Domestic Abuse, by establishing the number of victims and survivors of domestic abuse who have no recourse to public funds, the cost of supporting those who need support, and the cost benefit of such interventions.2’ LSE was asked by the Commissioner to provide a detailed technical analysis to underpin her response. This LSE report provides our views on relevant evidence but no policy recommendations. Such recommendations are made in the Commissioner’s report, having considered our evidence alongside the evidence of other stakeholders and people she has consulted. Our analysis required us to make strong assumptions, and there is significant uncertainty around many of them. There is considerable uncertainty in published validated statistics about the numbers of migrants in the UK with each visa status, and even more uncertainty about numbers of undocumented migrants and of ‘visitors’. There is also uncertainty about how many of these migrants currently experience domestic abuse, and an added and independent uncertainty about the proportion of those people who would present at services. We have also had to make various assumptions which affect the costs and benefits. We address and mitigate these and other uncertainties through our sensitivity analysis and additional modelling in this report. In addition, an accompanying Excel workbook sets out in detail our assumptions and how they underpin our conclusions, which allows for further sensitivity analysis and modelling to be done.