The Policy Effects of Disability Benefit Reforms: A Comparative Analysis of Denmark, the UK and Sweden – Working Paper

Abstract

Introduction: Numbers of claimants of disability benefits have seen a sharp rise, with reforms introduced to reduce this. Sweden and the UK have introduced reforms to increase conditionality for disability benefits (reduced generosity and stricter eligibility) whereas Denmark has pursued a more supportive approach (expansion of the subsidised employment scheme ‘fleksjobs’) combined with some increase in conditionality. We examine the effect of reforms in these three countries using a difference-in-difference-in-difference (DDD) design.

Methods: Data from Survey of Health, Aging and Retirement in Europe and English Longitudinal Study of Ageing were used for two time periods: 2004-2007, and 2011-2017. For the DDD approach, the first difference represents the difference between those with and without a health condition, the second difference is the difference between the first difference and before and after the policy intervention, and the third difference is the difference between the second difference and the study country and the comparison group. Multinomial logistic regression was used for labour market outcomes, logistic regression for poverty and linear regression for the standardised depression measure. The DDD are measured using three-way interactions between binary variables for country vs. comparison group, independent variable, and before and after the policy intervention.

Results: For people with severe mental health problems, employment rates have decreased and poverty rates have increased between the time periods for the UK and Sweden, whereas the opposite trends are seen in Denmark. From the DDD analysis, Danish reforms reduced economic inactivity for those with severe mental health problems, while employment and unemployment both increased, while in the UK, unemployment increased. For people with physical limitations in the UK, increased employment and decreased inactivity was seen. Risk of poverty increase for people in Sweden with severe and mild mental health problems, while it fell for people with physical limitations in Denmark. There is a clear increase in mental illness for those with physical limitations in the UK after the reforms.

Conclusion: There is a clear risk that policies aimed at moving people from disability benefits might have the effect of moving them onto less generous unemployment benefits as has happened in the UK and Denmark for people with mental health problems. Moreover, as can be seen in the UK, policies which have helped people with physical health conditions move from inactivity to employment have not been effective for those with mental health conditions. In Sweden reforms were associated with increased poverty amongst people with mild mental health problems. In Denmark reforms were associated with reduced poverty amongst people with severe mental health problems and physical limitations.

Keywords: Disability benefits, mental health, employment, conditionality, policy