Friday, August 07, 2015

UK Benefit Sanctions: Ethics and Evidence

These are personal views on an ongoing policy issue. They are provided for the purpose of discussion. They are not intended to provide an institutional view.  

There has been a great deal of recent debate about the use of benefit sanctions in the context of employment activation. The Guardian recently published a piece outlining the annual number of sanctions over time. There is a debate about how to calculate these numbers but even looking at the DWP's own raw spreadsheets makes it clear that sanctions are now a normalised form of intervention in the UK benefit system rather than a "last resort for a tiny minority"  as has been unconvincingly claimed by a number of senior politicians (See David Webster from Glasgow University for detailed discussion on sanctions numbers which have prompted the UK statistics watchdog to ask DWP to provide greater clarity to their figures - see here).

The obvious narrow question is whether sanctions are effective in these contexts in terms of people gaining employment. A textbook economics model might suggest so but there is obviously a volume of evidence as to the potential perverse effects of incentives employed in a punitive manner (one of most cited reviews here). A number of literature reviews point to what is basically a very mixed literature (see JRF review here). There are some interesting studies showing positive effects on employment search and outcomes (see Blundell et al for a detailed study of an early Blair-era policy) but amidst a sea of evidence finding not very much at all. The context of the economy in terms of vacancies and matching of applicant skills and regional demographic characteristics is also clearly relevant.

One question for a sanctions policy is how it is being applied. The Oakley review pointed to a number of features of job activation services that could be improved. Many of these reforms fall squarely with a behavioural policy type framework and experimenting with design of communication, instructions etc., potentially offers a mechanism for improvement of these services. Sunstein's recent discussion of ethics in the context of soft interventions is still very relevant but it seems clear that there is some role for behavioural design in the context of job activation to attempt to make the processes clearer and more effective.

Lord Layard and colleagues have put the potential for improved access to CBT and other psychological therapies squarely on the UK agenda. There is a great deal of debate surrounding this agenda with critics arguing it simplifies the treatment of mental health. The role of psychological support services in a conditional benefits framework is ethically very complex. This has been raised in a recent paper by Friedli and Stern that I have linked to here on a number of occasions. Friedli and Stern are sceptical in general about the use of any psychological intervention in the context of employment activation. This debate is extremely important. There is certainly abundant evidence that unemployment is stressful and potentially damaging to mental health and scarring in the long-run. However, if mental health services are provided in the context of potential sanctions for non-compliance whether explicit or implicit there are clearly a host of both practical and ethical issues.

A wider question is the impact of being sanctioned on those who are sanctioned. This question was asked many times during the Select Committee's questioning of the then employment minister Esther McVey and a senior DWP official but was answered with a general reply about the improving numbers of people in employment. In general there is very little evidence on what happens. The Guardian newspaper has published several reports of suicides, food banks usage etc., Loopstra  and colleagues have come to closest to documenting causal linkages with a clear relation between sanctions and the rise in food bank usage (discussion here) but the causal connection is still difficult to test with the existing data. The Guardian reported in January 2015 that DWP has investigated 60 suicides surrounding withdrawal of benefits. Various groups have cited a correlation between sanctions and suicides e.g here. There is an obvious possibility for sanctions to hit harder on those experiencing mental health problems. All of this is worthy of discussion but far from solid evidence of effects of sanctions.

It raises a question about the role of evidence in policy and who should collect it. The DWP's response to the BMJ paper by Loopstra et al was:

 “As the authors admit themselves the data does not give a full picture. What we do know – according to independent figures from the Office for National Statistics – is that we now have a record number of people in employment in this country and there are two million more people in private sector jobs compared to 2010".

There is no doubt that both internally and externally DWP have access to people who could produce a better response than this non-sequitur. It is also clear that there is no impetus to collect the type of information that would enable serious analysis of the effects of sanctions leading to a call last month for an urgent review into the sanctions regime from the independent advisors to the government (the Social Security Advisory Committee details here). It is plausible that these policies are leading to severe psychological distress, crime, suicide and other outcomes. At present, simply no-one knows as the data is not available to test this. A precautionary principle would suggest greater phasing of such a policy in line with an evidence-strategy. There are several bodies in the UK who could do this research with requisite support from the DWP themselves. (See Prof Martin McKee's comments on this).

As several others have pointed out, sanctions are not just used in unemployment and work disability contexts and are becoming an increasingly used form of intervention in other areas such as housing. In that sense it is even more important to understand both their effectiveness from a narrow outcome point of view and wider spillover effects they might engender. Such an exercise requires interdisciplinary knowledge combined with detailed understanding of the mechanics and implementation of benefits and sanctions.

It also is worth thinking about these policies through the lense of political communication. The popularity of these policies with centre voters is a very interesting aspect of this debate and the extent to which sanctions will be used a way of signalling toughness to winnable voters will be interesting to observe over the next election cycles in different countries. It also raises the question as to how this will influence attitudes to unemployment. What do people interpret from the communication of these policies about the extent of social welfare cheating? Furthermore there is the question as to the welfare function a government maximises. It may be the case that the welfare of the unemployed is not a priority for government for political and related reasons which further heightens the degree of ethical tensions faced by people tasked with designing a system like this subject to political constraints.

Some Further Reading: 

Beatty et al (2015). Benefit sanctions and homelessness:a scoping report. Crisis (the national charity for single homeless people).

Etherington & Daguerre (2015). Welfare reform, work first policies and benefit conditionality: reinforcing poverty and social exclusion? Centre for Enterprise and Economic Development Research, Middlesex University in London.

Fitzpatrick et al (2015). Destitution in the UK: an interim report. York: Joseph Rowntree Foundation.

House of Commons Work and Pensions Committee (March 2015). Benefit sanctions policy beyond the Oakley Review.

Martin (2015). Activation and active labour market policies in OECD countries: stylised facts and evidence on their effectiveness. IZA Journal of Labor Policy mFebruary 2015, 4:4.

Norman & Uba (2015). Austerity measures across Europe. In Defence of Welfare 2.

Perry et al (2015). Emergency Use Only: Understanding and reducing the
use of food banks in the UK. The Child Poverty Action Group, Church of England,
Oxfam GB and The Trussell Trust.

Spencer, A., Ogden, C. & Battarbee, L. (March 2015). #cheshirehunger Understanding Emergency Food Provision in West Cheshire. Research Report by West Cheshire Foodbank, The University of Chester, The Trussell Trust, Cheshire West Citizens Advice Bureau, DIAL West Cheshire (DIAL House), Chester Aid to the Homeless, The Debt Advice Network and The Salvation Army

No comments: