What working in mental health services has taught me about reforming social security

Back in 2018, I finished up an 18-month secondment at the Department for Work and Pensions (DWP) and decided to retrain as a mental health social worker. After a decade of policy work, I wanted to take a break and gain some frontline experience. I was also deeply frustrated by the lack of progress I had been able to make during my time at the department on issues around benefits and employment support for ill and disabled people.

As a new Labour government grapples with the challenge of record numbers of people who are classified as ‘economically inactive’ due to disabilities and poor health, many of the issues I worked on at DWP remain pertinent. However, back in a policy role at the New Economics Foundation and trying to propose solutions, I find myself reflecting as much on my time in NHS services as my experience in the centre of the department.

I had been asked to come into DWP from the charity Mind, in a bespoke role as a mental health policy advisor. Although this position afforded me freedom to work on a range of areas of both policy and delivery, I struggled to overcome the institutional and cultural barriers to the changes I felt were needed. I wrote about my frustrations subsequently in a paper for the think tank Demos.

Frontline mental health work felt like it would give me the opportunity to have a much more direct positive impact. It was also something that I had wanted to try since being supported by mental health services myself when I was in my early 20s. The appeal of training as a mental health social worker specifically was that it would allow me to make use of my knowledge of relevant law and social policy, and in turn enhance my practical understanding of them. 

Adjusting to frontline work

The biggest adjustment in moving from policy work to a frontline role was losing the relative safety and comfort of speculating on, or even actively shaping, how a service should work and becoming personally responsible and accountable for how it is experienced. My clearest memory of this transition is walking up the steps of a tower block on my way to my first solo home visit to someone and thinking “what on earth do I have to offer this person?!”.

The reality is, however, that when you turn up at someone’s front door with an NHS lanyard on, it’s the institution that’s setting their expectations as much as your personal credentials. On the one hand this tends to grant you a degree of credibility, on the other it can come with a lot of baggage. This is particularly true in mental health services where people are often experiencing a heightened sense of threat and may be fearful of potential coercion.

Your ability to build a positive relationship with someone is dependent on harnessing the best of this inheritance while overcoming the worst. Ultimately, I found that it’s the quality of the relationships you can form that determine how effectively you can support someone. If someone feels heard, respected and supported by you, they are more likely to trust you, open up to you, work with you, and be receptive to your questions and advice. 

Relationships in the social security system

There are some clear parallels to be drawn with the task facing DWP when it comes to supporting the growing proportion of those on universal credit with mental health problems, as well as the wider cohort. In a recent research project at the New Economics Foundation, we ran workshops with people in the social security system to hear about their experiences of interacting with DWP and jobcentres, and explore how support could be improved.

People expressed a lot of fear and mistrust of DWP, this was particularly strong among those who are ill or disabled. There was a sense that jobcentres were trying to push them off benefits, rather than focusing on supporting them into good jobs. It was clear from the workshops, and from the wider evidence we reviewed, that successful employment support is dependent on building good individual relationships between those providing support and those receiving it.

Accounts of more positive interactions with DWP or jobcentre support were often framed in contrast to people’s preconceptions or wider experiences of the system – which they feel is primarily focused on checking up on them and ensuring compliance with rules. Good members of staff have to overcome these expectations in order to build a connection with someone, often by working around the negative aspects of the system they have to operate within.

Recourse to coercion

People’s relationship with the DWP as an institution and its staff as individuals is fundamentally mediated by their receipt of benefits and, for many, the ever-present threat that these could be reduced or stopped for failing to comply with rules and expectations. In a similar way, many people’s experiences of mental health services are strongly shaped by the knowledge that professionals have recourse to coercion if they deem it necessary.

Within mental health law, services and professions, there has been a conscious and explicit effort to only use coercion as a last resort and to promote choice and control wherever possible. This stems from an increased understanding that it is both more ethical and also a more effective way to support people towards good outcomes. This is not to suggest that mental health services and individual professionals always get the balance right – they often don’t – but it is a dynamic that most try to actively consider and manage.

In my frontline work, if I started a meeting with someone struggling with their mental health by suggesting that they need to follow my advice or they may end up being detained in hospital, it would likely completely undermine any effort to build rapport and trust with them. Of course, they may be mindful of this risk even if I didn’t make explicit reference to it, but doing so makes it central to our interaction rather than on the periphery.

The impact of conditionality

In contrast, the use of conditionality in the benefits system is front and centre in people’s interactions with the jobcentre. The relationship opens with people being essentially obliged to agree to a ‘claimant commitment’, which sets out the requirements people must meet to avoid a potential benefits sanction. In theory, these requirements are supposed to be tailored to the individual, but participants in our workshops reported that the process felt like a tick-box exercise where they had no real choice or control.

The long-held assumption within the DWP has been that strict and prescriptive conditionality is necessary to ensure a minimum floor of compliance, to reduce the risk of someone receiving benefits but doing nothing in return. However, this approach hugely underestimates and underprices the ceiling it puts on the quality of engagement. As one of our workshop participants put it: “The real problem with conditionality is that it’s so limiting to interactions with your work coach.”

Many people engage with DWP in a very cautious and defensive manner – doing what is required of them to avoid punishment but not seeing it as an opportunity to open up and seek genuine support. This may well be interpreted by some as resistance to moving towards work, when in fact it is a fearful and conditioned response to the terms of engagement set by DWP.

If someone is not engaging with mental health services, a good team or individual professional would first reflect on why this might be happening and whether there is anything they could do differently to build trust and rapport, rather than jumping to simply blaming the service user.

But the social security system does little to meaningfully encourage work coaches to go beyond their core function of monitoring compliance. In the absence of strict and prescriptive conditionality, it would be much clearer whether work coaches are doing a good job of fostering genuine engagement, which in turn could provide insights to improve practice across the workforce.

The politics of mental health

The general shift over time from a more coercive to more collaborative approach in mental health services has reflected and shaped public opinion which has, broadly, become more compassionate towards and less fearful of people with mental health problems. However, this transition has, at times, caused tensions in terms of whether policymakers are being seen to get the balance right between promoting autonomy and managing risks.

Progress has stalled or been set back when someone experiencing mental health problems causes harm to themselves or others and it is seen by some as indicative of this balance being out of kilter. In reality, such incidents are often linked to a variety of factors, such as the underfunding of services or poor practice by an individual or team. But the narrative of a general failure to sufficiently prioritise the management of risk can easily take hold.

This has meant that the politics of mental health has sometimes felt more focused on being seen to respond to public concerns (or even just assumed or perceived public concerns) than on fostering the services, and the underlying social and economic conditions, that will be most supportive of people’s mental health, wellbeing and safety.

Conditionality and public opinion

Social security policy has also been heavily shaped by political interpretation and, at times, exploitation of perceived public opinion and demands. Alongside the enduring belief among politicians and policymakers that strict and prescriptive conditionality is effective, this approach has also been maintained on the basis that it is assumed to be popular with the public. 

A more positive, social democratic view of the social security system might justify conditionality on the basis of the importance of reciprocity – ensuring a kind of collectivist accountability and maintaining public consent. But this risks conflating the underlying principle of conditionality with the need for a compliance-focused and often punitive system, while over-estimating public demand for such an approach.

As part of our recent research, we commissioned public focus groups and polling to dig under the surface of well-documented headline public support for conditionality. We found that people wanted there to be some mechanism of accountability in the system but could also recognise, and wanted to avoid, the risk that overly strict and prescriptive policing could undermine the quality of employment support and outcomes.

People tended to underestimate how strict and prescriptive the current system is. Polling respondents thought, on average, that people are required to spend 13 hours a week seeking work (when for most it’s 35 hours) and could spend three months seeking a job in a preferred field (when it’s just one month). Given the choice, they tended to opt for an approach that created as much space as possible for genuine engagement and a focus on getting people into good jobs – only invoking more strict and prescriptive conditionality if later deemed necessary.

Our findings suggest that there is political room for manoeuvre if a government wanted to take a more progressive approach, prioritising good quality voluntary engagement and using conditionality as a backstop rather than the default. If such an approach improved outcomes, as our research suggests it could, the public seem more than willing to accept a rollback of more strict and prescriptive conditionality.

The findings also imply an opportunity to make a positive, social democratic case for a more relational approach – maintaining that reciprocity is critical but will lead to better experiences and outcomes if it can be achieved through genuine human relationships rather than impersonal rules and process.

Improving engagement

The most straightforward way to test a more relational approach to employment support, built on genuine engagement rather than compliance, would be to start with those people currently exempt from conditionality due to disabilities and poor health. DWP currently does little to proactively reach out to this group, due to a tendency for officials and politicians to assume that people will only engage if there is some underlying requirement for them to do so. 

The previous government had planned to tighten the eligibility criteria for certain benefit categories so that more people would be subject to conditionality. The new Labour government is yet to announce whether it will take these plans forward. Meanwhile, an influential (and otherwise positive and welcome) report from Barnsley Council included the suggestion of a new “duty to engage” with an annual appointment.

Given the lack of trust between ill and disabled people and the DWP, and the fact that many in this group will be experiencing mental health problems, I’d argue that a shift towards more conditionality would not only be ineffective but often counterproductive. People are likely to hunker down in response, look to protect the financial support they receive, and feel much less inclined to try to genuinely engage with support.

Much more could be done to offer voluntary support, de-risk the prospect of engaging with this support, and begin to rebuild trusting relationships. In a recent policy paper, I set out how DWP could take a rapid, agile test and learn approach to designing and iterating a model of support that more ill and disabled people would engage with voluntarily and that would be more effective at helping people into good jobs.

Rethinking conditionality

Rethinking the current approach to conditionality should not be limited to resisting the siren call to apply prescriptive requirements to ill and disabled people. Our research draws on and adds to evidence that the approach for those subject to full conditionality is pushing many towards poor quality jobs and others towards seeking protection due to disabilities and health conditions. Many people who are out of work experience mental health problems, and interactions with a daunting and disempowering social security system can exacerbate these. 

Taking a more relational approach with everyone on universal credit, and looking to open up as much opportunity as possible for genuine engagement with support, could fundamentally alter perceptions and experiences of DWP and jobcentres. Ultimately, it could help more people into better jobs. But this will require the institutional instinct to prioritise compliance to take a backseat.

The first step would be to roll back the most prescriptive aspects of the current conditionality regime, such as being expected to spend 35 hours a week seeking work and only having one month to focus on seeking jobs in a preferred field before being required to apply for any job suggested by a work coach. These policies encourage performative compliance and movement into any job going rather than good jobs. Our public opinion research suggest they are also far from politically necessary.

But DWP should go further and trial an approach where conditionality is used as a backstop rather than the default. This could involve an initial period (possibly three months) where the focus is on fostering a trusting one-to-one relationship and genuine engagement with support. If there is sustained evidence of someone simply refusing to engage, some basic conditionality could be gradually phased in (but only for people not facing additional barriers such as disabilities and health conditions).

Such an approach would change the exam question for work coaches from “what is it reasonable to require this person to do” to “what will help to maximise this person’s engagement with support”. It would also allow for a much more personalised and flexible approach, with a greater degree of triaging. If someone seems motivated and able to find work, why not just have a check-in phone call after a few weeks rather than a mandatory jobcentre appointment after one?

A reality check?

Arguing for such a fundamental shift in thinking around conditionality, in the face of deeply entrenched practice and assumptions, I’ve often felt out on a limb, wondering if it would be more worthwhile to make the case for more modest and incremental reform. However, the occasional days that I still work in a mental health crisis team help to strengthen my conviction that such a shift is necessary.

The most effective appointments I have with people in crisis are those where we are able to establish a level of rapport and trust that facilitates an open and honest conversation and a willingness to take (often daunting) steps towards improved mental health. It is viscerally obvious in these situations that genuine human connection is the most powerful tool at a professional’s disposal when trying to help someone.

The key question, then, for any service concerned with helping people should be “how do we create the most favourable circumstances for positive and effective relationships to flourish?”. If there is deemed to be a need for recourse to some kind of coercion to manage risks of harm or ensure a degree of accountability, the aim should be to do this in a way that is sufficiently robust while having as small an impact as possible on the primary objective of genuine engagement. 

Many mental health services are still a long way from getting this balance right, but where progress has been made the benefits are clear. In contrast, within DWP and jobcentres the primary objective remains to ensure compliance and manage risk, with attempts to foster genuine engagement having to fit around the approach this entails. It’s time for a change.

Tom Pollard is Head of Social Policy at the New Economics Foundation