Universal credit ruling: a victory for welfare in Britain
A recent ruling by the British Court of Appeal sided with welfare claimants after a flaw in the income calculating system was revealed.
- A problem in the Department for Work and Pensions' automated benefits delivery system resulted in claimants being paid less benefits than they were entitled to
- The government refused to correct the problem on the basis of technical complications and cost
- The Court of Appeal found the government's failure to correct the position "irrational"
A unanimous ruling by the Court of Appeal confirmed earlier findings that the rigid monthly income assessment regime set up by the Department of Work and Pensions - the entity responsible for issuing benefits - to calculate the amount of benefits to be paid out resulted in significant and unwarranted cash losses to benefits claimants. The Court of Appeal found that the failure by the Department of Work and Pensions to rectify the issue was unlawful.
The case is a clear example of how automated welfare systems can, without effective supervision and timely human intervention, have adverse effects on the very population that they seek to serve.
Universal credit is an income-based benefit available to a segment of residents in England & Wales. Introduced in 2013 in an attempt to simplify the welfare system, universal credit rolled six distinct benefits - ranging from tax credits to housing support - into one, including unemployment support (Job Seekers Allowance) and income “top-up” support (Employment Support Allowance).
The issues with universal credit are structural and manifold. In addition to the unduly restrictive elegibility requirements, universal credit claimants must grapple with opaque decision-making and unclear review and appeal processes. The case before the Court of Appeal concerned an altogether different issue: the inherent shortcomings of the income assessment tool.
The foundational flaw in the award calculating system
Because universal credit is a means-tested benefit, the amount paid out varies depending on the income received by the claimant, which is assessed over a fixed monthly period. An automated computer programme then calculates the award on the basis of the reported earned income.
The claimants, four single working mothers entitled to universal credit, were not paid on the same day every month. Like many other paid employees in Britain, whenever their pay date conflicted with a bank holiday or a weekend, the claimants’ pay was brought forward. Because the monthly assessment tool was fixed and did not have the flexibility to account for these slight variations, two salaries could be captured in a single assessment period, artificially inflating the claimants’ monthly income. Accordingly, the claimants’ universal credit payments would be drastically reduced. As a result, the claimants would struggle to make ends meet during the months when their pay was erroneously double-counted, incurring significant rent arrears and relying on food banks.
Though conscious of the adverse consequences of the rigidity of the monthly assessment period tool on the claimants, the Department for Work and Pensions (DWP) failed to take action, in part because rectifying the system would be too costly. The DWP argued in court that the automated system could not be fixed without devising a new calculation system from scratch, which would cost at least £7.5m and require thousands of manual calculations.
A false dichotomy uncovered
The government’s case was that the structural problem could be resolved either by overhauling the entire computer programme, or by introducing a temporary fix in the form of a manual intervention for each claimant affected (as many as 85,000 according to estimations). In any event, any fix would be resource-intensive, and the cost of addressing the structural flaw would outweigh any benefits to the claimants.
The court decidedly rejected this argument, as it did not see any reason why the programme could not be modified to ensure that it was able to read two roughly equal payments received within a single month as salaries corresponding to two different assessment periods. The court found that the failure by the DWP to rectify the system was irrational, and therefore unlawful. Significantly, the DWP could not rely on the cost of the necessary adjustments to justify its inaction.
The scope of the ruling is clear: where an automated system falls short of legal standards, the government cannot hide behind the complexity of the programme or the cost of rectification in order to escape liability.
A chink in the austerity armor
Automation is a solution that governments are increasingly turning towards for the efficient delivery of welfare benefits. The Court of Appeal itself acknowledged its value in generating substantial savings in the costs of administration, thereby releasing more money to be paid out in benefits. The drive for efficiency is a key part of the over-ambitious austerity narrative: doing more, with a lot less.
However, as the ruling proves, automation and austerity are not always a match made in heaven. Automated systems can fall short of their cost-efficient promise if hastily designed - particularly where their flaws do not lend themselves to easy manual rectification - and result in higher expenditure than anticipated. The lesson from the ruling is clear: there is no price too steep to rectify the unlawful consequences of an automated system.
This case illustrates particularly well the limitations of tech solutionism: pushed to its extreme we reached this preposterous situation, where a government chose to put people's livelihoods at risk to avoid making changes to a technology too complex for them to modify. While technology is becoming part of our lives and part of the way governments manage welfare systems, we must be adamant that algorithms are there to serve us and to facilitate our access to welfare. Not the other way around.
The DWP has announced that it will not appeal the ruling.