Will the UK's digital welfare reform project be able to tackle the complexities of human behaviour?

Derek du Preez Profile picture for user ddpreez June 19, 2014
Universal Credit has had a difficult time, but will changes to the way the system is being developed allow the digital version to overcome the complexities of combining a number of benefits into one?


The UK's Department for Work and Pensions has had a 'challenging' time with the roll-out of Universal Credit, its flagship welfare

reform project that aims to combine a number of benefits into one single payment. For those of you unfamiliar, Universal Credit was the only major IT project that was allowed to go ahead under the new coalition government, given that it has strong political backing from pretty much all parties, and was intended to simplify the complex benefits landscape of the UK's welfare system. However, problems with the project made the headlines last year after it was revealed that there were challenges with the system's flexibility and security and that millions of pounds worth of software was going to be written off as a result.

Universal Credit had been being developed by some of the world's largest suppliers (HP, IBM, Accenture, BT), but it was decided that this approach wasn't working and that the original development would be written off after five years and would simply be used in a number of pilot areas to test the reaction of claimants to the changes in policy. Instead, a new digital 'end-state' solution would be developed in-house and alongside the original system, which would be mobile first and would be the system ultimately rolled out to the whole of the country.

Last time I wrote about Universal Credit, I was questioning the secrecy surrounding the new digital project and asking why the Department for Work and Pensions was not releasing the business case it has put together for the 'end-state' solution. Its response is that the details are 'commercially sensitive', but I'd hazard a guess and say that the department is just sensitive about more bad press. However, the mainstream press has begun to push the issue again recently, after it was revealed that the UK's Major Projects Authority would not be commenting on the progress of Universal Credit in its annual review of transformative government projects. Instead of the 'amber' or 'red' rating that one might have expected, Universal Credit was being put into a 'reset' category – the first time this has ever been done and reflects the fundamental changes Universal Credit is undergoing. It is essentially being treated as a new project.

I also noticed earlier this week that Bryan Glick over at ComputerWeekly reported that the National Audit Office, the government's watchdog, is planning a second investigation into the system before the 2015 general election. This will likely send shivers down the spines of those working within Whitehall, given the damning nature of the previous one, which stated that “throughout the programme the department has lacked a detailed view of how Universal Credit is meant to work”.

So how is it meant to work?

The main problem facing Universal Credit is complexity and the diverse requirements of those claiming benefits. The idea behind the scheme is to make work pay. In recent years the government has been criticised for how the benefits system works, in that it often hasn't made sense for someone who is unemployed with a number of children, for example, to get a part-time job because when you take into consideration childcare costs and the benefits that would be cut as a result of new income, they would be worse off. By rolling all the benefits into one payment, and by making it less risky for people to take up work when they can by not seeing huge drops in their benefits, the government hopes people will transition back to the workplace. A recent government document stated:

“Universal Credit will improve financial work incentives by ensuring that support is reduced at a consistent and managed rate as people return to work and increase their working hours and earnings. People will generally keep more of their earnings for themselves and their families than is currently the case. 

“Universal Credit will also remove the distortions in the current system that tend to over-reward people for working a specific number of hours that may not suit them or their employers. Universal Credit will ensure that all amounts of work will be more financially rewarding than inactivity and remove the current barriers to small amounts of work.”

social media, social utility
This is all easier said than done, of course. I attended an event at the beginning of the year, which was looking at the complex policy requirements that the Universal Credit IT system is meant to support, and the general consensus from everyone in attendance was that this is really bloody difficult to do. Mainly because families are different and human beings aren't simple. They don't always follow a trackable path that you can map and easily build a system for – modern Britain isn't made up of families with 2.4 children. We're a diverse bunch, with many strange and wonderful definitions of family, and Universal Credit needs to cater to all of these – as you never know when you may find yourself out of work.

And although Universal Credit has been rolled out to a number of 'pilot' areas across the UK and the latest stats show that 6,550 people have started claiming on the system – the majority are men under the age of 25. They are also all non home owning, unemployed, single people without children – all requirements for the initial testing. The simplest of the simple claimants. This is being extended to couples over the summer and later on in the year will include families, which will be interesting to see how the original system copes with these changes in complexity.

Herein lies the problem – complexity and diversity needs to be established from the outset. And from what I'm hearing about the new digital 'end-state' solution, this is the plan. Instead of getting the simplest of claimants onto Universal Credit and scaling up from there and adding more diverse groups as things progress, the testing for the new system will look to address the challenge of complexity from the outset. In fact, a group of hard-working digital bods within DWP are going to try and introduce the new system with a sample of just 100 people, but those 100 people will represent all different possible types of claimant. Once the system can cope with these, it will then be scaled up in terms of quantity. This makes far more sense in my mind – pinning down and getting those most difficult cases right from the off will make scaling the system far more achievable. Because let's face it, there's no point in having a system with thousands of people claiming across the country, if they are all single and under 25.

Another bit of good news, I'm also hearing that the digital team working within DWP on the end-state solution is incredibly capable – not only in terms of their development skills, but with regard to being able to ignore all the political crap going on around them. Early signs suggest that the new solution could in fact do the job and that they are up to the challenge. 



  • Getting the complexity right will be the key to success for Universal Credit (this is a thought I've actually stolen from someone else, but I completely agree).
  • Human beings aren't predictable and we live complex lives, building a system that supports changes to a wide variety of peoples' circumstances isn't ever going to be easy. But rather focus on that first, instead of the number of people on the system. I would rather know Universal Credit can support all walks of life, rather than it can support 7,000 of the most simplest claimants.


A grey colored placeholder image