5Oversight
The review process
41.The PAR framework is designed to afford flexibility in scope and timing. In some of the reports we considered, however, that flexibility appeared to contribute to inconsistency of review approach and poor communications between the review and programme teams. The June 2013 PAR was conducted just three weeks after Howard Shiplee took over as SRO of the programme. He argued that the PAR was “ill-timed” as he had not had time to convert the blueprint into a delivery plan. The MPA had “insufficient evidence” to assign a new DCA to the programme in their report.133 The then UC Programme Director, Ann Harris, argued that the February 2014 PAR was “too short” to enable the MPA to make a full assessment of improvements in the programme.134
42.The scope of the March 2017 AAP was a source of contention between Neil Couling, the SRO, and the IPA. While Neil Couling favoured a narrow focus on the readiness of the digital service to be rolled out to 30 Jobcentres in July 2017,135 the IPA also reviewed progress in implementing wider recommendations made in its September 2016 HC.136 Signing off the report two months after the review, Neil Couling noted that reviewing a programme the size of UC in a two day period was “very difficult” and noted that it can come with an “urge to make superficial, pseudo-definitive point of time judgements and simplistic recommendations”. He said the IPA deserved “credit for attempting to swim in these shark-infested waters”.137 It was not clear why such an important review was only afforded two days. The next and most recent review, the September 2017 HC, focused solely on readiness to move to scale at the behest of the Department. It was explicitly agreed between the DWP and IPA that it could not consider whether:
- previous IPA recommendations had been acted on;
- UC would achieve its business case; or
- UC was delivering its policy intent.138
43.Recent IPA reviews of the UC programme have been linked to specific scaling events or business case processes. Delays in the programme have meant delays in oversight. The October 2015 PAR recommended the next round of formal assurance take place in Spring 2016.139 In the event, due to delays in the rollout, it was postponed until a HC in September 2016. The March 2017 AAP expected a PAR, to inform decisions on the full business case (FBC) for UC, in September 2017.140 The FBC was, however, delayed and is not expected until March 2018. The HC of very limited scope was conducted instead.141
44.The Government’s approach to oversight of major programmes is flexible, enabling each review to be tailored to the circumstances. This has its advantages, but it has made tracking the progress of UC more difficult. Key concerns moved in and out of scope, at the behest of the DWP programme team, and some important recommendations were not followed up in detail. We were surprised to learn the latest review, which considered the readiness of the digital service for accelerated rollout in late 2017, was explicitly excluded from considering whether previous IPA recommendations had been acted on, whether UC would achieve its business case, and whether it was delivering its policy intent. Furthermore, delays to the UC business case process mean that by the time of the next PAR, the UC programme will not have been subject to that level of IPA assurance for over two crucial years of its development. We expected UC to have been subject to more recent IPA oversight.
Business case and transparency
45.Major programmes follow a three-stage process for securing business case approval and funding from the Treasury. The first, the strategic outline business case (SOBC), was approved after some delay in September 2014.142 The programme had also reached this stage in 2011, before returning to square one with the reset in 2013.143 The second, the July 2015 outline business case (OBC), was approved in November 2015, informed by the positive October 2015 PAR. This secured funding for the programme until the end of 2017. The third stage, the full business case (FBC), has not yet been submitted. Since December 2017, the programme has been funded by the Treasury on an ad hoc basis, with the latest extension being until April 2018. The Department expects the FBC to be approved in March 2018.144
46.The July 2015 OBC forecast net benefits of UC of £27.1 billion. It did not, however, incorporate substantial policy changes made to UC in the November 2015 Spending Review and Autumn Statement.145 By the time it was approved, it was out of date. Since then improvements have been made to the design of UC, including at successive Budgets, and the programme has been further delayed.146
47.The business case includes gains to society of £12 billion from increased employment and £16 billion from redistribution (as people on lower incomes value a given increase in income more than those on higher incomes).147 Both those figures are highly sensitive to small changes in underlying assumptions.148 The DWP’s latest employment impact research, published in September 2017, showed that, compared to a control group of comparable Jobseeker’s Allowance (JSA) claimants, people on UC were four percentage points more likely to have been in work at some point within the first six months of making their claim.149 That compared with eight percentage points in the Department’s previous analysis, published in December 2015.150 Both those analyses, however, only covered single, unemployed claimants without dependent children. It quite possible that the employment impact of UC for people with more complex circumstances, who may face greater barriers to work, will be different. The then Secretary of State told us in November 2017 that the DWP “remain committed to producing robust comparative analysis of the employment impacts of Universal Credit” for couples and families.151 Neil Couling told us that it was “entirely possible” this would be available in early 2018.152 In the meantime, the employment component of the business case remains highly uncertain. The NAO cautioned that “this uncertainty is magnified” because the distributional benefits of UC are closely correlated to its employment benefits.153
48.The DWP has consistently been unable or unwilling to share statistics with us regarding the functioning of UC.154 Similar concerns were evident in the IPA reviews.155 In his sole comments on the September 2017 HC, Neil Couling quoted the Living Bible: “If you wait for perfect conditions, you will never get anything done”.156 Mr Couling told the Chair of the PAC that UC “still represents a value for money programme” and that, in any revision of the business case, its net present value “will remain substantially positive”.157 In scrutinising the UC programme from the outside there is, however, a danger we are being asked to take too much on faith.
49.In the eighth year of the programme, a full business case for UC has yet to be submitted. There is a need for robust statistical analysis to show whether the improved employment outcomes revealed in previous impact studies have been maintained for more complex claimant types. The effects of a slower rollout timetable and delays to automation on projected efficiency savings are also far from clear. Public, parliamentary and governmental scrutiny of this major reform would be better served by a more transparent approach by the Department. Given its confidence that the programme is on track, the DWP would also benefit from greater openness.
Future of Universal Credit
50.The UC policy changes announced in the November 2017 Budget were major improvements which we hope will make a big difference to the lives of many claimants. UC is still far from perfect and we will continue to consider how policy can be best improved. The IPA reports provide insight into the interaction of policy objectives with the challenges of running a huge change programme. While the UC programme has come a long way since it was reset in 2013, some of its biggest challenges—such as delivering an automated online service on a national scale—are still to come. Examining those concerns will form an important part of our ongoing work.
133 IPA, UC Project Assessment Review, June 2013, p8
134 IPA, UC Project Assessment Review, February 2014, p4
135 In the event, it was 29 Jobcentres.
136 IPA, UC Assurance of Action Plan, March 2017, p2
137 IPA, UC Assurance of Action Plan, March 2017, p4
138 IPA, UC Health Check, September 2017, p4
139 IPA, UC Project Assessment Review, October 2015, p16
140 IPA, UC Assurance of Action Plan, March 2017, p18
141 IPA, UC Health Check, September 2017
142 NAO, Universal Credit: progress update, HC (2014–15) 786, November 2014, para 7
143 NAO, Universal Credit: early progress, HC (2013–14) 621, September 2013, p40
144 Letter from Neil Couling to Meg Hillier, 18 October 2017
145 PAC, Nineteenth Report of Session 2015–16, Universal Credit: progress update, HC 601, para 1
146 Letter from Neil Couling to Meg Hillier, 18 October 2017. Flexibilities were also introduced for adjustments to UC to be made by devolved administrations in Scotland and Northern Ireland.
147 IPA, UC Health Check, September 2016, para 5.4.1. This works the other way around too - a given loss of income is felt more keenly by people on lower incomes.
148 NAO, Universal Credit: progress update, HC (2014–15) 786, November 2014, para 1.18. Appendix 6 (p59) gives an indication of the sensitivity of the business case to changes in assumptions.
149 DWP ad hoc research report no. 53, Universal Credit Employment Impact Analysis, September 2017 (based on claims made between July 2014 and April 2015).
150 DWP ad hoc research report no. 28, Estimating the Early Labour Market Impacts of Universal Credit: updated analysis, December 2015 (based on claims made between July 2013 and September 2014).
151 Letter from David Gauke to Frank Field, 13 November 2017
153 NAO, Universal Credit: progress update, HC (2014–15) 786, November 2014, para 1.18
154 See, for example, Work and Pensions Committee news story, 18 October 2018
155 For example, the February 2014 PAR was “surprised” by a labour market performance measures and a lack of statistical evidence on the effects of UC. See IPA, UC Project Assessment Review, February 2014, p14–15.
156 Ecclesiastes 11:4. IPA, UC Health Check, September 2017, p7
157 Letter from Neil Couling to Meg Hillier, 18 October 2017
29 January 2018