6 There have been a number of welcome developments to address the issues identified by us and the Committee of Public Accounts. These include:
• improvements to accountability with greater clarity about the roles of senior responsible owners (paragraphs 3.15 to 3.16);
• investment by the Authority and departments to improve the capability of staff to deliver major projects, with departments reporting to us that they are seeing benefits from these initiatives (paragraphs 3.1 to 3.8);
• increased assurance and recognition of the role that assurance plays in improving project delivery (paragraphs 3.9 to 3.12); and
• initiatives such as one-day workshops before HM Treasury approval to prevent departments from getting locked into solutions too early (paragraphs 3.13 to 3.14).
7 However, challenges remain and new challenges have emerged which could undermine or lessen the impact of these initiatives. For example, turnover of senior responsible owners has been high, with only 4 of the 73 programmes that had been in the Portfolio for 4 years having had a single senior responsible owner during that time (paragraphs 3.17 to 3.18). Likewise, there are still concerns around shortages of skills in specific areas such as risk management and behaviour change, which is worrying given the volume of transformation projects being undertaken (paragraph 3.4).
8 It is too soon to see the impact of some initiatives. For others, the impact seems to be variable.
• In some cases, changes have only been introduced relatively recently, as in the case of the one-day workshops prior to approval.
• In the case of assurance, the Authority has not yet established a link between review recommendations and project performance. Our earlier reports showed that departments' responses to assurance recommendations varied, being positive on Thameslink and Crossrail, but slow in the early stages of High Speed 2 and Universal Credit, for example (paragraphs 3.11 to 3.12).
9 The Authority does not publish full information on the size and cost of the Portfolio. The published whole-life cost is lower than the total cost because the Cabinet's transparency policy exempts some data from disclosure. However, both the aggregate and disclosed costs were higher in 2015 than in 2012. This is largely due to changes to the composition of the Portfolio, more of the costs being disclosed, the inclusion of costs previously unknown or uncertain and changes in methodology. There is still uncertainty about costs and it is reasonable to expect that the value of the Portfolio will change further (paragraphs 2.4 to 2.6 and paragraph 2.11).
10 Without reliable and consistent measures of project success, it is difficult to state whether performance is improving. Our previous reports highlighted progress on making information about major projects transparent in the Authority's annual report. The most recent annual report gives a more complete picture, but there are still a number of issues which make it difficult to form conclusions about trends in performance across the Portfolio. These include: the amount of turnover in the Portfolio; the limited data published by departments; inconsistent reporting of costs, with some departments reporting costs in real terms and some in nominal terms, or others using different index years; and because there is no systematic monitoring of benefit realisation (paragraphs 2.1 to 2.3, 2.6 and 2.7, 2.17 to 2.19).
11 However, delivery confidence, which the Authority does measure, shows a mixed picture with high risks to delivery in the next 5 years. The number of projects where successful delivery was in doubt or unachievable unless action was taken (rated red and amber-red) has increased since 2012 as more risky projects have entered the Portfolio. These ratings reflect uncertainty and risk as well as the performance of projects. Uncertainty reduces through the project lifecycle, but our analysis shows that not all project ratings improve over time. Of 56 projects which remained on the Portfolio from 2012 to 2015, 17 had red or amber-red ratings in June 2015 compared with 12 in 2012, although the number of projects considered highly likely to deliver on time and on budget (rated green or amber-green) also increased from 16 in 2012 to 25 in 2015. Of particular concern is that 35% of projects due to deliver in the next 5 years are rated as red or amber-red (paragraphs 2.12 to 2.16).
12 This is worrying given the number of projects to be delivered within this Parliament, and the large proportion of these which are transformation projects. Our report on lessons learnt from welfare reform commented that the Department for Work & Pensions took on an unprecedented number of reforms, and any large portfolio was likely to experience problems. Nearly 80% of the Portfolio projects due to be delivered by 2019-20 are to either transform or change the way that services are delivered or accessed. We commented in our briefing note on Lessons for major service transformation that transformation programmes can present the greatest risk of failure and that there is a need to balance ambition and realism in setting goals. For instance, the Better Care Fund, which is within the Department of Health's portfolio, was a challenging initiative which ministers paused and redesigned after the early planning and preparations did not match its scale of ambition (paragraphs 1.6, 1.11 and 1.12).
13 Moreover, progress in improving portfolio management is disappointing. There is increased assurance through the Portfolio and other central departments have an increased role in assuring, approving and improving quality of delivery. But an effective mechanism still needs to be developed for prioritising projects across government or judging whether individual departments have the capacity and capability to deliver them. We have reported in the past on the difficulties caused for government projects by unrealistic expectations and over-optimism. The National Infrastructure Commission will make recommendations about future priorities for infrastructure. But there is also a need to prioritise transformation projects where, as for infrastructure, the impact of change may be felt in other parts of the system or on other projects (paragraphs 1.6 and 1.8 to 1.10).