Government has reduced the number of major projects it manages, but – given the need to prioritise what it does – there are still too many. The cost of projects is increasing, and confidence that they will be delivered successfully is in decline. Infrastructure projects receive better ratings than information and communications technology (ICT) or transformation projects. Some departments – notably MoD – have more, and more costly, projects than others.
The Infrastructure and Projects Authority (IPA) has oversight of a number of major projects, which departments manage alongside their day-to-day operations, and awards each a delivery confidence RAG (red/amber/green) rating. These projects range from updating computer systems at HO to building high-speed rail systems at DfT.
The Government has reduced the size of the major projects portfolio, but more work needs to be done.
The IPA has reduced the number of projects it oversees, from 188 in 2015 to 143 in 2016. However, the chief executive of the civil service, John Manzoni, has said that the civil service is ‘doing 30% too much to do it all well’. Crudely, taking 2015 as a baseline, this would mean the IPA overseeing 132 projects worth £342bn, and not 143 projects valued at £405bn. The rate of decrease in 2016 has been particularly great, but also reflects the fact that 45% of new projects in 2015 left the portfolio after one year – a far higher proportion than in previous years. (Most projects leave the portfolio because they have completed or become business as usual, although the IPA does not publish the reasons on a project-byproject basis).
The total value of projects being managed (excluding those that are exempted) has also decreased (from £489bn in 2015 to £405bn) but remains higher than in 2013 (£354bn). The average value of a project in 2016 was higher than in any previous year on record: £2.83bn as against £2.6bn (2015), £2.45bn (2014) and £1.85bn (2013). New projects in 2016 included £29bn on a new model for managing the Sellafield nuclear decommissioning site and £7bn on MoD’s Future Beyond Line of Sight project (which covers satellite networks).
The proportion of at-risk projects is increasing.
The proportion of projects rated red and amber/red across the portfolio, in terms of confidence as to whether they will be delivered on time and within budget, has been increasing since 2013. In 2016, 31% of projects were rated red or amber/red. This may be due to inherently riskier new projects entering the portfolio; indeed, new projects in 2016 were riskier than those in the previous three years, with 43% of new projects being rated red or amber/red.
However, confidence in projects that have been in the portfolio since 2013 has decreased for the first time. Although the percentage rated red or amber/red has held steady, the percentage rated amber has increased and that rated green has decreased. This may be because the portfolio size has decreased, leaving more complex projects behind.
Of the two projects in the portfolio since 2013 that were red-rated in 2016, one – MoJ’s Electronic Monitoring (tagging of offenders) – has declined from amber in 2013, while the other, DfT’s Shared Services Futures (centralising functions such as finance, HR, payroll and procurement), has been red since 2013. The low confidence ratings in projects in the portfolio since 2013 suggest the IPA’s assessments are not succumbing to grade inflation. However, most projects rated red or amber/red in 2013 have improved over time (five are amber/green, eight amber, four amber/red and one red).
ICT and transformation projects are rated as riskier than infrastructure projects.
In its annual report, the IPA breaks down projects into four different types: transformation, ICT, military and infrastructure. These categories are only indicative and have changed over time, but are nonetheless useful. Nearly half of ICT projects were rated either red or amber/ red, compared to only 31% of the 143 projects in the portfolio as a whole. ICT projects included HO’s Immigration Platform Technologies and Digital Services at the Border, both of which were rated amber/red (and could be affected by Brexit). Confidence in infrastructure projects being delivered on time and within budget is much higher: none of this group was rated red in 2016.
Major projects are more concentrated in certain departments…
MoD (34), DH (21) and MoJ (16) have more projects than any other departments, which is broadly similar to the pattern in 2015. While these departments have some red or amber/ red projects, others – such as HMRC – with a relatively small number of projects, have a high proportion that are low-confidence rated. DCLG and HMT have no major projects in 2016.
…and some departments have many low-cost projects, while others have fewer but at much greater cost.
In 2016, MoD managed 24% of the total major projects portfolio by number, but 32% of the total by cost, the largest of any department. Similarly, DECC managed only 4% by number but 23% by cost, largely because of the costs related to nuclear generation at Hinkley Point and decommissioning at Sellafield. DfT also manages more of the portfolio in terms of value than number of projects. Some departments, then, may not have experience of running many major projects but are responsible for projects worth billions of pounds.
Turnover of major project leaders is high but in decline.
Keeping the right people in place is essential to successfully delivering projects. Turnover of senior responsible owners (SROs) of major projects has declined between June 2013, when one in four SROs changed in the previous quarter, and late 2015. However, turnover of project directors remains around one in eight per quarter. It is unclear whether high SRO turnover causes low RAG ratings, or results from them.
Data on turnover since 2015 could help clarify this point. The IPA started publishing quarterly updates on SROs, but has not published any updates since January 2016 (with data as of September 2015). The IPA’s publications offer good examples of transparency – the annual reports are welcome, and show that performance data is not something government should be afraid of – while also having room for improvement (some data is not published openly in the reports, such as reasons for projects leaving the portfolio, or for the classification of certain projects).