Many public sector projects struggle to stay on time and on budget, and civil servants are under growing pressure to improve their game. Joshua Chambers asks the experts how government can avoid project management pitfalls.
Project management in the public sector can seem to defy the fundamental laws of accounting. Take just one example: the procurement of the RAF’s Eurofighter jets. They cost £3.5bn more than the £16.7bn initially budgeted, and the UK will receive a third fewer aeroplanes than initially ordered. That £3.5bn overspend alone could have purchased 13 hospitals, 140 academy schools, or – in defence terms – paid for the deployment of forces in both Afghanistan and Iraq during 2007-8.
So-called ‘major projects’ aren’t the only important ones, however. Good management is vital even in relatively small projects, because even quite small budgetary problems can distort the space around them, absorbing large quantities of time, staff and money from the rest of the department. As public money becomes ever scarcer, the need for good project management will grow still more acute.
Of course, there are numerous success stories, where public sector projects have steered clear of disaster and been completed on time and on budget. For example, the development of the Olympic park is now almost fully complete, creating world-class facilities and regenerating one of England’s most deprived areas, on time and under budget.
CSW has spoken to a set of project management experts, drawing out the key elements of good project management. Looking broadly at the best approaches, and then learning from three markedly different schemes – the Olympics site; the development of the National Offender Management Information System; and the procurement of the Eurofighter – we’ve identified both some crucial elements of best practice, and some of the most dangerous traps that lie in wait for public sector project managers.
Where do I start?
For 20 years, until his retirement in 2008, Sir John Bourn headed the National Audit Office (NAO), analysing projects great and small, triumphs and disasters. Good project management is mainly just common sense, he says – but while the key questions to ask at a project’s inception may seem obvious, they are often forgotten when the pressure is on to deliver.
First, he says, the feasibility of the proposed project must be assessed. The latest NAO guide, Initiating Successful Projects, has a simple checklist of specific questions to ask throughout a project. For example, are the project’s aims realistic? Some past public sector projects have had overblown expectations of the capabilities of mooted IT systems. And are the desired outcomes of the project understood in detail?
This latter question is particularly important, because there are often numerous interest groups involved in a project, and not all of them will want the same outcome. Tim Banfield is a director at the NAO, and wrote the guide to initiating projects; he explains that “you’re going to have ministers with one set of expectations; officials who might have a different set; most projects use large commercial organisations, that also might have different expectations; and there’s a fair chance that the end user of the project will have their own, as well.” While a project manager can never realise the hopes of all those involved, Banfield recommends managing expectations in advance and consulting with all stakeholders before starting out.
If the project’s aims and approach are realistic and sensible, the cost and budget then need to be discussed. Banfield warns that public sector approval systems encourage project managers to under-estimate the costs. “You get what you could term ‘entryism’,” he says. “If there is budget capacity for a small amount of money, and you can start to demonstrate that your project will cost that, the chances are that you will get it approved.” However, he warns that the budgetary consequence of over-optimistic cost projections is that departments will have to juggle ongoing projects, either delaying them and thus increasing costs, or cutting their scope and weakening their impact. The NAO guide therefore states that “successful delivery requires strong leadership, setting the tone by encouraging honesty in estimates, challenging optimism bias and assumptions, and being willing to stop projects which no longer make sense.”
While it is never pleasant to dwell on failure, project managers do need to be absolutely clear about the risks of a project: what could go wrong? Not all risks are unacceptable, but there should always be a contingency fund designed to absorb manageable risks that do go bad. The dangers can be tested and minimised by running pilot schemes, and both officials and ministers should understand what could go wrong and make a plan for handling problems that arise.
This might, in a worst case scenario, involve scrapping the project: too often, public sector projects have continued long after they’ve evidently failed, mainly because ministers and officials don’t want to be the ones in charge when such dysfunctional schemes collapse. However, the government wants to give doomed schemes the coup de grace at an earlier stage. Speaking at a recent seminar on project management run by CSW and sponsored by HP, the deputy director of the Cabinet Office’s new Major Projects Authority (MPA), Steve Mitchell, said that “many projects have lingered on the brink in the past, but there’s not an appetite for these projects to linger any longer. If there’s a valid case, then the case needs to be made to ministers to continue; if not, move on. Non-viable projects will be terminated.”
How do we deliver this?
If a project makes sense, then it’s time to start thinking about how to deliver it. That means planning who will deliver it: who should be on the team, and who will be in charge? The commercial director of the Department for Work and Pensions, David Smith, is a member of the Major Projects Review Group – a body that must approve all proposed major projects and can question project managers up to three times a year – and has responsibility for examining projects’ commercial elements. He says he’ll be looking to ensure that the team includes people with “a proven experience” of dealing with similar procurements, or commercial expertise that will provide an understanding of what the market can deliver. A lack of technical understanding has led to a number of IT projects going wrong, including the National Offender Management Information System (see below).
Once the delivery options have been examined, the department will need to decide what to buy and how to buy it. It may, for example, adapt an existing supplier contract, or hold an open competition. Aileen Murphie is a director at the NAO, and says that “there’s nothing wrong with using an existing contractor, provided you use some way of injecting competitive tension into the way you’re letting the contract”. There must be suitable incentives for good contractor behaviour, and a framework that provides flexibility in case project requirements or technological capabilities evolve. Banfield says that around 35 per cent of PFI contracts cost more than expected because the scope of a project changes and the contract isn’t sufficiently flexible.
Bourn warns that “often, firms have been rather more adept at thinking about contracts than the government side.” For example, some contracts allow for costs to rise if the price of raw materials or labour increases. These provisions are normally based around a cost index, perhaps measuring the price of a number of goods – and designing these indexes can prove fiddly, providing scope for later contractual disputes.
Finally, public sector managers must plan how the delivered project is operated. Who will run it, and are suitable governance and performance-management systems in place? Banfield cites “the 20/80 rule”, pointing out that “the last 20 per cent of the project probably puts 80 per cent of the risks in.”
Take your time
Few projects are easy to manage, but public sector projects have an added complication: politicians. Bourn warns that planning “often takes longer than ministers want to allow,” because ministers are thinking of the next election and want to see their ideas put into practice by then. Project management manuals (and newspaper articles) may set out what needs to be considered, but “in real life, there’s a pressure from ministers and senior civil servants to get on with it. When people down the line refer to the need to plan it all out, there’s a disposition to say: ‘Oh, that’s just bureaucracy. For God’s sake get on with it!’”
Bourn cautions strongly against taking shortcuts. “Nearly always, you find that more time and money is required up front to plan than is allowed. Be prepared to spend time and money putting arrangements in place before you put the spade in the soil,” he says.
Once a project is under way, there may be new pressures – often to cut costs. Such cost-cutting often leads to a scheme’s specifications being changed mid-project, says Bourn, but this is one of the most common reasons for project failure: “You get new ministers and new people at the top, and they want to alter it. There’s no better way of raising costs and causing delays than to change what you’re doing – ‘horses in mid-stream’ is the relevant proverb.”
Don’t panic
Speaking at theCSWseminar on project management, Dr Harvey Maylor, the director of the Project Management Centre at Cranfield University, said that the success rate on major projects has remained stubbornly low, at about 30 per cent. Yet while the two horror stories among our case studies are all too real, this feature isn’t intended to scare. The government wants to improve project management, tightening up decision-making processes to ensure that projects start off on the right track.
The MPA is set to play a key role here, and its head David Pitchford explains that a Major Projects Leadership Academy will be launched later this year to train civil servants. “Ineffective or inadequate leadership is a key issue inhibiting greater rates of success,” Pitchford says. The academy will focus on improving leadership skills, plus technical and commercial understanding.
Meanwhile, Smith wants civil servants to see the Major Projects Review Group as a helping hand rather than a hurdle. Civil servants might appear before the group up to three times a year, but an appearance shouldn’t be seen as an ordeal. “I’m here to answer questions,” he says. “They should come to us with as many questions as they do answers.”
If you’re confident of your project, proceed with confidence, and be ready to battle for the time and money you require. While it’s an inevitable reality of public service that poorly-managed projects receive more publicity than good ones, successful projects will be recognised and praised inside government. After all, by saving the taxpayer £910m, the ODA has released enough cash to purchase three hospitals. Good project management: it can save lives.
The good: London Olympics 2012
A fine example of a successful project is the building of London’s 2012 Olympic Park. Over 90 per cent has so far been delivered exactly to schedule. It must be noted that, like many public sector projects, an inaccurate budget was submitted to the Treasury when it was initially proposed – and the revision of the budget will have taken resources from other areas. However, once the bid to host the Olympics was won, the newly-established Olympic Delivery Authority (ODA) set about ensuring that an accurate budget was secured. This budget has stayed the course.
The chief executive of the ODA is Dennis Hone, who took over the role this year and was previously finance director. He explains that there were a number of “tough conversations” with the Treasury to secure the appropriate level of budget, and it took time to determine the right sum. “We won the bid in July 2005, the ODA really got rolling by about April 2006, and Tessa Jowell [then Olympics minister] made the budgets announcement on 15th March 2007.” To date, though, the ODA has spent below its budget. The most recent NAO progress report says its final expenditure will be £7,189m, significantly less than the £8,099m forecast.
Early on in the process, the organisation recognised that it would need a wide range of skills, and commissioned a consortium of major contractors to help manage the project. This allowed the organisation to access the expertise of the contractors immediately, and the contract allows the ODA to hire specialists from the consortium for very short periods; these could be land decontamination experts one week, and IT specialists the next.
The ODA also retains its own team of project managers in order to guarantee its capabilities as an ‘intelligent client’, and they considered its delivery contracts very carefully. Incentives between contractors and subcontractors were aligned, so some weren’t trying to build as cheaply as possible while others were aiming for architectural awards. Other provisions ensure that “the contractor must notify if there are delays, or if they need additional resources,” says Hone. This enables the ODA to manage its contractors’ work, ensuring that any delays in one part of the system don’t feed through into additional costs for other contractors. “It’s no good getting to the end of a job and finding that people are three months late and planning to claim against you for delays,” Hone points out. Performance bonuses are paid for delivering on time and budget.
There is a huge amount of pressure on the ODA, and a large number of interest groups to reassure. Hone says that the ODA regularly publishes its performance against set targets, to prevent unnecessary questioning and pressure. It also specifies exactly what it will and won’t deliver, ensuring that stakeholders can’t try to get the ODA to deliver extra projects.
The bad: National Offender Management Information System
Public sector IT projects have developed a bad name. There have, of course, been many success stories – from London Underground’s Oyster card, to the Environment Agency’s fishing rods project. However, it’s fair to say that there has been a tendency for public IT projects to become over-complex.
A good – or rather, bad – example of this is the National Offender Management Information System (NOMIS), designed to streamline offender management across the probation and prison services (NOMS).
Aileen Murphie audited the NOMIS scheme, and explains that the problems started right at the beginning because of a lack of understanding and awareness within the project team: “They didn’t really have a clear idea of the scope of the project, which was enormous.” Further, they decided to purchase unnecessary technical capabilities that were expensive to provide – asking for real-time access to offender files, for example, when that isn’t required for the vast majority of offenders.
Rather than tendering the project afresh, it was decided to use existing contractors. This enabled the team to move forward without having produced a fully worked-up business case and a proper budget, so an appropriate framework wasn’t put in place and the scheme was managed on a cash basis, year by year.
Another major error was failing to adapt the business to cope with a new system. “Usually, an IT system isn’t just dropped into an organisation which then stays the same around it. Generally, what you’re talking about is an IT-enabled business change of some sort, and sometimes they can be massive,” Murphie explains. At NOMS, however, the business wasn’t altered to take full advantage of the emerging IT system.
Overall, NOMIS was supposed to cost £234m, but ballooned in price to £690m until it was stopped in 2007 and a re-scoped programme was begun that cost £513m. Due to delays, the affected public sector organisations started to create their own solutions, rendering some of the scheme unnecessary.
The ugly: Eurofighter procurement
As procurement projects go, the Eurofighter is a big one: the government is buying 160 planes, in its biggest current military aircraft purchase. But the cost of each plane has risen by 75 per cent, and consequently the Ministry of Defence has had to cut its order by a third just to keep costs under control. What’s more, the Public Accounts Committee has expressed fears that the maintenance costs during the aircrafts’ service could be up to £3bn more than originally anticipated.
The NAO’s Tim Banfield audited the Eurofighter programme, and explains that the costs of the programme were significantly under-estimated from the start. With complex projects there will always be some uncertainty, he explains, but use of historical-trend analysis as well as bottom-up estimates – trying to gauge the cost of each component and adding them together – provide useful methods of estimation.
Unfortunately, costs spiralled further because there was no central governance for the project; instead, the management was shared by four countries, each with their own objectives and a need to bring work to their own businesses. Some countries wanted to use the project to boost their manufacturing base, says Banfield, rather than build the aircraft as cheaply as possible.
Banfield says: “If this had been a UK-only programme, we would have had some underestimates of cost, without a shadow of a doubt, but it would probably have cost far less than the extra complication put in by saying: ‘Not only have we got to do all this, but we’ve got to share everything between four countries and four sets of industries – and not all of the industries have the same capability’.”