The roof on the main building of Heathrow’s Terminal 5 was completed last month. Given that at 400 by 150 metres, this is the largest single span roof in Europe, it was a tricky operation. The six sections each weighing 2,500 tons had to be jacked up over a 10 month period.
To minimised any chance of mishaps, BAA, the company which owns the airport and was privatised in 1987, did a dummy run in Yorkshire to see whether the concept was feasible. According to Mike Forster, the BAA’s development and design director, the test in Yorkshire, which cost £2.4m ‘saved three months work on the Heathrow site and taught us all kinds of things about tolerances and sequencing’.
This operation was illustrative of BAA’s meticulous approach to the whole T5 project which is informed by its overall approach to risk. Rather than taking a conventional approach by trying to pass on risk to the myriad of contractors involved in this enormous project, from the outset BAA took the key decision of saying that it would take on all the risk.
This is in sharp contrast to conventional contracts which attempt to pass on risk – or rather, as Mathew Riley, the company’s commercial director for Terminal 5, ‘the financial consequences of risk’ – to contractors. BAA realised that such contracts do not really work because ultimately either the risk would always fall back on itself, or the attempt to pass on risk would result in massive legal claims and counterclaims.
The novel approach to risk, set out in a single contract called the T5 agreement to which all contractors have to sign up, is an attempt to avoid the failings of other massive projects. When the company was looking at how it would approach the task of building the terminal, which including tunnelling, moving two rivers, constructing the link roads and various bits and pieces will cost a total of £4.2bn, the project team studied the failings of various major schemes in the public sector such as the Jubilee Line Extension, the Scottish Parliament and the Millennium Dome.
The company had, of course, a long time to do so since the public enquiry started in May 1995, after two years of preparation, and heard evidence for a staggering four years. It cost government, the local authority objectors and BAA in excess of £100m and the decision which, as widely expected, agreed to the construction of the terminal, with various conditions, was eventually announced in November 2001.
Indeed, all the numbers to do with T5 are mind-boggling. The site is 260 hectares, the size of London’s Hyde Park. Touring the site, among the 4,500 workers, one gets a real sense of excitement. There is a ‘wow’ factor as the sheer scale of the enterprise and the complexity of an airport terminal with its baggage systems, security requirements and, of course, plenty of space for retail become apparent. Each of the five floors is big enough to house ten football pitches and there will be 131 escalators and 178 lifts. The new terminal alone costs £1.2bn and is designed to handle 30 million passengers per year, which, on its own, would make it the fourth largest airport in Europe.
The team setting up the project was most concerned to avoid the pitfalls of the other projects they studied. The Scottish Parliament, for example, had increased in cost from the unrealistic £40m to £400m. The cost of the West Coast Main Line upgrade also jumped by a similar factor while the Jubilee Line Extension opened two years late and at a cost of £1bn more than projected.
Mathew Riley says that BAA simply could not allow a similar disaster to happen to the company which is, of course, a regulated industry and therefore vulnerable to cost increases since its income is essentially fixed. Its funding is determined by five yearly reviews of landing charges by its regulator who allows BAA a set rate of return and has also ensured that BAA has sufficient income to fund BAA. ‘The regulator allows us a certain rate of return. But to satisfy our shareholders, we have to beat that’, says Riley. Massive cost overruns or long delays to T5 would have wrecked the company’s reputation and sent its share price plummeting.
If the performance of T5 had been in line with those schemes in terms of timing, cost and safety, according to Riley, ‘it would have been 1-2 years late in completing, been 40 per cent above original cost estimates and we would have killed six people.’ So BAA looked in detail at why these projects had not met their targets.’
It was not technical competence which was lacking in these failed projects, but their organisation and management, made worse by the fact that the parameters were often not properly defined. Rather than attempting to assess and define risks in advance, they tended to be dealt with after the event leading to massive claims and counter claims. And as Riley points out, ’40 per cent of the cost of claims are the legal expenses’.
At the heart of the terminal 5 agreement is the concept that BAA retains the risk while suppliers work as part of an integrated team to mitigate potential risk and achieve the best possible results. Teams of people from different companies work cooperatively on the endless series of smaller projects that make up a massive scheme like Terminal 5. Instead of handing over a set amount to each contractor for risk, which often ends up as profit, these project teams are allocated a small contingency which, if unspent, is then available for another team. Moreover, rather than simply handing over the work to these project teams, of which there are around 20, BAA has an active management role in each one as part of the team. These teams act as ‘virtual companies’ responsible for the task and working to the overall milestones of the Terminal 5 project.
If there is a failure by a contractor – say, as a minor example, a ceiling needs to be replaced – then the work is carried out again with no blame: ‘If the ceiling has to be done a second time, then the team will pay the cost with no profit margin. If it had to be redone a third time, then the cost would be down to the particular contractor’. Essentially, it is a no blame culture aimed at getting the best approach through cooperation, rather than the conventional adversarial approach.
Assuming the risk does not mean that BAA lets the contractors do what they want, knowing that they will not get the blame. Quite the opposite. It is precisely because BAA assume the risk that it has to ensure that all its suppliers and contractors provide services to the highest possible standard. According to Riley, ‘the agreement is predicated on best practice. That is the minimum standard’, though he accepts it is not always easy to define what that is in reality.
Taking on the risk does not mean, as BAA’s leaflet on the agreement points out, that there is an open cheque book for suppliers. It just means a sensible managed approach to risk, rather than any attempt to pass it on regardless of the cost and consequences: ‘BAA has chosen to manage this programme and accept the risk rather than contract the programme out to a company to manage for us – a more common approach’. Indeed, there is no main contractor, unheard for a project of this size. It is a very open process with the books of the suppliers being made available to BAA, a very different relationship than in many large public sector projects where companies hide behind considerations of ‘commercial confidentiality’.
A further benefit of the co-operation at the heart of the project is that it gives suppliers opportunities to work together to reduce costs through buying in bulk. Riley cites the example of low-voltage electrical switchgear which would normally have cost about £22 million but ended up at £15 million because there was one client, rather than several.
The 41 first tier suppliers have signed the agreement, but BAA expects second and lower tier suppliers also to conform to the agreement and work in a similar cooperative way, so that risks are not simply transferred down the chain to the smallest suppliers who can least afford it.
The new approach to risk had an early test. Because of a wet winter in 2002/3 the ground preparation got badly behind schedule, losing around 14 weeks which, translated through the ensuing five years could have meant a two year delay. With other suppliers waiting to go on site, the consequential claims would have been enormous but instead, the fact that BAA had assumed the risk together with the cooperative approach meant that over the next 7-8 months, the delay was caught back and there have been no legal claims.
Contrast this approach to the way that the London Underground refurbishment has been contracted out to two infrastructure companies, Metronet and Tube Lines under the complicated . Projects are undertaken by the private companies for London Underground which is still publicly owned, but the interface between the two is a constant source of acrimony.
The original idea was, in a way, the same as with T5. Conventional contracts, notably the Jubilee Line Extension, had led to massive cost overruns and therefore it would be better to hand over the management of the project as well as the work itself, to private companies, who would then be best suited to deal with the work flow efficiently. But under the Tube PPP, the contractors have a vested interest, as with conventional contracts, to do as little work as possible knowing they will still receive a guaranteed income under the arrangement. Moreover, the process has exacerbated the blame culture with claims and counterclaims flying back and forth.
Of course, the PPP contracts are supposed to be performance based but it has proved impossible to calibrate the performance in relation to the real cost for the contractors. There has been a series of engineering overruns that have delayed passengers which clearly demonstrates the penalties to the contractors are not sufficient in deterring them from such behaviour. Although notionally the risk has been passed to contractors, in effect it always lands up with London Underground which is seen by the public as being responsible for delays anyway. The advantage of the BAA contract is that there is no pretence that the risk has been passed on and, consequently, the suppliers charge less.
BAA’s approach is, in a way, the antithesis of PFI. Instead of trying to buy expertise from outside and assuming that the contractors will bear risks, which are often ill-defined, instead it accepts that ultimately its own reputation is at stake and that therefore it may as well assume the risk. There is the not inconsiderable advantage that, so far, there has been no recourse to lawyers with any of the suppliers.
There are, then, major lessons to be learnt from the T5 project for the public sector. BAA, itself, has also found it a learning process. Mathew Riley admits that the company underestimated the number of managers that would be needed to ensure the project proceeded quickly: ‘A couple of years ago, we had just 60 people out on the site working with the various contractors to ensure that everything was progressing smoothly. Now we have three times that number, and it may well still not be enough.’
In terms of safety, the approach has undoubtedly proved effective, with no one killed on the site, a good record for such a large project. In terms of progress and finance, the project, which is 55 per cent complete, remains, according to Riley, ‘on time and on budget’. If that is the case when the terminal opens at 4 am on 31 March 2008 – a time and date determined by the changeover to summer schedules and recorded by a countdown clock in the project team’s reception area – then the T5 agreement may well become the norm for major projects.