Over the past 20 years there have been repeated examples of major capital project overruns. But has anything been learned from both these and the current National Children’s Hospital fiasco to warrant confidence that they can be prevented in future?, asks Caroline O’Doherty.
THERE is a pattern to the progress of the National Children’s Hospital project that is painfully familiar.
First came the proposal, then the planning, then — once various rows were resolved — the start of building, followed by the cost overrun.
Of course not all major construction projects end this way but they do often enough, and have done for long enough, that anyone presuming a different outcome should be knighted for services to optimism.
Over the past 20 years alone there have been repeated examples of overruns, the rows they create, the analyses they prompt and the warnings they produce.
In 1999, the Comptroller and Auditor General (CAG), the national public spending watchdog, carried out a value for money examination on the development of the national roads network which at the time was undergoing substantial investment.
The National Roads Authority had been set up in 1993 and a programme of works was laid out for the period 1994-99 which promised ribbon-cutting and plaque unveiling opportunities for politicians starved of such pleasures during the grim eighties.
By 1997, however, €792m had already been spent — close to €200m over budget. The EU had pledged 85% of the original budget but with the escalation in costs, the proportion coming from Brussels dropped to less than 60%. The shortfall would have to be made up by the taxpayer.
There was no allowance for inflation, no checks to prevent chopping and changing of designs after NRA approval, and no fixed price contracts in use. In effect, there was “a bias towards underestimation of costs”.
No-one should have been surprised. In 1997 it was already clear things were going financially awry and private consultants DKM, were called in to give their opinion — possibly the start of the endless stream of costly consultants reports into costly overruns that so maddens the public.
DKM identified the same weaknesses that would later be confirmed by the CAG — failure to make accurate cost estimations, failure to control costs throughout the project and too much re-drawing of designs and specifications post-approval.
To try to keep a tighter rein on spending, a new system of quarterly project reports was introduced by the NRA along with guidelines for local authorities on project cost estimation.
Now everyone would know what they were meant to do and the public purse would be safe from plunder.
Just five years later, the CAG was back with another report, once again examining roads, once again finding that costs were out of control, once again making recommendations that had relevance for all major capital projects.
The overrun reported in this 2004 analysis was even worse — road projects planned for 2000-2006 were meant to cost €7bn but had risen in a very short space of time to €16.4bn.
Challenges specific to certain projects — such as the Dublin Port Tunnel — accounted for a lot of the overspend but the most reasons were failures to manage.
“Underestimation of prices” and a “systematic failure to cost certain elements of schemes at the planning stage” were highlighted as well as projects being allowed to grow in scale after they had been approved. These were the same points raised previously.
A baffling arrangement also existed whereby consultants got paid a percentage of the final cost of a project, giving them little incentive to stay within budget.
But then it had also baffled members of the Public Accounts Committee when it held hearings on the Campus Stadium Ireland overruns — two years earlier.
While the CAG was preparing his assessment, then minister for the environment, Martin Cullen, was talking tough about cost overruns, citing an average increase of 40% between the acceptance of tenders for major works and the completion of the same works.
He invited various groups to make suggestions as to how to improve cost control. Among the submissions was one from Engineers Ireland which raised many of the issues troubling the CAG but which also suggested what would ultimately become the Strategic Infrastructure Act to enable major works sidestep local planning offices and go straight to An Bord Pleanála for adjudication.
Other suggestions were not taken up. The body recommended that preliminary estimates for projects not be made public so as to avoid overly simplistic comparisons with end costs. It also recommended loosening up ‘nuisance’ rules to enable construction continue at night or in other circumstances considered intrusive to the public.
The CAG and the minister’s exercises also identified problems with the two forms of construction contract commonly in use, neither of which was fixed price and both of which offered too much wriggle room for additional expenditure.
So a decision was taken by government to introduce the fixed price contract, although it did not come into effect until 2007 and it was not mandatory.
In relation to the National Children’s Hospital, Health Minister Simon Harris gave the following explanation as to why the project was not subject to a fixed price contract:
“The Government Contracts Committee, the cross-government consultative committee for construction procurement matters, accepted that given the project’s scale, complexity and its programme, the circumstances were such as to warrant a deviation from the standard form of Government contract and agreed a derogation from its use.”
In other words, when a project is so complex that it carries a high risk of cost escalation, one of the key safeguards against such escalation can be abandoned on the grounds of complexity.
The success or otherwise of the fixed price contract is also difficult to judge because of the timing of its introduction — just as the recession bit, major public capital projects all but ceased and the few that went ahead had the benefit of a starving and consequently competitive construction industry. Even then, Terminal 2 at Dublin Airport went 20% over-budget before it opened in 2010 and by 2014, overruns on the Poolbeg Incinerator included €30m on consultants originally contracted for €8m and €50m on buying land worth €6m — all before construction had even begun.
Now that there is some more money around — or at least there was until the children’s hospital devoured it — the question is, has anything been learned from all the overruns of the past and the current fiasco to warrant confidence that they can be prevented in future?
The academic research on the subject is as depressing as the carry-on behind the scenes of the children’s hospital. One of the leading authorities on major project management, Professor Bent Flyvbjerg, who coined the droll phrase “over budget, over time, over and over again” has a theory about the psychology behind megaprojects that he calls the four sublimes.
It suggests that getting heads around the financial minutiae of a major capital project is only half the battle — getting the mindset of those heads right in the first place is the key challenge.
The technological sublime is the excitement engineers feel about getting involved in something that’s going to be the biggest ever, best ever project of its kind. The economic sublime is the thrill felt by industry and trade unions at the prospect of big money contracts and jobs.
The aesthetic sublime is the pleasure that people who appreciate good design get from building something iconic that will become a new landmark.
And then there’s the political sublime — the rapture politicians get from building something that they’ll be remembered by.
Unfortunately, as is often said, there’s only one step between the sublime and the ridiculous.
It was the year 1999 and Bertie Ahern, still bathing in the post-Good Friday Agreement glow, seemed brimful of new millennium optimism.
He’d made his mark on the landscape of history and now he was going to make it on the landscape of west Dublin.
He was going to build Sports Campus Ireland, a sporting wonderland of training and playing facilities with the country’s first 50-metre swimming pool, a velodrome and, most importantly, an 80,000-seater stadium that would host all the main sports and attract international competitions.
The FAI were already planning their own mega-stadium but that didn’t deter the then-taoiseach, whose “Bertie Bowl”, as it was nicknamed, became a personal mission.
He even managed to link it to the peace process, saying he had informed then-First Minister David Trimble of his plans and his hope that the stadium would become an all-island venue.
The original cost was estimated at €280m, but Bertie’s PD partners were not convinced and an independent review was ordered. Completed in late 2001, it warned the cost could actually hit up to €1.1bn.
Still, Bertie clung on to the dream until the end of 2002 when he finally had to concede that a dream is all it would ever be. They were costly imaginings, however — some €43m had been spent on plans and reviews — quite apart from the millions spent relocating Department of Agriculture facilities which were then the main occupants of the Abbotstown site.
Determined to salvage something from the debacle, the Government went ahead with the swimming-pool element but the cost of the National Aquatic Centre, as it would be called, spiralled from €30m to €62m.
That did not take into account the millions in annual subsidies it would take to offset the running costs thereafter, the €1.5m settlement with an underbidder for the project after it was revealed the winning bidder was a shelf company with no assets that never should have got the contract, several million in legal fees, and the revenue losses when the stadium had to close in 2005 after part of its roof blew off in a storm.
Abbotstown has belatedly begun to resemble how the early plans saw it, with many sporting bodies moving their headquarters there and establishing training and research facilities, most notably the FAI, but with the major renovations at Croke Park and the Aviva Stadium in recent years, the idea of a national stadium on the campus has become an expensive footnote.
When the Bottlehill Landfill site was completed in North Cork in 2007, the €6m overrun was the least of the county council’s worries.
It had, by then, cost a total of €45m so, on the scale of overruns in this country, this one was quite modest.
The bigger problem was that the landfill — built to the highest specifications with the aim of making it the cleanest dump in the country (no irony intended) — was no longer needed.
Since the idea was mooted a decade earlier, national waste policy had changed and dumping was no longer acceptable.
Critics argued this should have been obvious but somehow, the Bottlehill bull kept on charging forward.
Since then, the price tag has only risen, with annual maintenance costs running at €200,000 a year and millions being paid out on loan repayments which gate receipts were meant to offset.
Over the years there have been numerous suggestions as to what to do with the site including turning it into a receptacle for ash from waste disposal incinerators or as a final resting place for knotweed once the pernicious plant was removed from the county’s gardens and construction sites.
The latest proposal, currently under consideration, is to locate a windfarm on it.
It started out as a mundane, purely functional project —but the decision to upgrade Irish Rail’s ageing signalling system was to become one of the most bitter controversies in the history of overspends.
The project was initiated in 1996 and contracts were signed the following year, with a IR£14.6m (€18.25m) budget agreed.
Two years later, CIÉ’s chief finance officer became aware that costs had risen, but only by 15%, which he considered manageable. But in seeking to find out what the new end outlay was likely to be, the contractor was asked to come back with new costings.
They quoted IR£40m (€50m) — almost triple the original estimate. PWC, the go-to consultants in an overrun, were drafted in and produced a highly critical report.
The subsequent Rail Signalling Inquiry, set up by a sub-committee of the Public Accounts Committee, was a tense affair. The former CIÉ chief executive had left on early retirement and then took his own life shortly before public hearings began.
It was also known by this time that at least part of the problem with the project was that it had been complicated by a side-deal with the telecoms company, ESAT, which wanted to lay fibre optic cables in the ducts Irish Rail were creating for their new signalling cables as a handy way of criss-crossing the country.
Eventually, the inquiry was suspended due to a legal challenge by the widow of the former CEO, and then abandoned as a result of a successful legal action by gardaí involved in the Abbeylara incident, who objected to a separate Oireachtas inquiry.
Their action, subsequently backed by the Supreme Court, had significant consequences for all Oireachtas inquiries as it found they had no right to make findings of fault against ordinary citizens.
What was learned before the whole thing came to a halt, however, was that the complexity of designing the signalling system was vastly underestimated and there were major deficiencies in accounting and management oversight, all exacerbated, to an unknown degree, by the eagerness to accommodate ESAT on the basis of payments into the future that might or might not materialise.
CIÉ ended up buying out the contract and continuing with the works itself, by this stage with an estimated completion cost of €58m —not including legal bills.
Thirty-five years after the last trams ceased operation in Dublin, they were back on the agenda, with ideas being tossed around for ways to improve public transport in the city.
It took about three years for the jokes to die down and the project to be accepted as a serious proposition, but, by 1997, there was a plan in place and a costing of £227m (€285m). And very quickly after that, the Luas project became a runaway train.
By 2001, revised estimates put the cost at £500m (€625m) — more than double the original price — attributed partly to inflation in construction costs and partly to the decision to scrap the original plan for a single route from Tallaght to Ballaly, through the city centre, and to go with two free-standing lines instead.
There were warnings that a linked-up line might be more useful, but given the disruption and cost of digging up the heart of the congested city centre, it was decided that two separate lines would suffice. By 2003 — the year the Luas was supposed to start running, but didn’t — the cost had been revised upwards again, to €765m, a jump blamed on the discovery of many undocumented pipes when digging began, and which had to be removed or diverted.
Seamus Brennan was transport minister at the time and defended the increases, arguing that comparisons should not be made with the 1997 costing, because that had been just a guesstimate.
The first Luas line began running in June 2004 and the second the following September, but, soon enough, the decision not to have the lines linked, as per the original plan, was regretted.
In 2017, after four years of construction, Luas cross-city opened with much fanfare and trumpeting about the fact that it had come in on time and under-budget. It cost €368m, not the anticipated €370m, although the spare €2m didn’t quite make up for the near tripling of the costs for the original project.
As the 150th anniversary of the Great Famine approached in 1995, an idea was floated in Kerry that the occasion might be commemorated by building a replica of one of the famine ships that ferried escapees from Tralee to North America during the grim 1840s.
That was in 1993, but it took five years before a financial package, involving Tralee UDC, Shannon Development, FAS and private industry, was put together to back it.
State grants of €2.5m were to cover most of the building, which was to provide experience for young carpentry and construction trainees.
But after two years of building and the floatation of the ship, faults were discovered and remedial work had to be undertaken. By the time the ship was declared seaworthy in 2002 it had cost the State €10m.
With some fanfare, it made a voyage to the US and Canada and then around Ireland, but after that it wasn’t clear what purpose it was to serve.
Eventually, in 2005, the Dublin Docklands Development Authority bought it for €2.7m, but the authority had no clear plan for it either, so it remained tethered to the quayside in Dublin, hosting birthday parties and corporate events.
The DDDA offered it to the Department of Defence for sail training, but, ironically, it was found to be suffering significant water damage, and the department swiftly declined.
In 2015, when the DDDA wrapped up its work, Dublin City Council inherited the ship which by then had cost the State €14m and was assessed to be worth just €150,000.
In the last few years, it has had a comeback and now serves as a Famine museum that is popular with tourists, but its future remains a concern.
Dublin City Council, currently devising plans to rejuvenate the George’s Dock area of the quays, recently considered moving the Jeanie Johnston, and associated facilities, to this area but was told the costs of doing so, given its condition, could run to €3m.
The Dublin Port Tunnel has the ignominy of being included in numerous textbooks for students of construction management thanks to an overrun that saw the cost either double or increase seven-fold, depending on the starting point taken.
At the very outset, in the early 1990s, it was believed the tunnel, which links Dublin with a northerly section of the M50, taking heavy goods vehicles out of the city centre, could be built for about £100m.
That very quickly jumped to €220m when concrete plans were drawn up, and by 2000, when the tender was awarded to a construction consortium, the expected price was €448m.
When it eventually opened in December 2006, almost two years late, it was heading for €800m and when all the dust had settled (in other words, when all the lawsuits had been settled) the final figure was put at about €804m.
Additional building works were part of the problem. In the closing stages of construction, large cracks were found in the roof with rainwater spilling through and substantial remedial work had to be undertaken.
It was also discovered that the roof was too low to accommodate the new generation of high-sided super trucks many distributors were beginning to use for delivery so a hi-tech detection and warning system had to be installed to prevent collisions.
But the biggest driver of additional costs were the compensation claims by householders whose properties were damaged by drilling vibrations, and by the construction consortium which claimed it had to carry out more work than expected, as well as fees for lawyers, consultants and reviews to deal with the myriad of issues that arose during construction. It is estimated that expenditure on paper and pleadings, as opposed to bricks and mortar, accounted for more than a third of the final bill.
The tunnel has undoubtedly made a difference to traffic in the capital but not to the extent envisaged. It has the capacity to handle 80,000 vehicles a day and it was expected to be handling 30,000 after a few years but it hasn’t yet reached 22,000.
The Glen of the Downs road widening scheme didn’t just widen a narrow, congested stretch of the N11 in Co Wicklow, it stretched the budget too.
The project was first mooted in 1990 but ran into objections as the planned dual carriageway would eat into a nature reserve.
When it finally got the go-ahead in 2000 after a High Court challenge, it was expected to cost €49m. When it opened three years later, it had swallowed up €85m.
Part of the increase was down to delays after eco-warriors set up camp in the woods and tied themselves to trees, defying court orders to leave until eventually dozens of gardaí were dispatched to the scene.
But that was mainly during the planning and preparation phase. Costs continued to soar during construction amid disagreements with the contractors that ended up going to arbitration.
Having to make provisions for fish “discovered” in a local stream was blamed for some of the additional cost although after chasing eco-warriors through the woods for three years, this perhaps should have been noticed.
Obstacles with water mains were blamed as was the fact that work had to take place while 30,000 vehicles a day continued to travel the route.
Once the section of road opened, motorists enjoyed a short-lived relief from the chronic delays but congestion on the stretch route has returned and it is now part of a longer stretch of the N11 for which widening plans are being drawn up.
When a flood relief scheme was devised for Kilkenny city in 1998, it was meant to withstand the extreme water levels of a one-in-one-hundred-year crisis event.
But it was only a few years before it would be hit by a deluge of criticism, instead.
The original costing was €13.1m, but by 2004 it had rocketed to €48m.
When probed, the Office of Public Works said that the costs had escalated partly because of unexpected archaeological finds.
But that didn’t really wash, because the medieval city of Kilkenny was renowned for its archaeological riches (those already uncovered and those yet to be).
It emerged that only €630,000 was set aside for archaeological excavations, but that the cost had risen to €4.82m.
Other factors blamed for the escalating costs included the annual salmon run, which prevented construction during the summer months — again, something that should have been well-known in advance — and the unexpectedly high costs of compensating landowners.
The OPW admitted that the budget had been blown so catastrophically that similar plans for Carlow, Clonmel, and Waterford, would have to be put on hold.
The minister who had responsibility for the OPW, Tom Parlon, however, dismissed the criticisms, telling the Dáil: “The Kilkenny flood relief scheme has, at every stage, been subject to a process of rigorous cost control and cost-benefit analysis and has fully complied with the criteria laid down by the Department of Finance for the appraisal and the management of capital expenditure in the public sector.”
Even when there’s been a colossal overrun on a major project, the instigator can usually cling to the hope that after all the rows and finger-pointing, there’ll be something tangible at the end of it that will look impressive enough to take peoples’ minds off the financial pain it represents.
It wasn’t so with PPARS.
The PPARS (Personnel, Payroll and Related Systems) computer system installed by the HSE was intended to computerise all personnel records. That would mean taking the myriad of different payroll and human resources systems used by individual hospitals, health boards, community care facilities and other health agencies and integrate them into one big, beautiful database.
At the touch of a few buttons, HR and payroll managers in all the different agencies and facilities would be able to organise rosters, calculate leave entitlements, monitor headcount and wage costs, assess manpower needs and track vacancies, plan pension funding and do many other administrative duties required to run an organisation with about 100,000 staff in an efficient and standardised manner.
The idea was first proposed in 1995 and when it was settled upon in 1998, the anticipated cost was just over €9m and it had a two-year completion date.
How it mushroomed to €220m a decade later with still only a third of employees being paid through it and its software out of date, is best summarised by the Comptroller and Auditor General who prepared a damning report.
He said there had been no clarity about how the system would operate, too much haste in opting for it, a lack of research into or appreciation of just how different and scattered the multiple systems already in place were, an inability to make hard decisions when things started going awry, an inability by the operators of the existing multiple systems to accept change, and a failure to properly assess a pilot project before attempting to roll the system out nationally.
Throughout the process, some five different reviews took place, with consultants being paid around €57m, and still no-one shouted stop or thought to cap the budget available for fixing the mess.
In 2017, the HSE commissioned research on a new integrated financial management system to incorporate payroll, saying it would cost €30m and the PPARs mistakes would not be repeated.
At 165 years old, Cork Courthouse was long overdue a facelift when it was decided in 1997 to undertake €6.5m refurbishment.
By the time the job was completed in 2005, however, the bill had risen four-fold to €26.5m. Meanwhile, the court had been using a leased property costing almost €762,000 a year — three times the amount agreed in 1999.
But at least the building was now a “21st-century courthouse” that would serve the people of Cork “for the next two centuries”, as the Courts Service put it.
Ten years later, the spruced-up courthouse was deemed unsuitable for the volume of cases it handled and a further €35m had to be put into refurbishing a second court building in the city so that the workload could be split between the two.
When the Comptroller and Auditor General asked about the initial overspend in 2003, the Department of Justice protested that the original estimate had been “very provisional” and “not based on actual plans”.
The department also argued that construction inflation was to blame for 60% of the cost hike and that the original costing did not include consultancy fees, site investigations, furniture, fittings, or equipment.