Saturday 23 February 2019

Colm McCarthy: 'These figures add up to one big capital scandal'

Political decision makers approve major schemes based on estimates with no relationship to cost, writes Colm McCarthy

The National Children’s Hospital could cost as much as €2bn, up from the initial €650m, and the €3bn Metrolink is still in the pipeline
The National Children’s Hospital could cost as much as €2bn, up from the initial €650m, and the €3bn Metrolink is still in the pipeline

Elected politicians are being given the run-around by the promoters of large public capital projects. The political decision-makers have been solemnly approving major schemes based on cost estimates which bear no relationship to the bill which is ultimately presented.

This is not just a threat to the sustainability of the State capital programme - even if you believe that the scarcity of public resources has magically been abolished - there is no democratic accountability when sheepish ministers are reduced to mumbling acquiescence as the costs escalate.

There have been regular controversies in Ireland about costly public capital projects over the past few years, including the proposal to spend €3bn on a single tram line in Dublin, the MetroLink project, and the long-running saga of the National Broadband Plan, supposed to cost €500m but now due to cost a large, and still undisclosed, multiple of the initial estimate.

Please sign in or register with Independent.ie for free access to this article.

Sign In

Neither scheme has commenced. But it has just been revealed that the National Children's Hospital (NCH), already under construction on a brownfield site in the inner city of Dublin, will cost almost double the figure on which Government approval was based just two years ago.

For a small country, Ireland displays an impressive willingness to dispense public capital money on a prodigious scale.

I am reminded of a county council chairman many years ago, cheerfully allocating the county road improvement grant to purely political priorities, explaining that "... you can get away with anything in this country provided you call it capital".

In his case, the amounts were in the tens of thousands, and at a time when the country was solvent. The current public investment splurge is overseen by a State which is heavily over-borrowed and when the mistakes to be avoided run into the billions.

Fianna Fail leader Micheal Martin described the latest over-run as "beyond comprehension". Mr Martin is mistaken on this point: cost over-runs on State capital projects are all too easily comprehended, the rule rather than the exception.

In Dublin alone, the five major projects completed in recent decades - the Red and Green Luas lines, the Cross-city line, the Port Tunnel and the M50 widening - all cost well in excess of the figures on which the political go-ahead was based.

The total over-runs on these projects would more than pay for the long-delayed M20 motorway connecting Limerick and Cork. These persistent cost over-runs are not some kind of victimless crime and, contrary to Mr Martin's comment, news that a large scheme had been delivered under budget would be a national sensation.

The NCH qualifies, since it now costs in excess of a billion, as a 'megaproject' in the language popularised by the Oxford economics professor Bent Flyvbjerg. This Dublin experience would not have surprised Professor Flyvbjerg.

With his associates, he has studied cost over-runs on hundreds of similar projects around the world and has enunciated what he calls the Iron Law of Megaprojects: they end up "over budget, over time, over and over again".

Studies stretching back to the 19th Century have documented a consistent pattern: large one-off public projects almost always cost well in excess of the initial estimates on which the political decision was based.

Where estimates of project benefits are prepared, the benefit estimates regularly exceed what is eventually delivered.

This is not just an Irish problem, although Ireland appears to be afflicted with a particularly virulent strain of the virus.

Flyvbjerg has conceded the inherent difficulty in estimating the out-turn cost for big one-off projects: this is not as easy as doing costs-per-kilometre for roads, or costs-per-classroom for schools, repeat projects where experience builds up and errors should be eliminated. But a century is a long time and there appears to have been no historical improvement in cost estimation for megaprojects. And there is hardly any evidence of under-estimation.

If the task was merely difficult, but undertaken with candour and honesty, there would be under-shoots as well as over-shoots. Since there are hardly ever any under-shoots, hence no evidence of learning from mistakes, Flyvbjerg is forced to conclude that the source of the problem is not its inherent difficulty, it is the deliberate deception of decision-makers by the project promoters. The people making the misleading cost estimates must be doing so consciously.

Given the record of repeated over-shoots on major projects and the failure to improve cost estimation over time, you would imagine that embarrassment alone would have encouraged politicians to address the problem. To the extent that they have, their efforts have made no difference whatsoever.

Ireland is in the happy position of possessing a splendid official statement of national policy on the evaluation and management of capital projects.

It is called the Public Spending Code and covers all the bases. Its genesis goes all the way back to the early 1980s when the Department of Finance first became alarmed about cost over-shoots.

The most recent iteration of the Public Spending Code requires that a full cost-benefit study be prepared for all publicly financed projects costing €20m or greater. This requirement is being enforced, at considerable cost and inconvenience, for some of the smaller projects but is being blithely ignored for the larger ones.

For the National Broadband Plan, there has at no stage been a public figure available for project benefits and there is currently no definitive figure for costs. Nonetheless, the Government has recently re-committed to proceeding with the plan.

The case of the NCH is exceptional even by Irish standards. According to Paul Cullen and Martin Wall, writing a week ago in The Irish Times, the cost will likely reach €2bn, twice the figure on which the Government signed off two years ago and treble the figure of €650m agreed as recently as 2012.

Hospitals internationally have been costing about €1m per bed, so the €650m budget should have been plenty for the 500-bed project. But the constricted site chosen in the inner south city, as well as offering poor access for a national facility, appears to be a factor in the near-vertical cost escalation.

Campaigners for alternative and more accessible greenfield sites warned about costs to no avail. Important projects elsewhere will now have to be scrapped or delayed, unless the Government has received secret permission from the European Central Bank to print euros.

The Government's capital spending plans for the next few years already look rather ambitious and the NCH has just blown a large hole in the available budget. Cullen and Wall described the cost over-run as "a scandal in the making", which is far too kind. This is already a scandal fully-formed, the biggest and quickest over-shoot in Irish history.

The Public Spending Code should either be formally withdrawn to end the pretence, or placed on a statutory basis, with defined accountability for cost estimation and consequences for misleading the Government.

The signal from above, yet again, is that misleading cost estimates have no personal consequences for anybody. The National Broadband Plan should be put immediately on hold pending a serious estimate of benefits and costs.

It's too late the stop the NCH but a week is enough to find a judge for the sworn inquiry.

Sunday Independent

Today's news headlines, directly to your inbox every morning.

Don't Miss