High Cost of Flawed Projects Sustained by Official Fictions

By Henry Ergas, (“The Australian,” 26/10/2013)

ADELAIDE, August 28, 1993. Prime Minister Paul Keating launches the first Collins-class submarine. As the gleaming vessel hits the water, Australia’s technological prowess glistens as brightly as its military strength. Led by the ABC, the media hails the event as a triumph.

But it is a hoax. The steel plates are timber painted black. The engine has never been tested in salt water. The pipe fabrication is not finished. Nor is the design of the vessel. And the combat system doesn’t work. Far from intimidating our enemies, the greatest threat the submarine poses is to Australia’s taxpayers.

Coffs Harbour, July 4, 2013. Beaming with pride, deputy prime minister Anthony Albanese pushes a big yellow button. “Done,” he exclaims: Coffs Harbour is on the National Broadband Network. Thanks to “the greatest nation-building program in Australia’s history,” Labor has propelled one more electorate into the digital age.

But this too is a sham. The building isn’t connected to the NBN; only Telstra’s copper wire links the big yellow button to the outside world. And with problem after problem crippling the NBN, ever greater doses of fiction have been required to disguise mounting delays. Had Keating and Albanese been company directors, the stunts could have landed them in jail.

But they are above all that. And like a recurring nightmare, the disasters keep coming: from subs and the NBN to pink batts and school halls, governments seem unable to deliver. True, the greatest disasters bear Labor’s mark; but as the Adelaide to Darwin railway shows, no party has a monopoly on folly.

With billions of dollars slated to flow to major projects over the next decade, understanding the failures and their implications should be a priority for the Commission of Audit. Not that the pathologies are particularly mysterious. Tolstoy’s unhappy families were each unhappy in their own way; but program failures are depressingly similar.

Like lemmings hurling themselves over a cliff, governments are propelled by forces as powerful as they are pervasive. At the beginning come the hygiene factors: or rather their absence. Invariably, governments would rather laugh with the sinners than cry with the saints. So the goals being pursued by new programs are not clearly defined, and the difficulties and risks involved not rigorously and transparently assessed.

In Collins, for example, the commitment to proceed was made without proper cost-benefit appraisal either of the program as a whole or of key design choices within it. As the Coles review of the program noted in 2011, this led the Hawke Labor government, “perhaps without fully appreciating the consequences”, to specify a vessel “quite unlike any other in the world”.

And as a review of Collins by RAND found, options to reduce the resulting risks by “backing off requirements slightly, especially for the combat system” were ignored, as were the “trade-offs between operational requirements and technological risks, and their associated cost and schedule implications”.

Equally, in the case of the NBN, the decision to proceed was largely based on unpublished advice from the Australian Competition & Consumer Commission, which lacked the capability to evaluate the options and bore no responsibility for the outcomes. By the time a more thorough analysis was commissioned, the government’s decision was well-entrenched, and the analysis was limited in ways that precluded that commitment being questioned.

A proper cost-benefit appraisal was therefore never undertaken. But even when some type of cost-benefit appraisal is carried out, the multiplicity of objectives any program can serve makes the benefits easy to inflate, while optimism trumps realism in evaluating the outlays that will be required.

That benefits typically flow from the outset, while burdens only become fully apparent in the distant future, means the temptation to commit those errors is all the greater. Sure, from an economic perspective, the inputs programs consume are mere costs: they are what must be sacrificed to achieve the goal, rather than the goal itself.

In political terms, however, those inputs are the exact opposite: they are highly valued jobs for marginal constituencies, contracts for favoured companies, patronage opportunities for cronies. That the program will deliver outputs as well is just cream on the cake.

Given such poor foundations, it is hardly surprising problems occur. In itself, that complex public projects run into difficulties is unexceptional: experience and common sense suggest the costs and time required to successfully complete projects of this kind will be hard to predict. As a result, it is simply unrealistic to expect each project to be completed on time and to budget. And as a famous RAND study found, the problems also beset similarly complex projects in the private sector, and at about the same rate.

In commercial markets, however, troubled projects are either fixed or terminated; but in the public sector, it is only good programs that die young, as governments, faced with projects in difficulty, throw good money after bad in a desperate attempt to avoid recognising defeat. That troubled programs prove such hardy survivors is partly because the “tyranny of sunk costs” comes into play.

Projects that would never have been undertaken if their total costs had been known at the outset are not cancelled because each evaluation concludes spending a (relatively) little bit more might still make the exercise worthwhile. As hope triumphs over experience, natural selection is put on hold.

But if that mechanism proves so compelling it is mainly because of the enormous scope governments have to postpone pain to the future. Each government would rather its successors bore the political grief of program termination, while it reaps any joy continuing the program has to offer; as a result, so long as it can get away with inflicting that “externality” on the future, it will. And everything makes it easier for governments to do so than it would be for private firms. After all, listed entities are under stringent obligations to disclose any material adverse events; and as their share prices reflect expectations of future earnings, a firm’s equity is marked down as soon as losses become likely, even if those losses are years away.

In theory, there are processes in government that should have similar effects: the requirement to properly value the assets and liabilities on the government’s balance sheet is especially important. But those processes work very poorly indeed. In the case of the NBN, for example, the government apparently knew from a report it had commissioned that it was likely to accumulate very substantial losses.

However, by keeping that report in draft form, it was able to refuse its disclosure. And as they were sitting on that advice, Stephen Conroy, the then minister for communications, echoed by Penny Wong, the finance minister, assured parliament that “this is a project which returns all of the government’s money and interest costs.” But it was not only parliament that was potentially being misled. The value of the assets on the government’s balance sheet must be confirmed by the Auditor-General; and public sector accounting standards specify that when it becomes “reasonably likely” that an asset will not provide “a pre-tax rate of return that reflects current assessments of risk”, the shortfall must be “recognised immediately.”

If the assessments available to the government are taken seriously, the assets committed to the NBN should have been written down to zero as they were being acquired and a net liability registered. However, the budget treatment of the equity injected in NBN Co suggests the project would more likely than not break even or come close to it. It is difficult to believe that conclusion could properly have been reached had the Auditor-General been fully briefed on the project’s financial prospects.

None of that is to say cover-ups are the only response when programs such as these run into trouble. Rather, further mechanisms kick in that often make matters worse. Perhaps the most common is the escalation of commitment to what would seem to be a deeply flawed course of action. As government circles the wagons, goal displacement sets in: what was a means becomes an end, cementing the support of the constituencies that are the project’s major beneficiaries.

The NBN again provides a textbook case. Deploying fibre-optic cable should be one way among others of providing very high-speed broadband, not an objective in itself; but the greater the obstacles the project encountered, the more strident became the government’s emphasis on fibre as the magic cure.

This was not a case of the tail wagging the dog; rather, the tail had reshaped the dog’s breed, altered its appetite and dominated the lives of its owners. And ultimately, in Conroy’s theatre of the absurd, the tail completely swallowed the dog, leaving the farce of the big yellow button as the only action that mattered. Yet if such failures can occur and persist, it is also because accountability is so weak.

Nor is the NBN an isolated case in that respect. In his recent review of the events which led to the early decommissioning of HMAS Manoora and the extended unavailability of HMAS Kanimbla, Defence Audit and Risk Committee chairman Paul Rizzo concluded that “accountability, authority and responsibility was misaligned, fragmented or simply misunderstood.”

As for Collins, the Coles review of the submarines’ sustainment could not identify a clear locus of responsibility for the key decisions that had been taken. And who has been held to account for pink batts and for school halls? No one. In a classic study of major program failures, economist David Henderson emphasised the “unimportance of being right”: the fact that there were few rewards for long-term success, and few penalties for those involved in projects that had plainly failed.

Twenty years on, that lesson has still not been learned. Far easier instead to rely on shams such as Keating’s and Albanese’s, while shifting the costs to future taxpayers. Little wonder getting these programs right has proved so intractable. And little wonder the dollars thrown at these vast programs result in more and more of worse and worse.