Last year, Bent Flyvbjerg, a Danish professor of economic geography specializing in megaprojects, and Canadian journalist Dan Gardner co-authored a book How Big Things Get Done. They investigate what they coin the “Iron Law of Megaprojects,” which holds they routinely come in well over budget, far past projected deadlines, and without the projected benefits.

Unfortunately for taxpayers, the book contains numerous examples of Canadian megaprojects that follow this Law of Megaprojects. The federal government’s infamous firearms registry is a textbook template for how IT projects can go terribly wrong, ending up 590 per cent over budget. The Muskrat Falls hydro project in Newfoundland is cited as a classic demonstration of what happens when hiring a firm with little direct experience managing such a large complex project. Most famously, the 1976 Montreal Olympic Games wins the title for the largest cost overrun in Olympic history, finishing 720 per cent over budget. The authors suggest Montreal’s “Big Owe” stadium “should be considered the unofficial mascot of the modern Olympic Games.”

One thing all these Canadian examples have in common is extensive government involvement. Not that governments learned from their past mistakes. The federal government’s Phoenix pay system fiasco demonstrates that IT remains a black hole, with the government recently announcing it would abandon Phoenix after spending $3.5 billion trying to implement it. Several light train projects across the country have gone off the rails, the poster boy being the system in Ottawa, which is years behind schedule and already $2.5 billion over budget.

There are several reasons why government projects are chronically prone to failure. One is that politicians, especially late in their careers, want legacies in the form of monumental tangible projects irrespective of whether they effectively meet a public need. You can see this dynamic clearly at work today in Canada, as the Trudeau government pushes for a prohibitively expensive (probably more than $100 billion) high-speed rail connection between Windsor and Quebec City. Meanwhile, Ontario Premier Doug Ford promotes a traffic tunnel underneath Highway 401 between Brampton and Scarborough, and Quebec Premier Francois Legault revives plans for a third link connecting Quebec City to the south shore of the St. Lawrence River. While Canada clearly needs more transportation infrastructure, these projects are not the most cost-effective way of meeting the needs of commuters.

Governments deceptively deploy several tricks to help get uneconomic projects built. They routinely produce unrealistically low-cost estimates to make wasteful ego-driven projects appear affordable. Another tried and true tactic is to just “start digging a hole and make it so big, there’s no alternative to coming up with the money to fill it in,” as former San Fransico mayor Willie Brown admitted. This approach preys on the mistaken belief that large sunk costs mean scrapping a project “would be interpreted by the public as ‘throwing away’ the billions of dollars already spent” when it is actually a textbook example of throwing good money after bad.

Unlike other studies of how major infrastructure projects typically are over budget, Flyvbjerg and Gardner have some concrete recommendations on how to manage large projects that respect deadlines and budgets.

These steps include careful consideration of the actual goals of the project (airlines can meet the need for fast transport in the Windsor-Quebec corridor without the expense of high-speed rail), detailed planning and preparation followed by swift execution to minimize costly surprises (summarized by their advice to “think slow, act fast”), accounting for the cost of similar projects in the past, and breaking large projects into smaller modules to allow projects to scale back when they run into trouble. A good example of these principles at work in Canada were several oilsands projects built before 2015, when severe shortages were addressed by firms using modularity and synchronizing their work schedules to free up scarce labour and materials.

However, one major flaw in Flyvbjerg and Gardner’s analysis is their failure to understand the economics of renewable energy. They cite solar and wind projects as examples of projects that routinely finish under budget, a major factor in their declining cost. But building renewable energy is not their only cost to the energy grid, as back-up plants must be maintained for those periods when the sun is not shining or the wind is not blowing, as noted in a recent article by Bjorn Lomborg. The expense of maintaining plants that often are idle raises overall costs. This is why jurisdictions that rely extensively on renewable energy, such as Germany and California, have high energy costs that must be paid either by customers or taxpayers.

However, apart from this mistake, there is much governments and taxpayers can learn from How Big Things Get Done.