Over at National Review's The Corner, Mercatus's Veronique de Rugy has a post on cost underestimates in infrastructure projects:
...here are some striking facts about government run public work projects. The most comprehensive study of cost overruns examines 20 nations spanning five continents. The authors find that:
- In 9 out of 10 transportation infrastructure projects, costs are underestimated.
- For rail projects, actual costs are on average 45%higher than estimated costs.
- For fixed-link projects (tunnels and bridges), actual costs are on average 34% higher than estimated costs
- For road projects, actual costs are on average 20%higher than estimated costs.
- For all project types, actual costs are on average 28% higher than estimated costs
- These same cost overruns exist in all public work projects
Remember the Capitol Hill Visitor Center? This ambitious three-floor underground facility, originally scheduled to open at the end of 2005, was delayed until 2008. The price tag leaped from an estimate of $265 million in 2000 to a final cost of $621 million.
How can we explain these cost overruns? The authors explain:
These cost underestimation cannot be explained by error and seems to be best explained by strategic misrepresentation, i.e., lying.
It happens in space transportation projects as well. On a cost-plus contract, the incentives are to low-ball the bid, and then get financially well with change orders. Before Constellation was canceled last year, the estimated costs for the Ares I rocket and Orion capsule had ballooned dramatically, and the schedule slipped far to the right (it was slipping more than a year per year). In fact, it tends to be even worse for NASA projects, because they're not even projects that the public will ever use, so it doesn't matter whether they ever actually succeed -- the program, with the jobs it creates in the states and districts of the only representatives who care about it (and for only that reason) is its own justification. Reagan announced the space station program in 1984, with the intent to have it flying by 1992, for a cost of eight billion dollars. The first hardware flew in 1998, and it has now cost about a hundred billion (though to be fair, the original cost estimate didn't include transportation costs of the Shuttle, which the latter number does). No one loses elections because launch systems don't fly, but sometimes they are won by the jobs from the local contracts that the representative brings home.
Veronique also points out that it's not just that costs are underestimated -- demand is overestimated. This happens in space projects as well. The Shuttle's per-flight cost estimates assumed that it was going to get all of the US launch business, and much overseas as well, giving it the high flight rate necessary to bring the average cost closer to the marginal cost, but many of the payloads never materialized, and many of them fled to other launch providers. What's truly amazing about the Senate Launch System is that NASA is admitting that it will fly rarely, and cost billions per flight. But until the rest of the Congress starts to care, and stops deferring to the porkers on the space committees, the waste will continue.