By Jim Schutze
By Rachel Watts
By Lauren Drewes Daniels
By Anna Merlan
By Lee Escobedo
By Eric Nicholson
The third, my favorite because of its obvious local parallels, was the Sydney, Australia, Opera House. Construction began in 1959 with an estimated budget of 7 million Australian dollars. When it opened in 1973, 10 years late, the full cost was $147 million.
But the authors don't merely offer scare stories. They present cumulative data for entire classes of projects, from rail to software games, and demonstrate how and why each class tends to break budgets and calendars in its own characteristic way and at its own pace. They argue that the human frailty that causes large projects to go haywire occurs equally in government projects and private ones.
The authors look at incentives, relationships, feedback, transparency and consequences. The baseline they use is the weather forecaster on TV. He or she has no reason to lie. The feedback is quick. Everybody sees how the predictions come out. The forecaster pays a cost if the predictions are lousy.
But take a very different scenario: Let's say, for example, a complicated public project with major impacts on private players. Why, we might as well take that downtown convention hotel we're about to vote on and use it as an example.
The incentives of some of the players pushing hardest for the hotel—The Dallas Morning News, for example, which owns real estate virtually ringing the hotel site—are quite distinct from the incentives of the ultimate shareholders, who would be you and I, the taxpayers.
The owners of the Morning News and their friends quoted in Cheryl Hall's story have a major incentive to see an investment made in downtown that might drive action to their own properties.
We taxpayers, on the other hand, need the project itself to make money, so that we won't have to bail it out, and then we need it to increase overall economic activity in the city so we can tax that activity and use the proceeds for public purposes.
We have different incentives.
We also have a glaring case of what the authors call a "principal-agent problem." You and I are the principal here. We are the ultimate owners. But we have this agent working for us, whose name is Tom Leppert. Now, he becomes the principal when he goes to city manager Mary Suhm and tells her he wants a hotel built, and she becomes his agent. She's the principal when she tells a construction company to build it and hotel company to run it, and they are her agents.
But here's what the authors of the article find when they look at relationships like this across the world: Every time you get differing or even conflicting incentives in a project and then spread those out over all kinds of principal-agent relationships, you wind up with lots of reasons up and down the line for people to fudge or delude themselves or flat-out lie.
People don't necessarily sit down and say, "Let's lie about it." They may. More likely, people at each stage of the game say to each other, "Let's be positive, and let's be upbeat, and let's say and do what we have to in order to get the deal."
Then in a place like Dallas, where the social structure of the business world is still a glass pyramid, slippery and steep, people have very strong reasons for getting along, for wanting to be thought of as right types and team players.
The result is a kind of group self-hypnosis which the authors describe as "delusion and deception." Part of it is rose-colored glasses. Part of it is lying. The two go together like horses and carriages.
"Delusion and deception are complementary rather than alternative explanations of the failure of large infrastructure projects due to cost underestimation and benefit overestimation," they write.
The thing I find most convincing in their argument is that they present simple solutions. For example, they suggest one good way to cut down on fudging, lying and rose-colored exaggeration is to make sure everybody, public and private, top to bottom, has real skin in the game.
"[T]o obtain more realistic forecasts and reduced risk, full public financing or full financing with a sovereign guarantee should be avoided," they write. "Whenever possible, the decision to go ahead with a project should be made contingent on the willingness of private financiers to participate without a sovereign guarantee for at least one-third of the total capital needs [authors' emphasis]."
The hotel deal we vote on May 9 is exactly what the authors say not to do. None of the ultimate risk is on the people pushing the hotel the hardest. The pushers all have incentives for self-delusion and public deception.
I think that's what Harlan Crow sees. He's the project's main opponent, the money behind the Vote Yes campaign. He sees a bunch of people all enthusiastic and upbeat about playing around with somebody else's money.
That's not a conspiracy, exactly. It's more like a fever, a kind of joy-juice war dance. Everybody's singing the same song, whipped up and sweaty, feeling great, feeling angry about the ones who won't sing along.
Actually I think I'd be a lot more comfortable with a conspiracy. There's some point at which you can reason with conspirators.