Failing to See the Forest for the Trees

New York University’s Transit Costs Project has issued its final report on why it costs so much to build transit infrastructure in the United States. While some of the answers appear reasonable at first glance, the report suffers from the researchers not asking the right questions.

Click image to download a 26.4-MB PDF of this report.

In its review of Boston’s Green Line, the report notes that “Understaffed agencies lacking experience with large capital construction projects struggle to manage consultants.” One result is less than half the costs of the project went into construction; the rest went to pay consultants. We’ve seen that happen with Honolulu and other rail projects as well.

In its review of New York City’s Second Avenue Subway, which has been called the most expensive per-mile transit project in the world, the report blames a similar over-reliance on consultants plus high labor costs, expensive mitigation measures needed to get buy-in from other agencies, and extra-expensive construction methods (such as boring instead of cut-and-cover).

The report is a stunning 424 pages long, yet amazingly it doesn’t manage to review any other U.S. construction projects beyond these two. This just seems inadequate as New York and Boston are completely different from, say, Phoenix and Honolulu, and what applies in one may not apply in the other.

More fundamentally, the report’s authors took the wrong approach. They asked questions like, “why was this contract worded that way?” and “why did project planners agree to absorb such-and-such an expense?” Instead, they should have asked, “what kind of political and decision-making process allows for such expensive projects to be approved in the first place?” and “what were the incentives facing decision makers and how might they differ from those in other countries that build projects for less money?”

The first question is crucial, as a truly rational planning process would never allow most of these projects to get past the initial proposal stages. Far from questioning this, the report is part of the problem, taking it as a given that the Second Avenue Subway, no matter what it cost, was worth building. It is exactly that attitude that allows planners to propose and eventually build ridiculously expensive projects whose results could have been achieved at a far less cost using other technologies.

The incentives question is important as well. The report writers also reviewed projects in Istanbul, Italy, and Sweden. Are those projects less expensive because of lower labor costs, more authoritarian systems that allow planners to ignore the impacts on other interest groups, and greater expertise in construction so less of a need to rely on consultants? The report’s answer is yes, to all of these, but despite that answer, it observes that costs are rapidly rising everywhere.

The real answer, I suspect, is that the transit lobby has persuaded the public that transit is all good and no bad. This in turn persuaded politicians that they can spend as much as they want on transit (unlike freeways) with no political backlash.

For politicians, the cost is the benefit. Phoenix recently finished building a 22-mile, 8-lane freeway that cost $1.4 billion, a few years after building an under-20-mile light-rail line for about the same cost. For many politicians, these are functionally equivalent since they cost the same even though the freeway will carry 10 to 20 times as many passenger-miles and infinitely more freight each year. (Despite transit propaganda, no light-rail line can move as many people as an eight-lane freeway; very few move as many as one freeway lane.) In fact, politicians may prefer the light rail because it is less controversial than the freeway. (These numbers are from the Phoenix Loop 202 South Mountain Freeway and its original light-rail line.)

Since the politicians who are supposed to be watchdogging agency spending are thrilled by projects that spend ever more money to get less results, the agencies have no hesitation to proposing insanely expensive projects. Sometimes they start out lowballing costs to get initial buy-in, but they know they won’t be held to those costs.

In at least two cases — Maryland’s Purple Line and the Twin Cities’ Southwest Line — a nominally fiscally conservative governor agreed to approve the project on the condition that costs could be reduced to less than $2 billion. The agencies dutifully reduce the costs and then, after the first spade of dirt was turned, costs rose by 33 to 50 percent, and hardly anyone, including the governors, made a peep. The reality was that, even at $2 billion, the projects were stupid.

At heart, and I keep harping on this, the real problem is the disassociation of costs from user fees. If costs have to be paid for out of user fees, then expensive and obsolete technologies will automatically be rejected. But if there is no relationship between costs and real measurable benefits, then there is no need to control costs at all. Any agency leader who supports lower-cost solutions loses out because their agency’s budget will be smaller and any politician who tries to control costs loses out because they bring less money into the pockets of potential campaign contributors.

Sadly, the term “user fees” doesn’t appear anywhere in the transit costs report, and it only mentions incentives twice in connection with the perverse incentives created by adding a large contingency fund to project budgets. There’s nothing about the incentives that led agencies to propose expensive projects in the first place, such as the existence of a federal multi-billion-dollar slush fund for these projects that gives the agencies money without questioning whether the projects were worthwhile in the first place.

In short, the authors of the transit costs report failed to see the forest for the trees. Their report implies that more experienced leaders, better cost control, and more productive workers would solve the problem of rising construction costs. Yet these are just symptoms of the real problem, which is the willingness of Congress and local politicians to throw money at projects regardless of their benefits or costs.