Why cost-effectiveness is sometimes the wrong decision rule

Federal transit-funding rules embody an elementary blunder in policy analysis: using cost-effectiveness in a situation where there are non-resource costs. The decision to build an elevated rail line to Dulles Airport, rather than tunneling under Tyson’s Corner, is an example of how bad analysis leads to bad choices.

Cost-effectiveness analysis is a good way to compare different approaches to providing a single benefit. The rule is to choose the approach that gives the greatest amount of the target benefit per dollar (or other resource unit, such as a work-hour) spent, or equivalently to generate a given level of benefit at the least cost.

Cost-effectiveness analysis has a huge advantage over benefit-cost analysis: it requires much less information. In particular, it doesn’t require putting a dollar value on benefits.

For that very reason, cost-effectiveness analysis can’t be used to choose among project where there are multiple benefits or non-resource costs.

For example: Congress required that mass-transit projects meet a cost-effectiveness standard in terms of passenger-minutes of travel time saved per dollar spent. Sounds reasonable, though of course the Congress puts no such requirement on the much larger highway budget.

But looking only at travel time saved ignores the other benefits and costs of transit projects: impact on land use, impact on commerce, and impact on the local environment. By that standard, an elevated rail line, which creates a desert around itself, looks “cheaper” than a tunnel, which doesn’t. In this case, the resulting decision is going to make a mess out of Tyson’s Corner.

As Plato said, we will never be well-governed until policy analysts run the government (a horrible thought) or politicians learn at least the basics of policy analysis.

Finger-Pointing After Death of Tunnel Plan

Supporters Blame Kaine’s Handling, Federal Demands

By Alec MacGillis

Washington Post Staff Writer

Friday, September 8, 2006; B05

In the vision of Fairfax County political and business leaders, Leesburg Pike in Tysons Corner would be transformed from a car-clogged strip of auto dealers and shopping plazas into an urban boulevard with a tree-filled median and sidewalks lined with shops and cafes.

But now it’s official: Running down the middle of the street will be an elevated train track, 35 feet high, with concrete pillars every 70 to 100 feet. Construction of the line, which is to reach Reston in 2012, will require a complete overhaul of Leesburg Pike (Route 7), one of the busiest streets in the region.

Fairfax residents and officials reacted with disappointment and finger-pointing yesterday to the announcement Wednesday by Gov. Timothy M. Kaine (D-Va.) that the planned Metrorail extension from West Falls Church to Dulles International Airport would run through Tysons on an elevated track instead of below ground. Critics zeroed in on federal standards for public transit funding, which they called overly rigid. Others questioned whether Kaine could have won approval for a tunnel if he had handled things differently.

The decision, tunnel supporters said, represented a setback for Fairfax’s hopes for turning Tysons into a walkable, vibrant downtown similar to Arlington’s Rosslyn-Ballston corridor. An elevated track, they said, will make it harder to walk around Tysons, add a new street grid and build to the edge of sidewalks, all keys to creating a successful downtown.

“It’s sad. The last thing Tysons needs is another silly barrier, and that’s what it’s getting,” said Clark Tyler of McLean, chairman of a county task force drafting a new master plan for Tysons. “We’ve got the Beltway and Route 123 and Route 7, and now we’ll get this thing sticking up that you can see from Pittsburgh.”

Kaine’s decision came after federal transit officials and Northern Virginia Republican Reps. Frank R. Wolf and Thomas M. Davis III warned him that switching to a tunnel for the four-mile Tysons stretch of the line, as Kaine was close to doing, would imperil the entire 23-mile, $4 billion project. The delay caused by the design change, federal officials told Kaine, could jeopardize the $900 million that the federal government is expected to provide, because the money might be lost in a future funding cycle.

Even more important, federal officials said, was the likelihood that the extra cost of a tunnel — estimated at $250 million by an engineers’ panel convened by Kaine — would put the project in violation of “cost effectiveness” standards used by the Federal Transit Administration. These standards, which Congress first adopted 20 years ago, are essentially a formula that determines how much time is saved by transit users for each dollar spent.

The formula looks at a project’s total cost, not just the federal share. Even if Virginia found extra money from local or private sources to pay for a tunnel, without seeking more federal money, it would risk exceeding the cost standards and losing the $900 million expected federal share.

Lawmakers say the rules help ensure that limited funds for public transit go only to sensible projects and not boondoggles. But effectively, the standards allow the federal government to influence the design of a project, like Dulles rail, for which it is covering less than a quarter of the cost.

To tunnel supporters, it’s simple: Public transit projects are being held to cost-benefit standards more stringent than those applied to highway projects. Federal funding of new transit projects is $1.5 billion a year, a fraction of highway spending.

“I don’t see why we have such trouble holding onto funds for rail in Northern Virginia when Alaska has done a great job of holding onto its money for the ‘Bridge to Nowhere,’ ” Tysons resident Neil Sullivan said, referring to a $223 million bridge that will serve a town of 50.

FTA officials pointed yesterday to the agency’s long-standing rules for funding only projects that are “meritorious” but declined to comment about the Dulles project.

Other supporters of a tunnel asked why Kaine waited to meet directly with federal officials until the last days before announcing his decision, after much enthusiasm had built for a tunnel. Perhaps, they said, the state might have won approval if it had taken the cost standards more seriously earlier. As it was, state officials were able to present the federal agency with only a few pages of cost estimates provided by companies seeking to build the tunnel.

“Some of us from months ago knew the federal funding issue was a huge unknown, and I am surprised [the governor’s office] did not get those answers much earlier,” said Del. Thomas Davis Rust (R-Fairfax). “He may have gotten caught up in the moment. A lot of people were pushing it, and he may have just gotten caught up in it and thought, ‘We can work around this.’ “

Kaine defended his handling of the decision. He said his office had been aware of the cost standards all along but had thought that they could be overcome after learning from the engineers’ panel late in July that the extra cost of a tunnel would be less than previously estimated. At first, he said, Federal Transit Administration officials indicated that a cost difference of $250 million might be acceptable. Only more recently did officials make clear that a tunnel would not get federal backing, he said.

“When we first made presentations [of the panel’s findings] to both FTA and congressional staff, they had the same reaction: ‘Huh? That’s not nearly as bad as we thought,’ ” Kaine said. “Over the course of the next four weeks, even that level of differential did become problematic.” This account differs from that offered by Davis, who says FTA officials never put much stock in the engineer panel’s tunnel estimate.

Some critics argued that an elevated track is so undesirable that Fairfax should revisit the possibility of serving Tysons with a separate streetcar line linked to a Metro line running straight to Dulles.

But a few officials argued that going with an elevated track should not be seen as disastrous because the decision for that design was made years ago after much deliberation. It was only recently that a tunnel again became an option, when Metro officials raised the possibility of using a potentially more efficient tunneling method that had not been considered before.

“None of these decisions were made in haste,” said J. Kenneth Klinge, a political consultant who chaired the initial task force to plan transit in the Dulles corridor in the late 1990s. “When this idea, that you could build a tunnel more cheaply, started six months ago, I remember saying: ‘These people are crazy. I don’t know if they understand what they’re getting into. The federal rules are going to kill this project if they keep this up.’ “

Author: Mark Kleiman

Professor of Public Policy at the NYU Marron Institute for Urban Management and editor of the Journal of Drug Policy Analysis. Teaches about the methods of policy analysis about drug abuse control and crime control policy, working out the implications of two principles: that swift and certain sanctions don't have to be severe to be effective, and that well-designed threats usually don't have to be carried out. Books: Drugs and Drug Policy: What Everyone Needs to Know (with Jonathan Caulkins and Angela Hawken) When Brute Force Fails: How to Have Less Crime and Less Punishment (Princeton, 2009; named one of the "books of the year" by The Economist Against Excess: Drug Policy for Results (Basic, 1993) Marijuana: Costs of Abuse, Costs of Control (Greenwood, 1989) UCLA Homepage Curriculum Vitae Contact: Markarkleiman-at-gmail.com

6 thoughts on “Why cost-effectiveness is sometimes the wrong decision rule”

  1. "To tunnel supporters, it's simple: Public transit projects are being held to cost-benefit standards more stringent than those applied to highway projects. Federal funding of new transit projects is $1.5 billion a year, a fraction of highway spending."
    And new transit projects carry a fraction as many people as highways do. Is it too much to ask that the relative fractions be compared? I'm pretty sure that's elementary school math, which shouldn't be beyond reporters…
    "Some critics argued that an elevated track is so undesirable that Fairfax should revisit the possibility of serving Tysons with a separate streetcar line linked to a Metro line running straight to Dulles."
    Ok, I'm just a dim witted engineer, but I'm having a bit of trouble figuring out why a train 35 feet up in the air, on a track that's supported every 70 to 100 feet, is more of an obstacle to pedestrians and vehicular traffic than a streetcar system with the same capacity, down at ground level. Care to enlighten me?

  2. "[W]e will never be well-governed until policy analysts run the government (a horrible thought) or politicians learn at least the basics of policy analysis."
    I don't recall Plato saying that.

  3. I don't know that I disagree with your cost effectiveness point, but there are truly some mega differeces between route 7 in Tysons and the Rosslyn to Ballston corridor. True, there was a lot of redevelopment in that corridor post metro, but we are talking a pretty massive distance here.
    Tysons was a fiasco when I worked there during the 1980s. Route 7 was sure the epitome. I could forsee this line running out to Reston and Dulles maybe blowing through Tysons on its way to elsewhere. Tysons is a huge sprawl, and route 7 really isn't where I remember most of the semi rise offices being.
    The above ground metros, such as the yellow line through Alexandria are above ground and were done without a lot of negative impact to pedestrian traffic. I worked on Eisenhower Ave when that station opened.
    Metro all the way to Dulles. Sounds more like a job for Amtrak!

  4. Seattle is writhing in the same agony. 50 years ago the city gave the state land on the waterfront to build what then served as a 'freeway'. In those days the city elites regarded Seattle as wasteland, the place in which unsightly and toxic people and businesses should be concentrated to magnify the contrast with the shining suburbs on surrounding hills.
    Today, that highway must be rebuilt or torn down, and naturally the state has no interest in relinquishing lands now worth billions.
    Tearing down the Seattle 'Embarcadero' (we call it the Viaduct) or replacing it with a tunneled roadway is the first and only choice of anyone with a brain that has entered the 21st century.
    Ironically, Seattle, the city of 'Century 21', has become a poisonous stew of would-be radicals who regard any public improvement as a conspiracy by rightwing millionaires, and rightwing millionaires who regard any public improvement as creeping socialism.
    Nobody can help them, they see their fate approaching but are powerless to resist. It is a terrible urban tragedy unfolding in the slowest of motions. In all probability, when it is done, the 'hipsters' and talk radio will have fastened a concrete necklace on the waterfront for the next century, a road that will be functionally obsolete and have no purpose within a decade of completion.
    It's like watching a city die of cancer.

  5. It seems to me that an elevated rail line does not need to produce a desert around it; the EL in Chicago certainly doesn't.

Comments are closed.