Image by Airman 1st Class William Johnson.

The Purple Line may be dead after federal judge Richard Leon refused to lift his ruling blocking the project. He says the FTA hasn't looked hard enough at some facts he decided are important. But this kind of reasoning seems to give a judge nearly unlimited power to stop projects they personally aren't convinced about, despite rules they are supposed to defer to agencies' judgment.

Last year, Leon responded to opponents' 7,182nd attempt at finding an argument against the Purple Line: their claim that declining Metro ridership required the Federal Transit Administration do redo part of its Environmental Impact Statement (EIS). The FTA said it thought its prior analysis was sufficient and the EIS didn't need to be redone. On Monday, Judge Leon nevertheless refused to let the Purple Line project move forward.

Judges are supposed to give "deference" to administrative entities like FTA in interpreting their own rules. However, under administrative law, the agencies can't act "arbitrariily and capriciously." This is a doctrine known as "Chevron deference" after a case involving the oil company in 1984.

Leon claims that it is "arbitrary and capricious" for the FTA not to do further analysis of the impact of declining Metro ridership. He acknowledges that the court "cannot substitute its judgment for that of the agency," but environmental law requires agencies to "carefully consider" the environmental impacts and prepare a supplemental EIS whenever "there are significant new circumstances or information relevant to environmental concerns" which "provides a seriously different picture of the environmental landscape."

The court is very restricted from deciding that an agency is wrong, but can "ensure that the agency has taken a hard look" at the environmental consequences." Largely, Leon says in this ruling, the FTA hasn't looked "hard" enough at his concern. His reasoning here is quite dubious.

"You didn't answer every counter-argument"

Another big danger in this line of legal reasoning is that it's very easy for a judge's own subjective view on the merits of the project to override a dispassionate legal analysis. If a judge is skeptical about a project, well, they can't supersede an agency decision and rule against it, but if they hear an argument from opponents that sounds compelling, hm, the agency doesn't seem to have sufficiently explained this in its materials! We need to remand!

There's little way to prevent judges from having outside views on any issue—if they're ruling on health care laws, what, we need people who never go to the doctor? But it's incumbent on judges to also ensure when doing something like asking an agency to do more analysis, that's on a sounder footing. Especially when, as in Leon's case, he lives in the affluent and exclusive part of the county which is generally fighting a project that would help people from the eastern side of the county and Prince George's access jobs in Bethesda. Even if he has no actual conflict of interest, it's very human to come to better understand your neighbors' and friends' points of view than those of other populations farther away, and thus give one more weight.

An agency analysis can't look at every single possible everything that might be criticized. Already, EISes are massive and take years to prepare because agencies have to try to consider virtually everything. But no matter how much they consider, sufficiently diligent opponents can probably think up something they didn't consider all that thoroughly... even if it's scarcely related to actual environmental harm.

There needs to be some reasonableness, but there are seemingly few clear rules about how much the judge can... not substitute their views for the agency, but decide that something needed to be analyzed even more.

The FTA gave a harder look; it just wasn't enough for Leon

Though the FTA doesn’t believe an SEIS is necessary, it did provide an updated ridership analysis which Judge Leon nevertheless found a way to reject. Among his reasons: that the FTA didn’t predict which of the scenarios is most likely to occur.

This makes no sense: even under the worst case scenario, there was still sufficient ridership to make the project worthwhile. Leon said the "purpose and need" of the EIS included Metro connections, so therefore, in essence, that's a mandatory part or else the project isn't valid. (This isn't how EISes are supposed to work.)

This kind of dubious logic not only leads to bad outcomes but emphasizes the catch-22 here; a judge who doesn't like a project can nitpick the EIS to death, and the longer the EIS, the more he can find one part which says something you can interpret as contradicting something else.

Echoes of McMillan

We saw something very similar with the DC Court of Appeals' decision on the McMillan development. Much of that opinion consists of the judges, Stephen Glickman, Kate Easterly, and Roy McLeese, saying that the Zoning Commission hadn't sufficiently explained its reasons for prioritizing some Comprehensive Plan elements over others. Some of the criteria, like that development could contribute to gentrification (um, nobody lives there now; new housing will only stabilize prices, not raise them), hadn't been significant factors in the zoning hearings for the project. The court in essence said that if opponents raise any objection to a project, if the judges themselves (who aren't planning and zoning experts or economists) think it sounds good, they could rule that the agency hasn't explained itself enough.

There's no clear roadmap in McMillan for how much explanation is enough. It's really just whatever is enough to convince Glickman, Easterly, and McLeese, but nobody can sit down with the three of them and try to clear things up; they have to file formal legal briefs on each side and so long as the three aren't persuaded, they can keep remanding for more information, delaying the project for years. There, like with the Purple Line, opponents threw every argument against the wall until one stuck.

I strongly suspect that these judges on this project have heard neighbors talking about "maybe there's too much development" in the city, read some articles in the popular press about displacement, and perhaps feel they personally haven't seen enough to convince them to support the project. But judges aren't planners; their background isn't in the field and rulings like this sound an awful lot like "I'm just not convinced I like this project" instead of "there's a real legal problem."

The same goes for Leon. If he personally is inclined to favor anti-Purple Line arguments, as seems to be the case, there's no certainty he couldn't say that a supplemental EIS isn't enough of a "hard look" as well. Probably if the FTA spends years doing a hundred-page follow-up that would be enough, though of course by then maybe there would be some other new information they have to consider. Or maybe not. In the Durant DC zoning case, which is similar to McMillan, the court sent it back to the Zoning Commission three times for being insufficient.

Certainly, sometimes governments try to blast ahead with a project that will legitimately harm the environment. Here, the Purple Line will only be a boon to the environment by reducing car trips by far more than any harm from cutting down a few trees, which will be replaced elsewhere anyway. It doesn't matter if the ridership is wildly off; the environmental benefit of this rail line far outweighs it. But it doesn't seem to really matter that the agency is entitled to deference if a judge doesn't like the project that much.

Decide projects in a public process, not the courts

Both McMillan and the Purple Line went through extensive debate in the political system at multiple levels of government and over many years. Both gained support from executives, legislatures, and agencies after detailed hearings and debate. It's just not right for judges to be making such broad public policy rulings, at such late stages of projects, and with such financial impact, at least without good reason. Here, their reasons just aren't strong enough.