Friday, August 28

Roundup of HSR discussion

I've been sucked into the vortex of High Speed Rail cost/benefit analyses this last week. The backlash has begun against Obama's decision to supplement the usual torrent of highway spending with about $15 billion in rail infrastructure improvements.

In the online New York Times, Edward Glaeser published a series of posts on a hypothetical HSR line between Houston and Dallas, arguing that the economic benefits don't pencil out. Eric Morris used the Times' Freakonomics blog to say basically the same thing for HSR in general. Then Robert Samuelson wrote an op-ed in the Washington Post on what he sees as the 'boondoggle' of rail. Two of the three, approvingly reference the perennial Randall O'Toole.

Prominent bloggers were on this right away. Ryan Advent responded to Glaeser blow-by-blow in StreetsBlog, and Yonah Freemark took a closer look at the numbers Glaeser used in the Infrastructurist. I should add that these criticisms were echoed by a number of really perceptive commenters. Here are some of the conceptual problems they found with the analysis:

  1. You have to consider alternatives. Glaeser and Morris compare the cost of implementing HSR to the cost of doing nothing, but, given projected population increases, doing nothing is not a viable option. Full cost accounting of continually widening highways and adding more flights needs to be part of the equation.
  2. Energy prices are a crucial factor. All are in agreement that HSR is more energy-efficient, so it's performance relative to other modes will inevitably be accentuated if energy costs rise. Glaeser bases his numbers on the assumption that energy costs will remain steady for the next 20 years, but there is no reason to believe this assumption is true.
  3. Just because conditions are not right everywhere right now doesn't mean that are not right anywhere. The Houston to Dallas route Glaeser uses to debunk HSR is not currently being proposed. Samuelson brings up the nonsensical fact that U.S. has an average density of 86 people per square mile, which makes it unsuitable for trains (I'm sure the whole world has an even lower population density, making the world unsuitable for trains). The point is that some places are denser than others, and those are places where HSR is being proposed.
  4. Major infrastructure improvements are always long-term investments that may not reap a return for many years. We have been willing to be more patient with other major projects in the past, including the federal interstate system.
Without pointing any fingers, it's worth noting, as this conversation continues, that there are enormous vested interests involved in our transportation system. Right now the average American household spends over 19% of the budget on transportation - mostly cars. That adds up. When Obama talks about "removing x number of cars from the road" with HSR, some people hear reduced congestion but other people hear loss of market share.

For the story of the automobile industry's early public awareness campaign to defeat rail, see the book Getting There. For their public awareness campaign to displace pedestrians from city streets, see Fighting Traffic.

Like always, this just means you need to make up your own mind about things.


Eric Orozco said...

I wish we were doing bolder moves on number 4. Instead we're spending our stimulus money on cheap tricks and worse than short term solutions....partly due to fact we got an economist to bat down every forward-looking proposal that has a chance.

I always take what an economist has to say with a grain of salt. These are the same people, mind you, who failed to warn us about the housing bubble. The kind of economists I will listen to are those who are daily exposed to development issues and regional planning.

On the latter point, we need more transportation and planning expertise in economic fields. Unlike many economists, planners first look at things from a whole systems point of view. Many economists are trained to approach things from the opposite vantage. But planners do not look first at ways of delimiting things. Economists like modeling, but models simplify. Models only tell you a part of the picture, and, if used spuriously and in fields economists know little about, their conclusions are likely to leave out crucial pieces of the story and miss their implications.

Daniel said...

I hear you Eric. Especially for something as complex and long-range as an entire transportation network, the level of precision to this 'analysis' seems way out of place.

Jebediah, who runs Infrastructurist, had this great nugget buried deep within the comments:

"So, I suppose, the answer is I don’t care if the numbers are “accurate” because I think that whole idea is meaningless. Glaeser’s whole exercise was sophistry. Because neither you nor I nor Ed Glaeser has any freaking idea what 2030 is going to look like, and I will say categorically that almost regardless of the conditions that actually prevail the US will be a better place if we build in a decent passenger rail network.

I think back to something Robin Chase, the Zipcar founder, said when Cassim Shepherd and I interviewed her a couple of months ago: you build good things that will last and prove adaptable and continue to be useful well into the future. That’s how you plan and build your society. And you don’t pretend to know more than you do, which is precisely what Glaeser is doing."

epar said...

It's really breathtaking to read that Samuelson actually cited the average national density figure to claim that trains are not a feasible transportation option. How can the WaPo still publish him? Do they have no standards?

I actually expect something of Glaeser, though, since he's known for his good housing and policy work. Inevitably he'll dismiss the criticisms as the ramblings of some know-nothing bloggers.

As a side note, I think this debate demonstrates precisely why newspapers are doomed. They are miserably failing at their job of proving informed commentary.

Eric Orozco said...

Great words from Jebediah, Daniel

Epar, for so long housing has been tied to auto-dependence that it is no surprise to me that Glaeser adopts some of the argumentation he does (HSR will create more sprawl). For him, rail died in the 20's. Of course, we're never going to become as streetcar dependent as we once where, but that doesn't mean people won't adapt to new conditions and infrastructures, and discover easily how make the best use of them.

NeilSWilliamson said...

As usual, excellent balanced comments from this blog. While I believe HSR may actually work in some areas, I believe when penciled out (even with your corrections) these will be the exception rather than the rule.

This does not mean the concept should not be explored. I see the future of American transportation to continued to be tied more directly to roads than rail.

This is driven by America's multi-generational desire for personal mobility. This is a very different (some may say selfish) mindset than Europe.

In the end some HSR projects may economically woprk, perhaps, but the vast majority of the proposals I have seen are both short and long term economically unsustainable.

Daniel said...

Thanks for commenting Neil. I agree with you that Americans will choose car-based living into the foreseeable future, but the margins are where all the action is.

For instance, if you'll excuse some crude math, say we currently have about 90% car-based and 10% transit/walk based housing stock. If that were to shift to 85% car-based and 15% urban, that would mean lots more urban than suburban development would have to be built. Even if, by raw numbers, suburban is still dominant, most new growth (and developers income) could still take a different shape and the federal government would want to shift their share of funding to reflect this shift.

CarFree Stupidity said...

Randal O'Toole is the Anti-Planner. Maybe you already know this, but I figured it couldn't hurt.