Welcome to the new version of European Tribune. It's just a new layout, so everything should work as before - please report bugs here.

Glaeser Hacks up the Numbers on HSR

by BruceMcF Fri Aug 7th, 2009 at 03:17:09 PM EST

Last things first ... after reading and commenting here, go ahead and comment at Running the Numbers on HSR by Edward Glaeser.

This last weekend, I looked at a low-brow attack on HSR by John McCarron in the Chicago Tribune. This week, I look at a high brow attack by the economist Edward Glaeser at the NYTimes "Economix".

However, the attack by Edward Glaeser is different. Even if some suspect a partisan motive, given Glaser's support for McCain ... this is not the kind of hackery we are seeing in the health care debate, where paid partisan hacks are just blatantly lying. Its the kind of hackery that is embedded in a frame, and which will bias the results of any honest analysis done within that frame.


Burning the Midnight Oil for Living Energy Independence

The hackery, in other words, is almost entirely in his framing of the problem, and the framing is a quite mainstream economic framing. Which means that shining a light on the biases of this analysis has the potential to give insight on how to pushback against a whole host of highbrow academic attacks on HSR in particular and public investment in useful infrastructure for a New Energy Economy in general.

Glaeser pursued an analysis that was biased against capital intensive transport ... and ignored the most capital efficient means of providing HSR between Dallas and Houston, as well as the most capital efficient alignment for providing Express HSR between Dallas, Houston, and San Antonio-Austin. It is taking the mainstream definition of economics as the study of the allocation of scarce resources to unlimited human wants and needs ... and adding a rider, "plus, ignore the most effective means of pursuing the particular choice you are talking about."

And, yes, clearly Glaeser timed his piece for the day before my first class of the new term, so that I am responding to his piece days and days after it came out. Which means I have not chance of getting linked into at the immediate reactions ... but on the other hand, gives me the benefit of being able to use the immediate reactions.


Basic Skeleton Outline of a Basic Skeleton Analysis

Its a short piece, so summarizing it makes it really short:

  • The project is justified if benefits more than covers costs
  • Immediate benefits are the benefit to the riders on the HSR
  • Costs are operating costs and corridor construction and maintenance overheads
  • In other words:
    Number of Riders times (Benefit per Rider minus Variable Costs per Rider) minus Fixed Costs.

  • Benefits are $40 for a saved hour of travel, $80 saving on ticket costs, and say $20 for added comfort and amenity, $140/traveller less variable costs of $72/trip, for a net benefit per rider of $68.
  • Times, Glaeser guestimates, 1.5m riders, is $102m for net benefits from the trips.
  • Based on a mid-range of costs from a CBO study of Express HSR, and mid-range maintenance costs per mile of corridor in European experience, Glaeser arrives at fixed costs of around $650m.
  • So unless there are external benefits adding up to around $550m/year, its not a worthwhile investment.


Critiques Glaeser's analysis

More than a few people have critiqued Glaeser's argument.

Ryan Avent has addressed Glaeser writing on HSR on two days in a row for Streetsblog Capital Hill: Glaeser takes an Unserious Look at High Speed Rail on Wednesday, which our own Robert Cruickshank ("eugene") used as a springboard at the California HSR blog, and then on Thursday looking at Glaeser channeling the "False Contest between HSR and Local Rail" talking point in Missing the Point on High Speed Rail. Oh, and at very least click through to Ryan's pieces ... they deserve the hits far more than Glaeser's piece does.

In the Wednesday piece, Ryan Avent considers the corridor chosen, and also the considerations that the analysis omits:

Why would he choose this corridor to examine? Why not begin with the most natural place to construct true HSR -- the Northeastern Corridor -- or the state moving fastest toward building its own true HSR network -- California?
...
What are his long-term assumptions? How quickly does he think the population of the Dallas and Houston metropolitan areas will grow? What will that population growth do to the number of people living within easy reach of a train station? How will that population growth interact with planned expansions of local transit systems?

How sensitive are his projections of changes in oil prices? Do they take into account the effect of changing demographics on demand for various kinds of housing and transportation?
...
And that brings us to a final point (which, again, Glaeser may ultimately address): What is the proposed alternative?

Is it doing nothing? Then at what point does the rising cost of congestion justify construction of something? Let's say an alternative is new airport capacity; well, how do the costs and benefits there work out, and how does that math change with oil at $150 per barrel?

Or perhaps an alternative is new highway capacity. Can we see a cost-benefit analysis for that, and how that varies with oil prices, congestion levels, and so on? If we assume that drivers will need to pay the full maintenance cost of the highway network already constructed via a user fee (and currently they're coming up well short), what does that do to expected demand for rail?

Even if you accept the numbers that Glaeser uses (and one shouldn't automatically do so), you're left with almost nothing -- an amateurish, back-of-the-envelope analysis for a corridor that's not even part of the current Obama administration plan. What is this supposed to prove, exactly?

Of course, this is just one of Glaeser's peices. In Friday in the Boston Globe, Glaeser trotted out the argument "invest in mass transit instead of HSR", as if killing off HSR is going to do anything other than reduce the ability of mass transit advocates to gain increased funding. And, despite despite Robert Cruickshank's title to the contrary, it was in this piece on Thursday that Ryan Avent truly and completely demolishes Glaeser's analysis. Indeed, consider it embedded in full right here, and if you do not have time to read it in full and also the remainder here, then click through now, and thanks for stopping by.


Why is it Fair to call this a Hack Job?

OK, welcome back.

Now, my real focus here is how to recognize that Glaeser's analysis of HSR cost and benefits is not just wrong, but is, indeed, a hack job. That is, the analysis is situated in a frame that is commonly used by mainstream economists when analyzing a new public investment that is not considered to be a status quo, "normal" government investment ... and that framework is egregiously flawed even from within the mainstream perspective.

That is, being a mainstream economist brings with it a lot of trained incapacities in analysing many aspects of the economy ... but for one of the biggest ones that Glaeser lapses into, being a mainstream economist is absolutely no excuse. Blind as they are to so much about the economy, even mainstream economists are supposed to be equipped to see this particular aspect of the economy.

Glaeser is assuming

  • On the one hand, that the economy will be continuing largely as it has over the past twenty or thirty years, while, at the exact same time
  • assuming that the population of both the Dallas / Fort Worth metropolex and Houston metropolitan areas will be stagnant.

How can we see that Glaeser is assuming that the economy will be continuing largely as it has over the past twenty or thirty years? Because Glaeser completely ignores the car trips that will be diverted onto HSR. And we know that if gas prices spike, one of the first reactions will be a spike in rail travel where it is available. People may not be able to quickly change their daily commute, but for "that trip", the price response of longer distance car travel is much more sensitive to gas price shocks.

However, make the technological cornucopian assumption that the economy will continue along pretty much as it has over the past twenty or thirty years, despite the fact that its physically impossible: Houston and Dallas have been among the faster growing metropolitan areas in the past twenty or thirty years.

Yet, Glaeser tacitly but quite directly assumes that if the HSR corridor is not constructed, and we do not provide transport capacity, then all of those construction costs are saved. Yet, in the real world, if large metropolitan areas within 300 miles of each other grow in population, so that the transport demand for travel between the two grows, that leads to political demands to provide for more infrastructure to support those trips.

The growth creates more travel, which creates congestion on existing infrastructure, which leads to political demands for more transport capacity.

And this is not something that can be "added onto the analysis later" ... embedded in the direct benefits of the rail travel is the assumption that air transport is not congested, and indeed that road transport is so uncongested that nobody presently driving will switch to the train if it becomes available.

Now, that is an absurd assumption. The fact is, when you limit people's choices, then people making a particular choice like driving include those for whom driving offers strong net benefits, and also includes those for whom driving imposed substantial costs. The benefit of the HSR is not the difference between the average benefit to all motorists if they all switch ... its the average benefit to all motorists who benefit enough to make the switch.

Which is why the assumption that Glaeser makes that HSR takes over all existing Houston/Dallas air travel is such a sneaky assumption ... it sounds so generous, but in reality is so clearly understating the potential ridership. A one and a half hour rail trip is likely to take over over 70% of the previous air transport market ... but fewer than half of the passengers will be those who would previously have flown. The rest come from people switching from cars, switching from buses, and new trips generated by the opportunity to take a more convenient, faster, and cheaper trip than flying.

So, maybe we are going to be experiencing sufficiently massive disruptions that growth in Houston and Dallas will be halted, in which case the HSR will have a much bigger share of the transport market than Glaeser's analysis anticipates ... or it will be business as usual, in which case Glaeser's analysis assumes that the airports and roads between the two metropolitan areas are Magic Land Fantasy infrastructure that grows itself for free.

So even if Glaeser simply refuses to recognize the possibility of dramatic change ... that implies he is making absurd assumptions about the infinite flexibility of existing road and air transport infrastructure.


Why is Glaeser engaged in Ignorant Misinformation

Now, this is, of course, not to say that this is Glaeser's only fault. He also engages in substantial misinformation in this piece.

The the fact that it appears to be the intellectually lazy kind of misinformation from not bothering to inform himself about the topic of his piece, instead of the deliberate misinformation of the kind of lies being spread in support of Big Pharma, Big Insurance and Big Coal ... well, he's getting paid to write this stuff, so he does not have an excuse for being intellectually lazy if he chooses to pick up a topic.

As Ryan Avent noted above, why Houston/Dallas? If you are going to ignore the impact on property values of having an HSR station in the vicinity, and ignore the protection against the risk of an oil price shock, and ignore the substantial number of motorists that will have a net benefit, and ignore the benefit to the businesses who pay for business travel of allowing businesspeople to get more work done, and are going to ignore the capital cost savings from not having to build more highway lanes and airport runways and terminal capacity ... if you are going to ignore all of the benefits that come from putting on the fastest HSR trip possible, which will attract the most ridership ...

... then why choose the massive capital costs of an Express HSR corridor?

The very same Government Accountability Office  report that Glaeser turns to for his estimate of $40m per route mile for an Express HSR corridor, reports that Emerging HSR corridors cost from $4m to $11m per route mile. Taking, following Glaeser's approach, a mid-range cost estimate of ~$8m, that means that after ruling out of consideration all of the reasons one would pursue an Express HSR over an Emerging HSR ... he analyses the more capital-intensive Express HSR anyway.

Cut the capital costs by 80% ... and that's ~$85m/year of benefit to generate. And then, for a much wider range of net benefit per rider and total ridership assumptions, there is a clear net benefit for Glaeser's "stagnant population and other than that business as usual" assumption:

0.5m1m1.5m2m
$25/rider(-$72.5m)(-$60m)(-$47.50m)(-$35m)
$50/rider(-$60m)(-$35m)(-$10m)+$15m
$75/rider(-$47.5m)(-$10m)+$27.5m+$65m
$100/rider(-$35m)+$15m+$65m+$115m
$125/rider(-$22.5m)+$40m+$102.5m+$165m
$150/rider(-$10m)+$65m+$140m+$215m

And now, supposed that an expansion of transport capacity is required. Obviously, at full capacity, an Emerging HSR corridor is cheaper per seat-mile capacity than road or air infrastructure, but suppose, conservatively, that the new transport capacity of the Emerging HSR corridor substantially exceeds the incremental cost of new road and air capacity, so that there is still a net capital cost of $25m/year. Even under that conservative estimate, the prospects look quite promising in the "rose colored glasses" scenario of things keeping going the way they have been:


0.5m1m1.5m2m
$25/rider(-$12.5m)$0m+$12.5m+$25m
$50/rider$0m+$25m+$50m+$75m
$75/rider+$12.5m+$50m+$87.5m+$125m
$100/rider+$25m)+$75m+$125m+$175m
$125/rider+$37.5m+$100m+$162.5m+$225m
$150/rider+$50m+$125m+$200m+$275m


Looking Ahead

With nothing more than Glaeser's hackneyed analysis, there is no way of knowing whether Express HSR between Houston and Dallas is justified as a stand-alone project. This would require establishing a range of population growth scenarios, and developing reasonable estimates of the cost of constructing the alternative transport infrastructure under those scenarios.

However, an Emerging HSR corridor ought to be able to put an Express version of its service through between Houston and Dallas in three hours. A three hour train ride will typically capture around 40% of an existing air transport market. And, of course, the Dallas/Fort Worth metroplex is not a typical air transport market ... it is a massive sprawling region encompassing a number of counties, which is far more easily served by five or six HSR stations than by one or two major airports.

Now, in the oil price shock, slow growth scenario, it seems like 1m trips on a mode of inter-urban transport that has far less exposure to oil price shocks ought is a conservative anticipation, as is a $100/rider benefit ... precisely because of the inflated cost gap between air and road on the one hand and Emerging HSR rail.

And in "business as usual" mode, its highly conservative to assume that an Emerging HSR has any incremental capital cost to cover, but even if it does, very modest assumptions about the average net benefit of those who benefit sufficiently to switch from air or car would cover those costs even at 0.5m passenger per year.

So it looks to me that if the numbers for a stand-alone Dallas/Houston Express HSR do not add up, the numbers for a standalone Emerging HSR corridor are very likely to do so.

So, in other words, it is quite amateurish of Glaeser to ignore the HSR option for two large metropolitan areas within 300 miles of each other. Fortunately for us, the Obama administration does not make the same amateur's mistake, and has positioned itself to support investment in whichever kind of HSR is suited to the variety of conditions that are found in inter-urban transport markets around the country.


What about the Texas T-Bone?

As a final note, as I have previously explained, the present momentum in Texas for an Express HSR corridor is "T" shaped Express HSR corridor between San Antonio/Austin in the South, Houston in the Southeast, and Dallas in the North.

So, why didn't Glaeser analyze the actual Express HSR proposal that is currently most likely to have an application?

I have not idea whether it is laziness or lying, but based on the balance of his piece, I am willing to assume laziness. As Napoleon is reputed to have said, never assume malevolence for what can be perfectly well explained by incompetence.

However, note that in terms of the "riders to cover fixed corridor costs" ... when the trains are going at 220mph, the "right angled" route from Houston will still leave the Express HSR trip as faster for most passengers than air travel in door to door travel time ... while, obviously, substantially cutting the capital cost by sharing the Fort Hood to Dallas/Fort Worth leg with the Austin/San Antonio trains. And then the Houston/Austin-San Antonio route is basically available for no additional capital cost.

So Glaeser pursued an analysis that was biased against capital intensive transport ... and ignored the most capital efficient means of providing HSR between Dallas and Houston, as well as the most capital efficient alignment for providing Express HSR between Dallas, Houston, and San Antonio-Austin.

Which is a head scratcher that suggests to me that I should conclude with a quote from Ryan Avent:

Ed Glaeser is a fantastic economist. He has done magnificent work analyzing the economics of urban growth and written indispensable papers on the connection between housing regulations and migration.

But when the man picks up his pen to write a piece for public consumption, he tends to take complete leave of his senses.

Display:
... of reputation and hack-job analysis trading on that reputation, it tends to spawn a lot of online reaction.


I've been accused of being a Marxist, yet while Harpo's my favourite, it's Groucho I'm always quoting. Odd, that.
by BruceMcF (agila61 at netscape dot net) on Fri Aug 7th, 2009 at 03:45:39 PM EST
It reads almost as if he had a clever but inexperienced, first year masters student with a research assistantship write the piece for him without really engaging in the final draft before it went to print.  Very embarrassing, but since "journalism" gets such little credibility among academics at his level, he won't find much professional fallout from it as he would have if he was presenting a paper at professional forum.

In addition to your valid criticisms, the general issue of treating transportation as primarily a business problem regarding transporting people economically, instead of as a real estate development business problem, also biases the findings against rail. (I think you're saying the same thing in different terms, actually.)

The history of rail is not a history of the business of transportation but rather a history of the business of real estate development.  The immense fortunes gained by the rail barons of the 19th century came from their real estate investments which were improved by connecting rail lines to them, almost all of which were operated at long-term economic losses.  The same model works, even if indirectly, for public financing of rail investment today in terms of offsetting public debt incurred with real taxable assets with which to repay public debt. That's why it's usually "profitable" for regional polities to subsidize transportation such as rail: they gain even more net revenue (which means either lower future taxes or better services) from the concentrated economic development opportunities that occur along the lines compared to what would exist without rail.

by santiago on Sat Aug 8th, 2009 at 02:32:40 PM EST
Glaeser's analysis is not commercial, per se ... he is trying to estimate benefit, not revenue ... but is stovepiped within the same limits as a commercial analysis would be.

Ryan Avent made an excellent critique of many of those limitations of the frame of the analysis.

However, even within the limits of the analysis, its hackery to assume business as usual in an area that has been growing but ignore the impact of population growth on benefits from new transport infrastructure. Its hackery to ignore the motorists, and evade the difference between the benefit to those who switch compared to the average benefit if a random sample switch. Its hackery to ignore the more capital efficient means of constructing an Express HSR alignment, and the far more capital efficient means of providing rail service within the HSR program that Glaeser is trying to undermine.

Its a limited analysis, but given its limits, its still a crappy job.


I've been accused of being a Marxist, yet while Harpo's my favourite, it's Groucho I'm always quoting. Odd, that.

by BruceMcF (agila61 at netscape dot net) on Sat Aug 8th, 2009 at 04:23:04 PM EST
[ Parent ]


Display:
Go to: [ European Tribune Homepage : Top of page : Top of comments ]

Top Diaries