The difficulty and pitfalls of transport problem solving
Transport systems are systems, and solving problems in systems is hard
Hey there! If you’re reading this, it means you’re a paid subscriber, which means a huge thanks to you for supporting my work, documenting, advocating for, and explaining transport systems all over the place. It means a ton, especially in these chaotic days!
Since I’ve gotten back from London, I’ve seen a number of pretty opinionated pieces on a variety of transportation projects that, in my opinion, reveal some common problems we can face when trying to analyze such projects. One piece discussed California High Speed Rail which I’ve criticized in the past - so let’s dive into that project and see some of the problems I’m alluding to.
The obvious disclaimer I need to provide and have provided historically is that I support high speed rail in California. It’s a fairly densely populated state, with several major cities with fledgling transit networks and booming economies mostly lined up, high speed rail makes sense! That said, too often in the transit advocacy community, support for a project concept is seen as needing to translate to support for a project. It’s seen as bad if you support high speed rail in California, but not California High Speed Rail. It’s sort of an issue of tough love: once you’ve established that you support the idea behind a project, you should be free to critique decision being made on an instantiation of that project.
Now, when actually looking at a project (California has a number like this - San Jose BART comes to mind as well), a common thing I see from advocates is to place blame for any project problems on external forces. This is often on consultants or construction firms, and rightly so in many cases, but it’s important to remember that state agencies not being able to manage consultants and constructors cause problems with higher levels of project involvement too.
Of course, many issues - like legal ones, are external to a project and in such cases I think the real problem is having any entity which has disparate elements of itself rowing in different directions. But at the same time, many problems with projects are made for shockingly minor reasons such as preference, or belief in something that hasn’t actually been validated.
A common issue is accepting all justifications at face value, something that should not go either way. When a freight rail company says “we can’t accept passenger traffic on our lines” they are typically rightly called out, but when CAHSR suggests the only way to serve the states central valley cities is via the mainline, it seems to be seen as taboo to question it. Transit agencies might be doing something we like, but that doesn’t mean they are doing it in the most effective way possible — we can and should be asking these questions.
That really ties nicely into the next issue here: while it’s often critiqued as overly pragmatic or minimalist, there are limitations to how much we can achieve at once, and the solutions to problems you arrive at when you accept that fact are often dramatically different to when you do not. Having CAHSR (or your chosen other very expensive project) cost 5% less might save several billion dollars — this is enough money to build a decent sized light metro system at reasonable costs, something which would have immense benefits in the right place.
At the same time, a project which is expensive cannot have the same flaws as one which is less expensive. A “flaw” is often something which needs to be seen as having some detracting or negative value, so some flaws are acceptable if they reduce cost or increase benefits. Too often, a project which is expensive does not deliver sufficient benefit to offset the additional cost. This ties into the issues of suggesting projects are worth it at any cost — which isn’t always a shocking conclusion to come to if you’re an advocate, but simply isn’t the type of thinking that leads to great outcomes, even if projects do end up actually getting completed. Sometimes this at all costs thinking is associated with a politician or other figure who pins their legacy to a project and this is something we really do need to try and avoid: some level of political interaction with projects is to be expected and is probably actually good (future post on that), but allowing someone to have their personal legacy too tied up in, or to have a project which is too easily sandbagged by a given politician, is bad.
I think a lot of the time, such issues can be traced back to a tendency to try and create overbuilt perfect projects, which meet every objective initially stated, and do not do anything that isn’t wanted (say, annoying NIMBYs). The issue with this approach is that it implies that all project features are equally valuable, and often doesn’t consider things like timeline and cost, very much elements of any infrastructure projects, as an important project feature. The issue here is that it’s impossible to have a good high speed rail line that a train never runs on, and while a never completed project never provides any substantial value, one which is imperfect does, and can always be improved later. The other issue with trying to build impossible perfect projects is it creates a temptation to reinvent the wheel, to try to effectively get blood from a stone. This leads to things like CAHSR using a pretty uncommon top speed of 220 mph, which seems like a less reasonable goal to aim for for your first major high speed rail line (or passenger rail line of any kind). Another example of this is San Jose BART’s giant single-bore tunnel, which seems to make virtually all aspects of the project worse or stay the same, but was seemingly adopted because it seemed like it was able to do the impossible, and abandoning it now is seen as walking back a commitment. This sort of escalation of commitment is exactly the type of problem you see when the excessive political tie in and extreme project goals come together. Of course, all of this wraps back to the issue of projects which are not perfect — but actually completed — being seen as equal or inferior to projects which are perfect but never usable. This sort of feels like the nirvana fallacy’s instantiation in the transit space.
Another major issue I see with contentious transit projects is a sort of project-integrated system of poison pills. One example that comes to mind is the Scarborough Subway Extension (a project I support — well, with some reforms) in Toronto, which is a project that often gets talked about as competing with the Eglinton East LRT (a project which I don’t support — except maybe with some reforms). As it turns out, now that the Scarborough Subway Extension has started construction, the design has been set such that the Eglinton East LRT cannot be built in the optimal way. At first this seems like it may be gross incompetence, but I get the sense that this might be a way of putting poison pills into the LRT project. This is bad even when it makes a project I dislike less likely to happen, because for one, it makes our overall options for projects narrower and worse, and for two, because allowing this type of stuff to be normalized gives everyone the ability to do it — including if the tables turn. A similar situation exists with California High Speed Rail: I’ve heard the justification that building the central segment first is key and a smart political move because it will sort of “shame” future politicians into finishing the project lest they have to preside over a massive white elephant. The issue I have with this is that firstly, the central segment of CAHSR is an order of magnitude simpler than the sections through cities and mountains on either end (and is still faced with lots of cost and technical problems), which means the pressure to simply can the project is still pretty high for a future non-believer politician, but also just that politicians preside over white elephants all the time, so I really just don’t see it as having as much impact as expected: It’s just really bad planning and governance. Suffice to say, there are better ways to improve a project, or avoid building a bad one then wasting money and putting our future selves in a worse situation.
All of this comes back to our tendency to put off fixing issues of process, even when their existence is not delivering any measurable benefits. A case can be made that an expensive project is worth it if it’s being done expeditiously. But I don’t think keeping something expensive alive unreformed based on hopes that our historic pattern with transit projects in North America and particularly the US is worthwhile. Not fixing things today rarely means you’re actually getting closer to your objectives in the long run, but it does have lots of consequences.
Of course, tying into that is the idea that maybe the solution to our problems is to resource up — if a project is moving too slowly, maybe it needs more people working on it or more funding. I take great issue with this thinking even though it seems intuitive. The issue here is that even when extra resources are used to complete something, it’s usually far less efficient than reforming the bad practices that are causing problems in the first place: at worst it can actually encourage inefficiency even more because budget and time pressures are reduced while managerial and organizational workloads increase. There’s a well-known series of essays in Computer Science (I had to mention) that fall under the name “The mythical man month” that talk about how increasing resources can slow a project down, largely due to increasing the complexity of managing the workforce, and I think this applies more generally to problems and trying to solve them by increasing resources. Counterintuitively, in the long run and with things as complex as massive projects, taking it slow and constantly reevaluating approach is often much more efficient than trying to work as fast as possible and throwing more resources at problems that simply do not respond to more resources.
So, those are some of the many issues I see with regard to our way of looking at transit problems and projects. I think it is often less interesting and fun to discuss, but people problems and intangibles definitely play a big role in many of the issues we face.
As always, thanks for your continued support, and stay tuned for the next one!