A lot of effort for a 5% solution to transportation.

Matthew Yglesias often comes off as your quintessential public intellectual with firm convictions on otherworldly issues.  But he is more than that.  Unlike most public intellectuals, he puts in the hard work of researching the issue.  He still arrives at bad answers too often, but there is at least value reading him for the research.

Then there are times when he has a reasonably good answer for the wrong or incomplete reasons.  

It is worth reading American transit agencies should prioritize ridership over other goals by Matthew Yglesias.  The subheading is It's the only way to get costs under control.

This is reasonably representative of his work.  The article is full of good information.  There are elements of good reasoning and logic.  But there is also a miasma of abstract intellectualism and academia.  A mismatch between reality and theory.  

Why does the United States struggle to create cost-effective rail infrastructure? Why do non-NYC American cities have such a hard time attracting mass transit ridership?

On one level, these are deep and complicated questions. But one thing I want to say to the growing community of people who are interested in them is that American agencies don’t deliver on these goals because this country’s high-level governance constructs don’t say they should deliver on them. Of course you won’t find a grant guideline that specifically says “don’t focus on delivering high ridership at an effective cost.” But wasting money is really, really easy when nobody is specifically telling you not to.

Well, yes, as far as that goes.  He keeps coming close to an insight but then backs away from the edge of realism and reverts to an authoritarian engineering mindset.

In his book “Human Transit,” Jarrett Walker introduces the idea of a ridership vs. coverage tradeoff in planning a city’s bus network.

A bus system offers a lot of flexibility, but it comes with high operating costs because each bus needs a driver. So one question that cities face is whether to prioritize frequent service on the most promising routes to maximize ridership or extensive geographic coverage (lots of routes) to maximize the number of people who live near a bus stop. This same tradeoff recurs with the question of how frequently the bus should stop. I, very conveniently, live a block and a half from a bus stop. If that stop were eliminated, I’d have to walk 3.5 blocks instead. I’d obviously prefer the shorter walk, but eliminating half the stops along the route would mean the bus could move faster, which would make it more appealing to riders. And because the bus could move faster, it could run more frequently without hiring more drivers, which would also make it more appealing to riders.

That's an engineering problem which actually resides within a more complex system:

The reason agencies don’t have a single mandate to promote ridership is, of course, that there are other things that people want to consider — things like environmental benefits, racial and socioeconomic equity, and economic development.

But I think that in the 21st-century United States of America, a basic ridership goal is a decent proxy for all that other stuff. 

Notice how he identifies ideological objectives for a transit system (environment, race, social justice, etc.) rather than rider concerns (convenience, cost, elapsed time, safety, reliability, cleanliness.)  And:

The same is true for environmental and economic development goals. You can create jury-rigged examples of a bus route that displaces bike riders and has no emissions benefit or something. But if you look at the Chicago/D.C./Boston/San Francisco/Philadelphia tier of American transit cities, it’s just clearly the case that any substantial increase in ridership would mean lower emissions. And you could try to claim that Amtrak’s plan for a slow, infrequent train between Cleveland and Columbus will promote some important economic development goal, but ask yourself: how that could possibly be the case if nobody rides the train?

"It’s just clearly the case," is not in fact the case.  I can think of plenty of scenarios where an increase in ridership does not lead to lower emissions.  If you frame it more narrowly along the lines of "If all increase in ridership represents a switch from high emission transportation to low emission transportation, then it is clearly the case that any substantial increase in ridership would mean lower emissions"  you are at least tautologically correct even though in the real world the now explicit assumption (high to low emission conversion) is not necessarily true.

His greatest insight and least controversial is at the end.

American agencies don’t prioritize ridership in their decision-making, and the grant programs that sustain them don’t tell them to prioritize ridership. Instead, everyone works with very complicated multi-criteria processes that sound nice but in practice are clearly failing. It’s time to try something else.

While he makes some good points, has some good data, and suggests some occasionally good ideas, he fails through a category error of thinking.  Mass transit is not an engineering problem and he is treating it as a static engineering problem.  In doing so, he reveals the authoritarian mind.  Nowhere does he really address the role of rider choice.  It is sort of sotto voce throughout the piece but not particularly present. 

He wants to increase ridership.  In the commercial world it is a classic piece of business - how do you get people to buy more of what we are selling.  

And that is a central issue for a high fixed cost capital service like mass transit.  The variable cost of each additional passenger is negligible but your fixed costs are huge.  If the system is to be socially viable, its public costs must be covered by what the public is willing to pay (directly and indirectly).  That is a harsh standard which most systems fail and that failure is hidden by grants and subsidies.  

Yglesias is right - ridership is an important proxy for the health of a mass transit system.  But it is not the only measure and even on that, it fails dreadfully.  Commuting trips in the US by mode:

Single person car trip - 76.4%

Carpool - 9.0%

Mass Transit - 5.0%

Walk to work - 2.7%

Bike to work - 0.5%

To be clear on three things.  These are only work commute numbers which represent only 20% of all travel.  These numbers don't address the population that have no commute because they work from home.  There is much debate as to what percentage work solely from home but it would seem 10-15% is broadly the lower bound.  Finally, the numbers vary hugely by location.

In New York City the mass transit figure is the highest, by far, in the the nation with up to 50% of commutes by public transportation (depending on territorial definitions).  There are several states where public transportation is 1% or less of work commutes.  There are only 9 states where more than 5% of commutes are by mass transit - Hawaii (6%), Illinois (9%), Maryland (8%), Massachusetts (10%), New Jersey (12%), New York (28%), Pennsylvania (6%), Washington (7%), and Washington, D.C. (34%).  

Which suggests that mass transit is an important issue to only five cities - Boston, Chicago, New York City, Philadelphia, and Seattle.  Of those, we really are only talking about Washington, D.C. and New York City with a material dependence on mass transit.  And again, these are only the 20% of travel which is work commute related.  

Why do only 5% of commuters use mass transit?  A worthwhile question.

Convenience, cost, elapsed time, safety, reliability, cleanliness would all be important considerations and in low density, high heterogeneity America, those considerations will be weighted enormously differently.  

Particularly given the 80% of trips which are non-commuting.  It is highly likely that the infrastructure necessary to meet the needs of the public for the 80% of their travel which is non-work commute is an important factor when considering where to invest in infrastructure.  

But back to the bigger picture.  Why is Yglesias so disconnected from reality?  Particularly when he is really close to the real issues.  

His diagnosis is that mass transit systems should focus on ridership as a measure of success, if not the measure of success.  "American agencies don’t deliver on these goals because this country’s high-level governance constructs don’t say they should deliver on them."

But that's not quite true.  As he acknowledges.  The better formulation would be.  

American agencies don’t deliver on these goals because this country’s high-level governance constructs don’t hold them accountable for delivering on them.

The measure chosen is less of an issue than the fact that there is no accountability.

Similarly, Yglesias focuses on trade-offs as an issue.  It is not an issue.  It is the issue.

You have to have clear measures and then you have to be accountable for making the necessary trade-offs between those measures.  

In commerce there is an old engineering adage.

Faster, cheaper, better - Pick two.

Yglesias alludes to the desirability of simplicity.  There it is and it illustrates exactly why trade-off decision-making is so uncomfortable and why, absent accountability, it so rarely happens.  

For those having a utopian unconstrained vision of the world, of course they want faster, cheaper, and better.  Simultaneously.  They are all desirable.  And with really good companies, they can be pretty sophisticated about balancing the three.  

But they always have a clear shared vision of the relative priority of the three which will drive the trade-offs because in the real world there is never enough time, capital, or talent to deliver on all three simultaneously.  There will always be a number one priority, a number two, and a number three and it is often a log relationship.  Number three priority is significantly less than the first and second.

What is also missing from Yglesias's piece entirely is the wants of the citizens (and their consent).  What do they want from a mass transit system?  Focusing on ridership numbers (Yglesias's recommendation) is a catalyst to the real issue.  What do they want and why is it they are not choosing to ride?  

The distinction is important.  In some locations, city mass transit systems are now switching to no-fare systems just to get ridership.  In Atlanta when they spent $100 million to build a downtown transit system, all their ridership and financial projections failed.  They ended up providing the service for free.  And it is still not used by many.

What commuters want is the core issue for which ridership is merely a proxy.

Talk about hard trade-off decision-making.  Particularly in middle class and wealthy communities, where mass transit is only a viable alternative when it meets their definition of convenience, cost, elapsed time, safety, reliability, cleanliness.  And in terms of commuting, elapsed time and reliability is critical.  If it takes me half an hour by car and 45 minutes by train, then car it is.  The higher the income the individual, the higher the cost of time, the higher the opportunity cost.  The top three quintiles of population by income (60% of the population) have different priorities and have different options which make the prospects of them choosing mass transit as it currently performs, unlikely.

Yglesias recommends imposing ridership as the measure of success.

I think it is far more complicated than that.  The necessary actions include

Pick a measure of success (financial, ridership, coverage, etc.). 

 

Ensure that these measures of success are supported by the citizenry. 

Impose accountability for achievement of the agreed measures of success.

Force trade-off decision-making.

Accept the consequences of those trade-offs.

If you are going to raise mass transit ridership in the US, I suspect you are going to have to increase the frequency of service, increase the reliability of service, and increase the quality of service.  That will cost a lot.  You will have to police stations more diligently.  And it might still not be sufficient.  

Imposing ridership as the necessary measure of success is at best an incomplete and simplistic public policy choice.  And it betrays the mechanistic and authoritarian world view.  Riders have choices and meeting their needs is the solution.  A solution which we have not come close to achieving in fifty years except in New York City (high density and mass subsidies) and Washington, D.C. (mass subsidies).  

Thats a lot of effort and time and money for a 5% solution to transportation. 

UPDATE:  It occurs to me that Yglesias is likely at least sympathetic to the goals of Stakeholder Capitalism, DEI, and ESG.  If so, that represents yet another trade-off.  Stakeholder Capitalism, DEI, and ESG are the antithesis of accountability.  In fact, they can be viewed as a sophisticated mechanism for executive and middle management to escape accountability for the real and core responsibilities of using resources efficiently to satisfy public demand.