Public Transit and the Postal Service Have the Same Problem
The USPS and U.S. transit agencies face the same impossible demand: Succeed as both a business and a public service.
If you want to understand the deepest structural problems facing U.S. transit agencies, look at the U.S. Postal Service.
The new Postmaster General is making sudden changes that are reducing the speed of mail delivery, which is certainly cause for alarm. But the Postal Service has always had a deeper problem. Our expectations of it are contradictory. To put it simply, we expect it to be available to everyone, as though it were a public service, and yet we also expect it to achieve high usage at low cost, as though it were a business. Even more urgently, we expect our leaders to protect us from having to think about how contradictory that is.
Transit agencies that provide rail and bus service over a vast metro area have exactly the same problem.
Let’s start with the most interesting thing that public transit and the Postal Service have in common: They are remarkably popular. A stunning 90% of Americans have a favorable view of the Postal Service. Meanwhile, when people in metropolitan areas are directly asked whether to fund transit agencies or new transit projects, they usually say yes. Even on the day that Donald Trump was elected president, voters passed most of the transit funding measures that were put before them in metro areas across the U.S.
Yet journalism often judges both services as being in decline because of falling usage. U.S. transit ridership drifted downward over much of the late 2010s, while postal volumes have been going down ever since the advent of email. These trends are indeed bad news to the extent that postal and transit services depend on revenue from customers. But that doesn’t mean that the justification for these services is getting any weaker. It may just mean that paying for them out of fares or postage revenue is inconsistent with the actual reasons that we value them.
A 2019 study by urban planning professor Michael Manville of UCLA asked the simple question: Why do people vote for transit but don’t use it? He looked at Measure M in Los Angeles County, which won more than 71% of the vote in November 2016. The measure expanded a sales tax expected to generate $860 million per year. Only 65% of that was for transit, but as Manville notes, “transit dominated both the coverage and rhetoric of the campaign.”
Manville found that support for the measure tended to be driven by “positive attitudes about transit” unrelated to whether the voter would use transit personally, or even to whether it would deliver congestion reduction benefits that were emphasized in the campaign. He concluded that “the political project of securing transit funding may be … at odds with … the policy project of encouraging transit ridership.”
If transit were a business whose success is to be judged by ridership, then this would indeed be a problem. But maybe actual voters don’t care about ridership that much, and maybe that’s fine. It could be that voters want to live in the kind of place that has effective public transit service, and that they are voting to push the city in that direction. They may also care, for example, about the ability of everyone to get around, including essential workers who are holding civilization together.
Ridership is not a measure of those goals. It goes up and down for all kinds of external reasons, such as fluctuations in the cost of driving or the availability of semi-competitors such as Uber. That doesn’t mean the value of transit to the city is going up or down. Measure M’s transit investments were mostly about long-term city-shaping strategies like major rapid transit lines. If these help foster the growth of dense and walkable neighborhoods, that will produce many benefits that ridership doesn’t measure.
In short, ridership is a helpful metric of some things, just as postal volume is, but if it isn’t connected to why people support the service, then it shouldn’t be our main measure of whether the service is succeeding.
What would it mean to measure postal performance, or transit performance, according the reasons we actually care about them? We would have to think about how we want to balance competing goals of availability with usage. Availability is the defining goal of a public service. Usage (and the revenue it generates) is the defining goal of a business. Of course, revenue isn’t the only reason to pursue usage. For example, transit only reduces car trips to the extent that it’s being ridden.
The Postal Service is unprofitable because it’s universally available. No for-profit service would ever commit to visiting every mailbox in the nation six days a week, as the USPS is bound by law to do. Likewise, no transit agency that was trying to maximize its ridership would promise regular transit service close to every home across a huge urban region.
That’s because both services are far, far more efficient in some places than in others, for purely geographical reasons that have mostly to do with density.
Suppose you live out in the country, in the only house at the end of a half-mile-long dead-end public road. The Postal Service may have to drive to your mailbox and back, a mile every day, just for you. If you happen to be in the service area of a transit agency that has promised service to absolutely everyone, then the transit agency will also have to drive that mile, just for you. (If it’s demand-responsive service, they’ll do that only when you ask, but if it’s a fixed route service they’ll go there whether you ask or not, just as the Postal Service does.)
In the time a mail carrier spends driving to your rural cul-de-sac, she could walk up to an urban apartment building and fill at least 20 mailboxes. Likewise, a transit vehicle driving that mile to get to you in the country could be driving a mile down a busy main street, where it might easily pick up 20 passengers in that time.
Because of this difference, postal and transit authorities both get accused of failure on usage-based measures — low usage or high operating cost per customer — because they are doing things that they would never do if usage were their true goal.
When these authorities commit to availability at the expense of usage, they aren’t just benefiting their most rural or outer-suburban customers. Transit services for disabled persons, which are a civil right under U.S. law, also express a commitment to availability even though a comparatively small number of people use them. As for low-income and minority populations, those who live in dense cities benefit from service designed for usage. But more of these groups are now living or working in suburban areas, where the geography is hostile to high-usage services. As a result, social and racial equity will sometimes require services justified only by availability.
What do we want, then, from our postal and transit agencies? How do we pay for it? And how do we measure whether we’re getting it?
In my work as a transit planning consultant, I ask decision-makers this question: “I know that you want both high ridership and service that goes everywhere, but your budget is fixed, so how much do you want to spend on ridership as opposed to universal availability?” Then, we can design a network that matches that balance of goals, and we can show clearly which services are designed for ridership, and which are for availability regardless of ridership. (Some services are a mix of the two, but we can quantify that too.)
Then, to decide whether services are succeeding, you measure them against their actual goal. Where an agency is explicitly trying to achieve high ridership, look at their ridership trends. But if availability is the goal of the service, you’d ask: What percentage of the population is near service? What are we spending to achieve that? In short, are we producing availability cost-effectively?
We should also stop using “efficiency” to mean “efficiency at delivering usage,” as we do when we say that the Postal Service is “inefficient” because it costs so much to deliver a declining amount of mail. Talking that way only gives efficiency a bad name. Efficiency is how much of any good thing you can deliver for the money you have. If the goal is usage, measure how efficiently you’re delivering that for the budget. But if the goal is availability, measure how efficiently you’re doing that.
It’s also absurd to judge postal or transit services by how well they are competing with the private sector, which is always a losing battle. UPS and FedEx compete with the Postal Service in cities, but when a package needs to go to a rural ranch, they sometimes have the USPS deliver it. The Postal Service’s availability commitment — which is central to why it isn’t profitable — also improves the bottom line of UPS or FedEx. Public transportation, too, faces competition from a for-profit sector but can’t act in a for-profit way in response. Transit agencies bear an availability commitment that Uber and Lyft can ignore, in addition to massive compliance requirements tied to federal funding. We can’t judge the results as though a fair competition were occurring.
What does all this mean for funding? If availability is the goal, then funding by the customer is the wrong tool. When we tell postal and transit services to chase revenue — postage or fares — we’re telling them to maximize usage. The way to do that on a fixed budget is to reduce availability, by focusing service only on the places where the highest usage is possible.
Some of the funding sources that transit agencies rely on, especially at the state level, further reinforce usage goals over availability goals. For example, California’s statewide source, the Transportation Development Act, ties some funding to whether local agencies pay a certain share of their costs with fares. That means they must plan for high usage, which usually means less availability. Does the funding authority really mean to tell the transit agency that availability is unimportant, as these requirements effectively do? Is that consistent with why the voters are ultimately supporting this funding? Maybe it is, but this question shouldn’t be hidden in technical documents. It’s an important value judgment that should be debated in public.
Postal and transit services have the same problem. We want them to attract high usage and we want them to go everywhere, but those goals imply opposite kinds of service. Pursuing either goal will cause outcomes that look like failure when judged by the other goal’s measures of success. It’s like we’re telling our taxi driver to turn right and left at the same time. When they can’t do that, we just yell louder and call them incompetent. Is that taking us where we want to go?
Jarrett Walker is an international consultant in public transit network design and policy, based in Portland, Oregon. He is the author of the book Human Transit: How Clearer Thinking about Public Transit can Enrich our Communities and our Lives, and blogs at Human Transit.