I live in Portland, Oregon, with a metropolitan area of about 2.5 million people. We have a pretty decent mass transit system here, called TriMet. I don’t have a lot of experience with other cities’ transit systems, but I do know a few things about this one.

A few years ago I wondered why TriMet made a particular policy decision, and when I looked into it I discovered that my intuition about government budgets was wrong. Today I’d like to walk you through the things that surprised me, using a case study on what transit riders pay to ride the bus here.

For 37 years (from 1975 to 2012), TriMet designated a small part of their service area as “Fareless Square”. This area primarily covered downtown Portland, which is the hub where most bus and light-rail routes meet. As long as you were riding only within that area, you didn’t have to pay anything to ride.

While I was a student at Portland State University in downtown Portland, Fareless Square was fantastic, both for students like myself and for guests the university hosted. For example, on the somewhat surreal occasion when I as an undergraduate student got volunteered to host a meeting of Larry Wall and the rest of the Perl 6 team, it was trivial for me to take everyone to a nice restaurant for dinner; we just hopped on the light-rail and didn’t have to deal with fares for a dozen out-of-towners.

There were several reasons given for ending Fareless Square, but officially the big one was that TriMet faced a budget shortfall and predicted they’d get $2.7 million more in revenue per year if they quit giving people free rides. (For more detailed history and other concerns about Fareless Square, see KGW’s article: What was TriMet’s ‘Fareless Square’ and should it come back?) My memory may be faulty, but I think I recall the total shortfall was expected to be around $15 million.

Compared to my household budget, those are big numbers! It made sense to me at the time that a budget shortfall of that magnitude would be a big deal. I was really unhappy about losing Fareless Square, but I accepted their claim that this was necessary to solve their budget woes. But was it really?

Because TriMet is a public agency, their annual budget report is also public information, and a few years ago in a fit of idle curiosity I actually read it. I’ve just updated my findings against TriMet’s FY 2020 budget, which I encourage you to look at if you have difficulty sleeping.

About 95% of TriMet’s revenue comes from three sources:

  • 60.1% from payroll taxes,
  • 19.6% from federal and state funds,
  • and only 16.1% from passenger fares.

Passenger fares are expected to come to about $112 million dollars this year, a figure which has not changed significantly since the 2013 fiscal year. (The budget’s exhibit 2 provides these numbers for every year from 1973 onward, if you’re curious.) And passenger fares are the smallest portion of their revenue. So it’s safe to say $2.7 million was a drop in the bucket.

Given these facts, I feel cheated. Fareless Square was awesome and they took it away for such a paltry amount of money? I’m left believing that finances were just an excuse they knew people wouldn’t argue with, not the real motivation. (Again, see the KGW article for other possible concerns that may have motivated TriMet management.)

When I realized that, I was angry. I wondered, what if TriMet had gone the other direction? What would happen if we made all of TriMet fareless? It wouldn’t be the first transit system to provide free public transport. There are well-known advantages for rider satisfaction and system efficiency if everyone can just get on and off without the hassle of dealing with fares. And of course as a matter of public policy it’s good for both traffic congestion and air quality if more people can be encouraged to use mass transit.

There are quite a few important questions that would need to be answered to make that policy proposal succeed, and I’m not interested in addressing most of those questions in this post. For example, a fareless system would have more riders and correspondingly higher costs, and any funding proposal would need to take that into account. But I want to continue focusing on things that surprised me when I threw out my assumptions and actually examined TriMet’s budget.

Given the earlier numbers, we know that today we’d have a $112 million budget shortfall that we’d have to make up somewhere. But wait—collecting those fares isn’t free. TriMet has budgeted over $10 million this year for personnel, materials, and services in the “Fare Revenue & Administrative Services Department”. There are some relevant costs elsewhere in the budget as well.

So let’s say the actual shortfall is about $100 million.

Here again my intuition was wrong. “Obviously,” I reasoned, “the system is funded by passenger fares and would collapse without them.” But funds from local, state, and federal taxes are almost 80% of the budget, and the fare revenue is only 16%.

For my whole life, a major controversy in Oregon government has been around the use of property taxes to fund significant parts of state and local government programs, so my first guess was that must be the taxes funding TriMet. (If you’re curious about the history of this controversy, you can read up on Oregon ballot measures 5, 47, and 50.) So based on that assumption, I did the math and worked out that if Portland Metro area voters would approve a property tax of 34 cents per $1000 of assessed value, that would make up for the lost fare revenue. But it turns out that hardly any TriMet funding comes from property taxes. As far as I can tell, it’s at most $3.75 million of the budget, and probably less than that. Given the controversy and the fact that property taxes are not a significant part of TriMet’s budget today, I didn’t consider this possibility further.

Where else could that money come from?

The largest portion of TriMet’s tax revenue comes from payroll and self-employment taxes levied on employers across the region, amounting to $382 million over the last year. What if businesses were interested in getting on board with a fareless plan?

The 2019 TriMet payroll tax rate is approximately 0.75%. For comparison, federal payroll taxes for Social Security and Medicare add up to 7.65% this year in employer contributions, which is ten times as much. A few years ago when I asked my employer’s accountant how much they actually paid in TriMet taxes, they considered it practically a rounding error in their overall budget. Now that I’m self-employed I pay TriMet taxes myself and they’re tiny for me too, especially considering the 15.3% tax I have to pay for Social Security and Medicare.

My first instinct when I discovered that TriMet payroll tax revenue was $382 million was to figure out how much that has to grow to cover the $100 million of fare revenue, which works out to a 30% increase. But asking businesses to pay 30% more taxes sounded like something they’d fight vigorously, so I knew I needed to look at this more carefully.

How much money are we really talking about? Increasing the TriMet payroll tax by 30% would bring it from around 0.75% to roughly 1%, which is still not much. We can break this down by how much more an employer would pay each year per employee based on that employee’s wages:

  • $60 for full-time minimum wage employees: The Portland Metro area minimum wage is currently $12.50/hour. Increasing the TriMet tax to 1% would cost employers about $60 more per full-time minimum-wage employee per year. In exchange, these employers could make it easier for both their employees and their customers to get to their place of business.

  • $150 for employees earning the average salary for the region: According to Exhibit 7 in TriMet’s budget, “Local Economic Trends”, the average pay per employee in this region is $62,717 per year. For each employee earning that much, increasing the TriMet tax to 1% would cost their employer about an extra $150 per year.

  • Less than the cost of monthly bus passes for employees earning up to $500,000 annually: Many employers in the region buy TriMet tickets and monthly passes for their employees, which cost up to $1,200 per year per employee. If TriMet went fareless, then employers would no longer need to pay that cost, but they’d pay more in the TriMet tax. Still, they’d actually save money on every employee who makes less than $500,000 per year!

Once I broke the costs down this way, I learned that on the scale of a metropolitan region of the size of the Portland area, even numbers like $100 million just aren’t that big. That was the biggest surprise for me.

Many people won’t even consider a proposal like fareless transit on the grounds that it sounds too expensive, but as I discovered, our intuitions about finance mislead us when we think about budgets on the scale of governments.

I don’t mean to suggest that fareless transit should be a no-brainer, because these sorts of proposals face many more objections than just “it’s too expensive.” What I would like you to take away from this story are these points:

  • If you would like to persuade someone that a government policy proposal is a good idea and they object that it would be too expensive, you may well be able to disabuse them of that notion. But bear in mind that they are likely to have more concerns hidden behind that one, and you should work to identify what those concerns are so you can have a real dialog.

  • If you would like to persuade someone that a government policy proposal is a bad idea, you should be very cautious about using the argument that “it’s too expensive,” and make a sincere effort to estimate what the proposal would cost. Otherwise you may find your argument quickly torn to shreds. I encourage you to think carefully about which of your values the proposal offends and make your argument based on those instead.