By Clay Shentrup
I had coffee with Portland City Councilor Mitch Green this morning. We talked about the Transportation Utility Fee (TUF) — the proposed $12 per month charge on every single-family household and $8.40 on every apartment to keep PBOT from going under. Like most people following this, we agree PBOT needs revenue. The question is whether the current proposal is the best we can do.
It isn’t. And the lot-size alternative floated by Strong Towns PDX, while well-intentioned, isn’t either.
I’m a software engineer and co-founder of the Center for Election Science. I live in a craftsman house in District 3 with my wife. We work remotely. We barely drive. I bike as much as I can. Under the lot-size proposal, I’d pay more than an apartment dweller who commutes by car every day—because my lot is bigger, even though my impact on the road network is smaller. That’s an ecological fallacy: charging individuals based on a statistical average of their housing type rather than their actual behavior.
The flat TUF is a head tax, and head taxes are bad. We’ve seen this movie — the Arts Tax charges every adult $35 per year regardless of income, has dismal collection rates, and requires a means-testing apparatus for low-income exemptions that costs almost as much to administer as it collects. The TUF as proposed would land in the same territory.
But scaling the fee by lot size doesn’t fix the problem. It creates new ones. Lot area correlates heavily with land value, which makes it legally vulnerable under Measures 5 and 50 — a plaintiff could argue it’s a property tax with extra steps. And lot size doesn’t cause road damage. Traffic volume, vehicle weight, and weather cause road damage. A 10,000 square foot lot doesn’t deteriorate the road in front of it any faster than a 5,000 square foot lot. You’re pricing the wrong thing.
So what should we actually do? Decompose the problem into two instruments, each doing one job well.
The Gift Card Fallacy and Free Parking
Imagine the city gives you a $100 gift card to a store you don’t like. You’d rather have $85 in cash — you’d get more value from it. The city spent $100; you got $85 worth of benefit. That $15 gap is what economists call deadweight loss. It’s invisible waste that happens whenever you give someone a benefit they value less than it costs to provide.
Free on-street parking is one of the most brutal examples of this. The city provides a scarce, valuable public resource — street space — for free. The cost of that subsidy is real: it’s the revenue the city doesn’t collect, the congestion it doesn’t price, the transit riders and cyclists who subsidize drivers through their taxes. But nobody sees it, because we think “yay, free parking!” without considering that we’re paying for it through higher taxes, worse roads, and fewer services elsewhere.
Donald Shoup spent his career documenting this. The late UCLA economist — widely respected across the urbanist spectrum — showed that free parking is essentially a massive in-kind transfer to car owners, and like all in-kind transfers, it’s worth less to recipients than it costs everyone else. Demand-responsive pricing fixes this by converting an invisible subsidy into visible revenue.
Expanded parking pricing is the single most impactful thing the city can do right now. It’s essentially congestion pricing by another name — you’re charging for the direct use of transportation infrastructure, in real time, with prices that respond to demand. The nexus is legally airtight. And crucially, it captures non-resident vehicles — which Joe Cortright has shown cause the majority of road damage in Portland. The TUF can never touch those drivers. Parking pricing can. Portland already does demand-responsive metering in several districts. The proposal is simply to extend the principle citywide.
Flat Fee + Universal Credit = Progressive Effective Rate
For the TUF itself: keep it flat, but pair it with a universal per-person annual credit. No means testing. Everyone gets it.
Here’s the part that surprises people. A flat fee plus a fixed-dollar credit produces a progressive effective rate automatically. Consider a flat 40% income tax with a $10,000 refundable credit. Someone earning $12,000 gets a net payment — a negative 43% effective rate. Someone earning $40,000 pays 15%. A millionaire pays 39%. The marginal rate is flat. The effective rate is deeply progressive. No brackets, no phase-outs, no means testing. (I’ve written about this in more detail.)
Same principle here: the TUF is the flat rate, the credit is the offset. Low-income households come out ahead. High-income households pay net positive. The credit can be delivered through the same infrastructure the city uses for the Arts Tax — same database, opposite direction. It more than offsets that $35 per year, effectively neutralizing the Arts Tax without touching it.
This isn’t some untested theory. Research from the UBI Center, drawing on OECD (Organisation for Economic Co-operation and Development) data, shows that the US already has the most progressive tax brackets in the developed world — more progressive than any Scandinavian country. And yet we achieve roughly half the inequality reduction. What those countries do differently isn’t tax structure. It’s transfer generosity. Countries that distribute a larger share of income as universal cash transfers consistently reduce inequality more, regardless of how progressive or flat their tax rates are. Generous universal transfers beat steep brackets every time.
Two Tools, Two Jobs
The framework does three things at once. Parking pricing provides the efficiency signal — pricing the actual externality of driving rather than a proxy like lot size or housing type. The flat TUF provides stable baseline revenue for PBOT. And the universal credit makes the whole package progressive without a single line of means-testing code.
The City Council Finance and Governance Committee of the Whole meeting on April 15th is the next decision point. Council has a choice: pass a head tax that will generate the same resentment as the Arts Tax, or build something that actually works — economically efficient, legally durable, and genuinely progressive.
The full analysis, including the math, the OECD evidence, and responses to common objections, is at wonk.blog/tuf.
Clay Shentrup lives in Portland’s Upper Laurelhurst neighborhood in District 3. He is co-founder of the Center for Election Science and has nearly 20 years of experience in electoral and policy reform, including successful approval voting campaigns in Fargo, ND and St. Louis, MO.
Thanks for reading.
BikePortland has served this community with independent community journalism since 2005. We rely on subscriptions from readers like you to survive. Your financial support is vital in keeping this valuable resource alive and well.
Please subscribe today to strengthen and expand our work.
A benefit of expanding parking fees is that it can also be scaled to personal trucks and SUVs similar to cities in France tripling the parking fees for SUVs.
By all means price parking! There’s a direct user fee. Raise the gas tax while we’re at it.
In your example you used a flat 40% tax plus universal credit which makes sense. Are you suggesting a flat dollar TUF, or a flat % tax TUF based on income? Maybe I missed that part. Please do not make me fill out yet another tax return.
I read through his link and it appears the plan is the city sets a target revenue, then the city collects the parking fees / permit fees, etc, and whatever money exceeds the revenue target, gets returned in an even distribution to each Portland resident, regardless of income, etc.
No new tax return, I promise! The 40% income tax example in the piece was just an analogy to illustrate the general principle that a flat rate + universal credit = progressive effective rates.
In practice, the TUF itself stays exactly as proposed—a flat dollar amount per household, collected on your utility bill. No income calculation, no filing. The only addition is a universal credit paid to every Portland adult, delivered through the same database the city already maintains for the Arts Tax. You don’t apply for it. You just receive it.
The parking pricing piece is completely separate—you pay when you park, just like you do now in the metered districts. The revenue from that is what funds PBOT and the credit.
So the net experience for a typical Portlander is: your utility bill has a line item (same as now), you get a small annual check or direct deposit (new), and if you park on the street you pay for it (expanded from current). Simpler than the current proposal, not more complex—because there’s no means-testing exemption apparatus to administer.
So if every head pays a fixed fee and also receives a fixed credit, how does that add up to progressive tax rate? Or do only the folks who are exempt from the arts tax receive the credit? That’s a means test, but it’s already in place. Is that your point?
Good question—I should have been clearer. The 40% income tax example in the piece was an analogy for how flat rates + universal credits produce progressive effective rates. In this case, the “flat rate” isn’t the TUF itself—it’s the parking price, which is effectively 100% of the cost of the parking you use.
The progressivity comes from the parking side, not the TUF side. Wealthier people tend to drive more, own more cars, and park in higher-demand (more expensive) areas. They pay far more in total parking costs than lower-income residents. But the credit is the same fixed dollar amount for everyone—it’s universal, not means-tested.
So: a car-free renter pays $0 in parking and receives the full credit. Net positive. A moderate driver pays some parking, receives the same credit, roughly breaks even. A heavy driver pays a lot in parking, receives the same credit, and it barely dents their total. That’s a progressive effective rate without anyone filing anything or proving their income.
The whole point is that you don’t need to means-test the credit. Everyone gets it. The parking pricing does the sorting automatically, because consumption of parking is correlated with income and driving intensity. It’s not a perfect correlation like in the income tax analogy, but in aggregate, flat universal transfers are radically more effective at redistribution than means-tested benefits. The OECD data on this is overwhelming.
My understanding is that part of the rationale of the TUF is that all residents benefit from the transportation network, regardless of how we use it, rather than developing a fee that directly correlates to individuals direct wear and tear on the system. (Please correct me if I’m wrong!) If that’s so, I am tired of the rhetoric around how it’s unfair to charge people living on larger lots who don’t personally drive a lot or perhaps don’t even drive a car themselves. The point is, someone is doing your driving and moving your goods. If you don’t drive to the store, but have Amazon deliver: you’ve gotten benefit from the transportation network. If you don’t drive to the restaurant but get door dash, you’ve benefitted. If the employees at the coffee shop that you can walk to arrived via the transportation network, you benefitted. It doesn’t matter who does the driving but whether the transportation network makes your goods and services available to you. And in that way, larger lots are roughly correlated to a larger, costlier system because they space everything out further.
It’s like how I asked a friend who is a Tibetan Buddhist how he could eat meat without generating bad karma. He explained that there was an ethnic minority in their village who did all the butchering so that the animals were already dead by the time he bought the meat, therefore he did no killing and thus no bad karma, even though the animals were killed precisely because he and others were going to consume it.
“larger lots are roughly correlated to a larger, costlier system because they space everything out further.”
The correlation here is rough and shouldn’t be discounted, but taxing based on correlation or association is very vulnerable to case by case comparison. Fees based on consumption, which people believe that they have some control over, are much more intuitive and fair.
Taxing or charging fees based on things that are associated with a costly system should make the effort to identify the strongest correlates of cost.
I actually agree with your premise—everyone benefits from the transportation network, even people who never drive. The coffee shop example is a good one. But I think you’re drawing the wrong policy conclusion from it.
The problem with a flat fee based on that logic is that it provides no price signal. If everyone pays the same regardless of how much road capacity they consume, there’s no incentive to consume less. That’s the tragedy of the commons: when a shared resource is unpriced, it gets overused. The whole reason Portland’s roads are deteriorating faster than PBOT can maintain them is that driving is massively underpriced relative to its actual cost.
Parking pricing solves your Amazon delivery concern directly. If the delivery driver pays to park, that cost gets passed through to the price of the delivery. The person ordering DoorDash pays more for DoorDash. The coffee shop whose employees drive in pays slightly higher labor costs, which get reflected in the price of coffee. The price mechanism transmits the cost of road use to the people who benefit from it—including indirect beneficiaries like you’re describing—without anyone having to fill out a form or prove anything about their behavior. It’s automatic.
The Buddhist analogy is actually a great illustration of why the flat fee *doesn’t* work. The whole point of the story is that the karma is real regardless of who does the killing—the consumer is responsible. Parking pricing makes the consumer pay through the price of the goods and services that required driving. A flat TUF charges everyone the same whether they order ten Amazon packages a day or none.
I don’t know that it’s *my* premise that everyone should pay the TUF because everyone benefits from the transportation network, more that I thought it was the premise being touted for this particular TUF. The gas tax already exists and (imperfectly) scales with consumption, so I don’t think the TUF necessarily needs to do everything here.
Also LOL that Amazon drivers pay for parking–they just stop in the middle of the street, duh!
The obvious downside of this proposal is that a very loud subset of drivers think it is the apocalypse if they have to actually pay for the infrastructure that they use or are mildly inconvenienced in any way and so that component of it (while clearly, obviously necessary) will be quite politically toxic
Yes and a major point to understand in this debate is that, for the city leaders who support the TUF in general, time is of the essence. There’s a real fear that anything that causes any sort of political liability or public controversy could derail the entire thing and given how severe the funding crisis is, there’s very little appetite for anything complicated or potentially controversial.
I hear the political concern, but I think it’s actually backwards. Why would the median driver prefer a flat head tax over parking pricing?
Run the numbers roughly. Portland has ~500,000 adults in ~270,000 households. If expanded parking pricing generates around $60M/yr (conservative middle of the estimated range), a substantial chunk—call it 30-35%—comes from non-residents: commuters from Vancouver, Lake Oswego, Beaverton. So maybe $40M from residents, $20M from non-residents. PBOT keeps $47M for roads. The remaining $13M funds a universal “parking credit” (but like the Idaho “grocery credit”, is really just a UBI that can be spent on anything) of roughly $26/adult/year—about $45-50 per household.
Under the current TUF, a typical household pays ~$144/yr regardless of behavior. Under parking pricing, a household breaks even when their annual parking costs exceed roughly $190 (the TUF amount plus the credit they’d receive). But driving and parking consumption is heavily right-skewed—a small share of heavy drivers account for a disproportionate share of total parking use. When you account for that skew plus the ~10% of Portland households that are car-free (who pay nothing and pocket the full credit), the break-even falls around the 60th-65th percentile of driving intensity.
That means roughly 60-65% of Portland households pay less under this system than under the flat TUF. A clear majority comes out ahead—and the ones who don’t are the heaviest road users.
The people who lose are the loudest. But they’re a minority, and they’re the ones imposing the highest costs on everyone else. The political question isn’t “will some people complain.” It’s whether you’re willing to design policy around the preferences of the people who cause the most damage.
New York City’s congestion pricing faced nine lawsuits, a presidential veto attempt, and years of political warfare. One year in: $550M in revenue, traffic down 11%, pollution down 22%, economy booming. Every congestion pricing program in history has been more popular after implementation than before—London, Stockholm, Singapore, New York. The political risk of doing the right thing is temporary. The political risk of passing another Arts Tax is permanent.
So much better than last week’s guest opinion. When they claimed “Considering apartment dwellers are also more likely to ride transit, walk, and ride a bike to meet their daily needs, leading to even less road maintenance costs…” I had to wonder if they have ever visited a neighborhood before and after one of those giant apartment buildings with no parking was built there. The huge increase of cars parking on street is impossible to deny.
If we’re charging road users by damage caused, shouldn’t Trimet be paying the most? Just look at the pavement in front of any bus stop, huge ruts get carved into the street, usually right where we are supposed to be biking.
Thanks—and you’re actually making my argument for me on the first point. If apartment residents without off-street parking are flooding the surrounding streets with cars, that’s exactly the problem on-street parking pricing solves. Those residents would pay for the curb space they’re using. The lot-size proposal would charge them *less* because their building sits on a small parcel—even as they’re consuming more on-street space than a homeowner with a driveway who never parks on the street. Parking pricing charges the behavior. Lot-size charges the housing type. Your neighborhood observation is evidence *for* pricing, not against it.
On TriMet: yes, and they should pay too. Buses are heavy and do real pavement damage. That cost should be reflected in fares, and it largely is—TriMet pays fuel taxes, and the cost of operating heavy vehicles on public roads gets passed to riders through the fare structure. That’s the price mechanism working correctly. Buses are still far more efficient per-passenger than private cars, but “more efficient” doesn’t mean “free of externalities.” The goal isn’t to exempt any mode from pricing—it’s to price every mode accurately so the actual costs are visible and people can make informed choices—which can include biking from home, living closer to work, biking, etc.
I appreciate the creativity and trying to target the taxes more. For discussion sake I see two issues with your argument:
The most direct would be to do a VMT fee but that seems impossible given that the state can’t even get that going.
Thanks for engaging substantively—these are both fair points.
On #1: the Arts Tax infrastructure is bad at *collecting* because compliance with a head tax is inherently low—people avoid paying. But the universal credit runs in the opposite direction. You’re sending people money, not chasing them for it. People don’t dodge receiving a check. The compliance problem is asymmetric: hard to collect, easy to distribute. The parking revenue itself comes from meters and permits, which is a completely separate collection mechanism with near-perfect compliance (you either pay the meter or get a ticket). The Arts Tax database just identifies who gets the credit. That’s a lookup and a payment—orders of magnitude simpler than the enforcement apparatus the Arts Tax currently uses to chase down non-payers.
On #2: this is actually the strongest version of the lot-size argument, and I think it’s partially right. Street frontage is a real cost driver—weather damage, utility cuts, and basic surface maintenance do scale with linear feet of road. But two things. First, the dominant cost driver for road repair is traffic volume and vehicle weight, not weather exposure. The AASHTO fourth-power law means a single loaded truck does roughly 10,000 times the pavement damage of a passenger car. Weather and utility work are real costs but they’re secondary to use-driven deterioration. Second, to the extent that frontage does drive costs, that’s already capitalized into land values and thus (imperfectly, thanks to Measure 5/50) into property taxes. A lot-size TUF is trying to recapture a cost that the property tax system already addresses, however badly. The right fix for that is property tax reform, not layering a second instrument on top.
You’re right that VMT is the first-best instrument. The city can’t do it. But parking pricing is the closest available approximation—and it captures non-residents, which neither VMT nor the TUF can do at the city level.
An interesting idea. I think the criticisms of the Strong Towns proposal miss the causality that does exist. Lower density directly causes higher costs of creating and maintaining roads. Demolish one apartment building tomorrow and build houses for all of those displaced residents. You will need to build huge amounts of new roads that need to be maintained. Demolish one house tomorrow and build a new house for the displaced resident. No new roads need to be built or maintained.
Is density the single factor that explains everything? No, but it does move in the correct direction.
The causality you’re describing is real but it’s the wrong unit of analysis for a fee. Yes, if you magically replaced an apartment building with single-family houses, you’d need more road. But you’re not charging the housing pattern—you’re charging an individual household. And the individual household on a large lot may drive less than the individual apartment dweller. That’s the ecological fallacy: inferring individual behavior from group-level statistics. It “moves in the correct direction” on average, but it’s wrong for a huge number of individual cases, and those people are paying a fee based on a false inference about their behavior.
Parking pricing doesn’t have this problem. It charges the actual act of storing a vehicle on public land. There’s no inference. There’s no average. You park, you pay. You don’t park, you don’t pay. The causal link isn’t “roughly correlated”—it’s one-to-one.
There’s also the legal issue nobody in the lot-size camp wants to talk about. A fee that scales with parcel area is substantially correlated with assessed land value. Under Measures 5 and 50, that starts to look a lot like a property tax—and Oregon courts have struck down fees that function as disguised property taxes. The flat per-unit TUF has survived legal challenge in 31 Oregon cities. A lot-area fee has not been tested. That’s a real risk for a city that needs this revenue to be durable.
The TUF is a monthly or quarterly utility fee administered through the same system used for water and sewer. The arts tax is an annual tax paid with your taxes. They’re not the same and comparing them conflates two very different things.
By all means hike parking rates, but the city needs a TUF now, and the perfect should not be the enemy of the good. We will be shooting ourselves in the foot if council ends up punting on this.
The comparison between the TUF and the Arts Tax has nothing to do with the collection mechanism. Nobody cares whether it arrives on your water bill or your tax return. The point is that both are flat per-person charges with no behavioral component. That’s the economic structure that matters, and it’s identical. Saying “they’re not the same because one is quarterly and one is annual” is like saying a $12 parking ticket and a $12 meter fee are fundamentally different policies because one comes in the mail. The price signal and incidence are the same.
On “the perfect should not be the enemy of the good”: I agree, which is why the piece explicitly says a transitional TUF with a hard sunset is acceptable if council needs revenue while parking pricing ramps up. But let’s be honest about what’s happening. “Don’t let perfect be the enemy of good” is doing a lot of rhetorical work here. It’s being used to foreclose any discussion of alternatives five days before a vote. That’s not pragmatism—it’s incuriosity. PBOT’s funding crisis didn’t materialize overnight. There is time to get this right, and “right” in this case means a system that actually prices the externality, captures non-resident drivers, and doesn’t repeat the structural mistakes of the Arts Tax.
This is a great article, thanks Clay!
I like the premise, though I am somewhat skeptical of market mechanisms to introduce fairness into parking prices. At least in some cases, residents and businesses view demand based pricing as unfair, and I think they are right to (in a very narrow sense). Parking definitely should be priced, the general question should be “what should the cost be”? To this end, I prefer something like linking high-demand parking to the effective cost of two adults riding transit. That way, it’s defensible to political challenges – is someone going to say that a transit rider should pay more? Anyways, that’s a bit of an aside and mostly relating to an issue I’ve been tracking in Seattle (you can read more on my own blog if you’d like)
I wasn’t previously familiar with universal credits, but I do really like that mechanism after reading your piece Clay. And especially for residential permits, where pricing issues (relative to the cost of something like transit) are extremely acute. At least in Seattle, the maximum cost of an RPZ permit is $95 per two years. Meanwhile, a monthly transit pass (that doesn’t cover higher-fare modes like the commuter rail or monorail or ferries) is $108. But people still complain about the cost! Anything that gets us away from this is a good thing. If people had to pay the same cost to park their car as they did to ride transit… surely more people would be riding transit.
And I’ll definitely be referencing your flat income tax with universal credit next time the income tax in WA comes up… incoming lawsuit over the uniformity clause, but I wonder if a flat tax + universal credit would get around that.
Like many here, I own a car but drive extremely infrequently. I currently am lucky enough to have a garage to store it in; but for the past decade I’ve parked it on the street. So if I am being charged to park on the curb, I have to feed the meter 24/7? I’d be driving to work – where there is a parking lot – every single day instead of riding my bike or taking the bus, because it would at least save me 40-50 hours a week of paying for parking.
If you think people hate the arts tax, just wait until you tell them that every single inch of curb space in their neighborhood is now “dynamically priced”. Dynamic pricing is one of the most demonic inventions of the 21st century. Seeing prices go up and down on those little e-ink price tags makes me feel like Ted Kaczynski. The TUF as currently proposed would require very little infrastructure because it would just be collected through the water billing system, which is already set up for every property, right?
Why would we give PBOT more incentive to maintain the car centric status quo? Tying their budget to the amount of curb parking they can charge for means that space never gets used for anything other than car parking. No new bike lanes, no new traffic calming, no outdoor seating, no new mid-block crossings. Can’t get rid of that precious revenue-generating curb space.
Why not also charge by length of vehicle too? My Honda Fit takes up half the curb space of an F150.
Both this and the Strong Towns proposal are great examples of “perfect as the enemy of good”.