The Ineluctable Politics of Transport Funding

» If we insist on charging car users to fund transit, we have to accept increasing highway spending in exchange for more public transport subsidies.

Here is a matter worth noting: The vast majority of federal subsidies for transit are collected from automobilists who pay taxes on fuel. Revenue sources at other levels of government are more broad-based, many of them relying on sales taxes and the like, but in Washington what matters right now is the fuel tax. Similarly, any future major revenues for transit would likely come from increased tolling on roadways or a vehicle miles travelled fee.

The economic theory explaining why a government might charge car drivers to help pay for the commutes of people who use public transportation is sound: Drivers produce a number of negative externalities like pollution, congestion, and sprawl. And subsidizing transit makes an attempt to ensure adequate mobility for people of all classes and with all sorts of disabilities, which is an important goal.

Nonetheless, the American transport funding mechanism is routed in the user fee, a product of a certain logic that assumes that people should pay for what they use.

The reliance on the user fee is a reasonable explanation for why it is politically necessary to devote the majority of fuel tax revenue to roads resources rather than transit. From an economics perspective, transit may be construed as simply another part of the greater transport system, so there is not necessarily a contradiction between the user fee and the transit subsidy. But in a country where the vast majority of people drive to fulfill the majority of their transport needs, it would be politically untenable to suggest that most roads money be transferred to transit users.

That’s why I find the recent discussion by Cap’n Transit on this issue so perplexing. In critiquing my argument in favor of increased funding for infrastructure in general, he suggests that the problem is not necessarily the overall lack of funds but rather the manner in which those funds are distributed. “It doesn’t matter how much transit funding is cut,” he argues, “as long as road funding is cut more. He points to several examples in which a large roads project could be replaced by a transit one and serve the same number of people, while producing a lower level of negative social costs.

I certainly agree that the American transport system would be more sustainable and safer were the portion of capacity devoted to automobiles diminished with respect to transit capacity. You do not necessarily need more funding to make this possible, you just need to find the political will to cut funding for roads.

In essence, this is what has occurred in countries like France, where of the planned national spending on transportation over the next twenty years, just 4.5% will be distributed to highway construction. A major increase in transport expenditures has not been proposed there, just a redistribution of overall funds, just as Cap’n Transit has promoted.

Unfortunately, that argument is politically unsustainable in the United States, at least under the existing funding system. For politicians representing people who drive (most), it is virtually impossible to suggest that we collect transport revenues from fuel taxes and then devote most of those funds to transit rather than roadways.

France’s focus on non-automotive modes arguably has two origins: A lower mode share in the transport system for drivers (though it is still quite large), and a disconnect between fuel tax receipts and transport spending. The latter is problematic to economists, who like to see a direct connection between user fees and spending, but it allows the country’s politicians both to extract a very high tax on fuels and to spend a lot on public transport. France’s overall fuel tax revenues are almost as large as those of the U.S. federal government, despite the country having only one-fifth the population (though American states also have their own fuel taxes). Decisions about funding for transportation in France do not begin with an evaluation for how much the fuel tax per se brings in, since the connection is indirect: Money comes out of the general budget, into which the fuel revenues as well as those from other taxes had been deposited.

It is perhaps unsurprising that local transport agencies in the U.S., which collect funds locally from sources other than the fuel tax and collect federal fuel tax funds, but indirectly, are able to act basically in isolation from the political demands of motorists. Take Los Angeles Metro, which is supposed to take responsibility for highways and transit in its region. Of the $40 billion the agency is expected to bring in over the next thirty years from the 1/2¢ sales tax approved in 2008, just 35% will go to roads, with the rest being spent on transit capital and operations expenditures.

What are the implications for changes in how the American national government spends its transportation dollars? The continued use of the fuel tax as the primary — and direct — funding source suggests that from a purely political perspective the federal government has no choice but to spend massively on roads, certainly more than on transit. Fundamentally, that is why I have been a supporter of increasing infrastructure spending overall.

But a right-headed long-term approach would require that either we pull the national government out of the transport financing game altogether, or that we pull away from the direct connection between highway user fee collections and spending. Only after that change has been made can we really focus on altering the manner in which we distribute our limited transportation dollars.

Image above: St. Louis Metrolink station at 8th and Pine Streets, from Flickr user Jeramey Jannene (cc)

69 replies on “The Ineluctable Politics of Transport Funding”

Yonah, I agree with you.

If transit dollars come from fuel taxes, it’s just not going to work to be OK with cutting transit spending as long as spending for roads is cut more. There are far more voters who drive than voters who use mass transit, and if those who drive see today’s high fuel prices and don’t see benefits to their roads from fuel taxes, their opposition to fuel taxes will only grow; it’s already high.

I disagree that if an item is taxed, then the proceeds must go to that item.

Look at “vice” taxes, like those on cigarettes.

Not a cent goes to making smoking easier.
A portion goes to reducing smoking (which will lower the tax revenue)
And the rest goes to random things like education or the general budget.

Frame the gas tax, as a vice tax, and the solution is obvious.

50 cents a gallon.
Not a cent goes to making driving easier.
A portion goes to reducing driving (transit, which will reduce the tax revenue)
And the rest goes to random things.

Why is gas a vice tax? Because it pollutes, causes political tension, and is bleeding the country dry. Sort of like a casino.

What about road maintenance? That’s where true user fees come in, like tolls.

Gas isn’t a user fee, you dont need gas to use the road, and gas can be used in non-road activities, like jetskis, lawn mowers, etc.

and driving on roads that aren’t supported by the gas tax. I don’t get a refund of my gas tax paid when I burn the gas on roads supported by property taxes. Or on toll roads.

Exactly. A toll road would be a user fee. I can use a road without paying gas tax, on foot, bicycle, horse, electric car, etc.

Most toll roads ban pedestrians, equestrians, bicyclists etc. Cars, trucks and usually but not always, motorcycles. There are tolled parkways that ban trucks, …… Things with license plates and inspection stickers on them are allowed.

This kind of political analysis is essential, but I think you overstate the case. In the Washington area, polls show a 2-1 majority thinks transit is a higher priority than roads for transportation funding. Given the existing mode share split, this means that a majority, or close to it, of drivers thinks transit is the priority.

The problem, I think, has more to do with the elite dominance of American political discourse. The elite drives, and doesn’t care about the cost of gas. Think about the Brooklyn and Queens politicians who killed the congestion charge, even though the great majority of their constituents are transit riders.

Damned free riders.

The elite not only drive, but many are DRIVEN with car and driver provided by their employer. That is, we taxpayers pay when govt big shots are chauffeured around. And corporate big shots use these deductible “business expenses” as part of the plan for our corporate owners to avoid paying any taxes. The elite don’t even have to stop at a gas pump and watch the figures spin up to a fairly impressive total.

It’s a pickle tho. I concede that the Governor and the Secretary of Transportation, even the Congresscritters, members of the Lege, and other officials, must be able to do other work even while getting from here to there, plus the security concerns.

But if they somehow had a day of reality now and again, something like the yearly “Bring Your Children to Work Day” I guess, where they had to live for 24 hours like normal people do. Get up and get to work, get to your meetings, live your day without a car and driver, or even a car, you must use public transit just one day even if it means unreimbursed $25 taxi rides fbetween your nice place in the ‘burbs and your big job downtown. Oh, the horror!

I give props to Mike Bloomburg for taking the subway to work, even if his driver takes him in an SUV from his mansion the few blocks to the station, ha ha h a ha. It’s not just symbolically important, but he actually experiences the subway himself.

Climbing stairs where there are no escalators or elevators — and at his age I’m sure he notices. Crowding into subway cars where the washed and the unwashed get very intimate. Hearing the sometimes painful noise and the mumbled announcements of mysterious delays. Wondering why the vomit from last night’s excesses hadn’t been washed from the floor.

But also enjoying the speed and convenience of public transit, and appreciating its service to millions, of every walk of life and every range of income, to get to where they need to go.

It’s a short and partly phony ride for NYC’s Mayor. But wouldn’t it be great if every official with a govt car got to share the real world experience from time to time?

So you want more money for transit. I’ve got a solution:

1) Charge more in fares
2) Spend less on costs
3) Do them at the same time


Our costs are out of control on the whole for this entire country. Labor costs, procurement costs, construction costs, administrative costs, environmental costs, bureaucratic/political costs. They are ALL out of control, and they all deserve a hard reality check.

And at the same time, we refuse to charge people what their ride is really worth. We keep average fares low because we are too afraid to impose a burden on the poor, who somehow get subsidized anyway. If you want to ease the burden on the poor, then ease the burden on the poor by directly means testing the subsidies. Charge higher prices on monthly passes…trust me, if people are taking transit $60 bucks a month, it won’t immediately not make sense for them to do so on $100 bucks a month. The alternative is still 4-5x higher at the very least! There is no reason that average fares for a large city should ever be below $.50, and that is what they tend to be.

I know this may be hard to believe for some of the people out there who refuse to accept the real problem, but here is truth: Half the rail-based transport systems in our nation could be self-funding, and possibly even funding their own expansions, if their costs and revenue levels were benchmarked to the costs/revenues of similar systems in Spain.

As of right now, the US transit system has about as much financial credibility as MC Hammer. Their funding woes won’t go away until that is fixed. Who wants to spend more money on something that is unlikely to bring more benefits?

Which large cities have average transit fares of 50 cents? Most that I’m aware of have fares that start at $1.50 and above.
San Francisco $2.00
Los Angeles $1.50
Honolulu $2.50
Chicago $1.54
Washington DC $1.60
New York City $2.25
Miami $2.25
Portland, OR $2.05

Average fares are not the same thing as the posted fare. They are commonly calculated with two different methods, depending on the data that is available from the transit agency. The first is the total fares collected divided by the number of rides taken, and the second method is the average fare per passenger mile divided by the average miles per trip.

Subsidized fares for elderly/disabled/poor/students, fixed price transit passes for an unlimited number of rides, and free trips for the school systems in urban areas all eat into the average fare.

Danny, doesn’t this alarm about student passes and half-fares for the elderly etc overlook the TIME cost of the fare to the patrons?

When NYC became one gigantic one-fare zone, instead of requiring transfers from subways to buses to reach almost all neighborhoods in the “other” boros, it was a huge boon to working class people, and the young and the poor. And to me it seemed very fair. One city, one fare.

And after all, if the rider spends half an hour on a bus to get to the station, then rides half an hour on the train, isn’t that citizen paying with her time to get to her job from a home in a low-income part of the city? She is paying more than the upper-income folks in Manhattan and those who are gentrifying closer-in areas like Williamsburg, Harlem, Astoria, even parts of Bed-Stuy. If she has to pay higher fares, she has to pay in both her time and her money.


I do suspect there’s enormous waste in the yellow bus subsidies, where private contractors are eager to see the program expand and their profits alongside. It’s about at the point where we’ll provide buses for the fat kids so they don’t have to walk.

Outside the City, suburban riders of the commuter trains get bigger subsidies than City residents (probably not on a per-mile basis, but surely if figured on a percentage of my-house-to-my-job basis). One nice thing about congestion pricing (or bridge tolls in NYC) is that it not only would reduce the “free streets” subsidy whereby City taxes maintain streets used by those living outside the City, but it could also let govt reduce the subsidies to suburbanites by raising the commuter fares. If the politicians chose to do so, ha ha ha ha. They usually prefer to raise the costs to the workers and the poor and extend the subsidies to the rich and powerful.

Sorry, my comment about average fares being below $.50 was incorrect. I meant to say average fares per mile below $.50.

BTW, just to make it clear, the difference between an average fare per mile of $.50 and an average fare per mile of $.60 is huge for the profitability of the system. It would raise FRRs by 20%.

Looking at per-mile charges and using them to set users’ fares would be fair to riders, even if their fares do happen to go up.

On the one hand, higher fares give riders leverage over defining the quality of service. If a system is only recovering 20% of fares from the farebox, that means 80% is coming from subsidies. (Most riders, especially poor ones, falsely assume that if a transit ride feels expensive to them, they are getting gouged on the fare.) There are far fewer people controlling that other 80%, and elected officials or bureaucrats can cut off that larger share for trivial reasons.

Also, this subsidy system screws riders because it is based on an invoice of operating costs for services that have already been rendered. That 80% goes to paying for last year’s crappy service, not next year. You can’t recover the time, money or resources from last year’s crappy service.

Riders threaten a boycott? No big deal. The transit system will just run the buses and trains empty, collect their subsidies next year and reduce service or raise fares next year to make up the difference.

I was going to get to the other hand, and here it is. The benefits are only going to be visible in systems that have high ridership from the start.

Rail lines are simpler to levy per-mile charges because variable fare schedules are easy to enforce when there are limited access points. Any gated system can do this. Washington, D.C. and BART already charge by distance, and it is standard practice for commuter rail systems.

It’s even possible to do on proof-of-payment rail systems. San Diego used to charge by distance on the Trolley (it went to flat fares a few years ago primarily because of management prerogatives).

Yet rail systems are only in a few metropolitan areas, and in most cases, they have lower per-passenger-mile costs already.

Most of the nation’s transit ridership is on buses, and generally local buses with very close stop spaces and many access points.

You can’t meter individual rides by the mile like a taxi, fixed geographic zones can penalize riders taking shorter trips across a zone (see: BART charging more for an Embarcadero-West Oakland trip of one stop over Embarcadero-Balboa Park that spans San Francisco’s length), and less-patronized transit systems don’t need the complexity of administering a complex fare system for few riders. Also, fare enforcement will drive up costs by slowing the driver or adding fare inspectors.

An option for buses, or for bus-rail systems where fares need to be integrated, might be to calculate a flat-fare-by-time schedule. Enforcing a time pass is easier than a distance pass, and it may help bring down the marginal cost of ridership. More boardings bring down the relative cost of vehicles that are going to be in service anyway.

I don’t think the point is that fares should be proportional to distance. The question of how to charge is independent of the question of how to reduce operating costs. Operating costs can be kept low both on systems where the fare depends on distance, for example in Tokyo, and on systems where it is flat or nearly flat, for example in various Central and Northern European cities. (The Helsinki Metro is profitable.)

In fact on urban rail fares either are flat or rise less than proportionally with distance. Large gaps between the shortest and longest fares occur mostly on systems that provide service over the distances of commuter rail, including BART and Metro. Again, this is independent of the question of costs.

That makes things much clearer when you talk about fare per mile rather than just average fare. Then it gets into flat fare versus zone fare. Also I don’t believe you mad it clear that you were you taking into account all posted fares and averaging them out. You are questioning the need to help out those who are less well off, which leads into a thicket of social questions that can lead nowhere but arguments.

Sure, but just one comment: Spain has very low construction costs, but also very low revenues. Both the subway and the HSR lines it builds would often be considered marginal elsewhere. MetroSur, which famously cost $50 million per km to build, gets so little ridership the cost per rider is $20,000, not much less than Second Avenue Subway Phase 1.

To be honest, I’m worried about operating costs right now more than construction costs. It’s not that they’re worse – they’re not – but that the only way to bring them down is to massively lay off existing workers, and this is hard, both politically and operationally. The existing workers wouldn’t be used to it, European consultants would not be familiar with how to make this work, and Japanese consultants would not know anything faster than decades-long attrition.

The best that can be done is to use modernized FRA rules and reduced construction costs to quickly raise the amount of mainline service provided and build more systems in the Sunbelt, and then work out an agreement with the unions in the Rust Belt and California in which their workers get reassigned to Texas and Florida. But that requires far more discipline and cooperation among all involved, including the federal government, than the US political system is used to.

There could be a way to cut costs on some transit systems without hurting the serivces on it. On some Amtrak Routes that go from rural area to rural area or link a rural area to a urban area they could have say a Rail bus or a rail speeder which would be the size of a streetcar which would be built to carry say under a 100 people or they could get two or three of them linked up to one another and have them drive around the rail system. They would be less bulky then having one or two big locomtives and passanger cars going around on rual lines that don’t even fill the train 30%. That way when ridership goes up they could always change over the rail speeder buses for something larger if it is need.

I think they could try and work something out like this with rail speeder buses in say the UP of Michagan or other rural areas that have active rail lines that link up small towns center’s

You need reform to FRA rules to do that. Period.

The fact is, however, that there aren’t any Amtrak lines which “don’t even fill the train 30%”. You may be thinking off a decade or more ago. All the current ones are busy busy busy.

Railbuses would be excellent for new start lines, of course.

Danny wrote:

Our costs are out of control on the whole for this entire country. Labor costs, procurement costs, construction costs, administrative costs, environmental costs, bureaucratic/political costs. They are ALL out of control, and they all deserve a hard reality check.

No argument here. Costs can be controlled everywhere. Labor, though, is asking for a bloodbath.

At best, a victory over front-line workers’ unions will be Pyrrhic. At worst, in most cases everyone would be worse off than before.

Larger transit agencies have the worst share of labor as a percentage of total operating costs. Paradoxically, they have the lower costs per passenger-mile.

Larger transit systems are going to be productive no matter what. A driver, even with higher compensation outlays, will cost less than in cities with relatively low transit usage. Well-patronized transit systems are in high cost-of-living areas, so wages will necessarily be higher to staff equipment.

As for the other costs, they could be managed far more easily and yield results quicker than slaying the dragon of labor costs.

Procurement cost problems? Solutions: Abolish Buy America and for buses, convert from a bulk buy model to a lease model. This takes depreciation costs off the ledger.

Construction cost problems? Solutions: Very few. There may be things like PPPs or DBOM contracts, but there are still significant public risk costs involved. Also, construction costs seem to be relative to prevailing local costs. If construction costs are high for public agencies, they tend to be high for private jobs as well.

Administrative cost problems? Solutions: Consolidation. The office end of transit systems is rather standardized throughout the industry. These can be spun off to larger, central offices or even privatized. Unlike front-line work (driving and maintenance) that’s location-dependent, administrative jobs can be done regionally, at the state level, or even taken over by the FTA and run out of region-level offices.

Environmental cost problems? Solutions: Don’t know. Transit agencies have unique sets of environmental cost problems, and generally will pay more for them because they tend to gain expertise in handling them. Transit fleets have gotten more expensive because of investments in alt-fuels, yet the market share for vehicles powered by engines other than conventional gasoline or diesel is higher for publicly owned fleets.

Political cost problems? Solutions: Don’t know. Bureaucratic problems are handled in administrative. Political costs are inherent to a political (i.e., a deliberative body guided by a framework of laws and processes) system. Decision-makers are free of liability for the costs of bad risks. Economists like to call that moral hazard. That doesn’t mean “running government like a business” is the solution. Managers in this vein tend to bring out the worst aspects of the government and business sectors.

Leasing always ends up costing more than owning, if done on a large scale.

Leasing should be for the “marginal” buses, the ones which you want some years and not others. For the “base fleet”, you want to own, it’s cheaper.

Of course, not-yet-built roads are by definition receiving gas tax funds paid to drive on entirely different roads.

So one approach to edge in the direction discussed by Cap’n Transit is to dedicate automatic formula funding only to repairing existing roads, while new capital works competing for federal capital subsidies have to compete for funds on a cost-benefit basis.

That’s not the main problem – in principle those roads could be the ones doing the subsidizing later, like with Social Security. The main problem is that gas tax-ineligible urban streets subsidize numbered highways.

Fix-it-first is a good frame for reducing total highway construction, but it’s not going to get around the user fee myth, and as far as transit construction goes it’s counterproductive; it’s true but hard to explain that New York is only building new subways now after a multi-decade fix-it-first program. The only way to detach road and transit funding is to detach their nominal funding sources.

The trainsit systems by their very nature can’t get as big as say the highway systems in that there is a cirtan number of people pur square mile they can go to. Also something like an Interstate highway needs a massive suport system of lesser highways to keep up and running while a subway station or a Amtrak station out in a rural needs a parking lot and to be along a good sized road and it’s in bussiness.

Fix-it-first is a good frame for reducing total highway construction, but it’s not going to get around the user fee myth, and as far as transit construction goes it’s counterproductive.

If the Highway Trust Fund component of the gas tax goes to fix it first, then the only eligible new construction is new construction that gets people out of cars, and only to the extent that is saves on road maintenance, so it selects for the most cost effective means of getting people out of their cars.

Spending money on new road construction is directly counterproductive for transport over dedicated transport corridors, so a formula based on fix it first that preserves the same amount of funding for transit improvements while cutting funding for new road construction would be an incremental benefit for transit.

You misunderstand the politics of this. As soon as fix-it-first becomes a slogan for highways, it raises questions about transit. I’ve had blog commenters complain about this very thing; it’s hard to explain that fix-it-first is the present-day policy for transit when new lines are being built.

Yes. And? I don’t see where in the comment there is any effort to argue that the pain is not worth the gain, in terms of undermining the constant onslaught on dedicated transport corridors in the guise of diverting road tax revenues from unfunded road maintenance needs to new lane miles.

A reasonable political compromise would be to spend a large chunk of transportation funding on road maintenance but relatively little on new roads. Road maintenance has tended to be neglected compared to new road construction in the US, as exemplified by the infamous I-35W bridge collapse in Minneapolis.

Yes, and I’ve heard that maintenance tends to have a better economic impact that building new roads anyway.

We’ve been heavily focused on road-building for about a century, and have probably built as many lane-miles as we’ll ever need—we’ve certainly put down more asphalt in general than we’ll ever need. If people want to add more highway lanes, they ought to at least be ripping out under-utilized parking lots from somewhere else. I wonder if an asphalt credit trading system (kind of like a carbon credit system) would be a good idea…

I’m not entirely sure how much funding should go toward highways, though. Cities have been getting shortchanged on street-repair funds for a long time, and I think it’s beginning to affect many small towns and older suburbs as well. The street I grew up on was probably put in during the 1960s. It is roughly 50 years old now, and both the road surface and the underlying water lines are disintegrating. Who would want to move into a place with 3-foot-wide potholes in front of it? I suspect the broken windows theory applies just as much to shattered pavement as it does to vandalism and other forms of decay.

I’ve seen a lot of transit advocates show disgust at virtually any road project, though we should be careful to choose our battles. After all, transit and intercity buses both need good roads and highways in order to carry their passengers comfortably, safely, and with a minimum of damage to the vehicles.

A lot of the serface parking lots in the last ten years in my home city have been eather turned into muti story parking decks that are mostly packed with cars and more then that the rest of the surface parking lots are getting turned into six and ten story buildings in my downtown area and areas that are with in five miles of it. I like how there are pently of surface parking lots for them to turn into new buildings and parking decks in that it makes them less likely go to knock down the old existing buildings that are cool looking.

My city has had no state or local support for road maintenance for about 20 years. The current mayor finally decided to bite the bullet and pay for all the repairs out of local funds. A lot of new brick has been installed, partly because *it lasts longer than asphalt*.

Unfortunately, our jackass state government (NY) is considering a property tax cap. While requiring local governments to handle stuff which is really the state’s responsibility, like Medicaid — these are “unfunded mandates”. This means eventually all the property taxes will go to the unfunded mandates, with none left for discretionary items.

The main “discretionary” items are road maintenance and fire protection. We’re sane enough in this county that road maintenance will be removed first (we’re not stupid enough to cut fire protection and will probably defy state law to maintain it if necessary). I expect to be driving on gravel and dirt roads in 5 years if the state government doesn’t get its head screwed on right, eliminate the property tax cap, and start funding everything it requires counties and cities to do.

Thanks for correcting citing my photo Yonah. Much appreciated.

As far as the actual article, I like your thoughts for a short term solution. Long-term, I think we as a nation (or world) need to develop better tools to track driving. Something VMT based (with a gas tax component as well) seems to make more sense in my head. Better yet would be something to track which roads are used, so as to more accurately fund roads.

I think the notion of the gas tax as a user fee is fine. The argument that is being missed is that a portion of that fee funds mass transit, which keeps a number of cars off the road (thereby allowing the road to have reduced congestion for you).

The American gas taxes have been more-or-less dedicated to roads, making people think of them as a user fee. European ones are different – they’re taxes on roads, just like taxes on alcohol and cigarettes. There’s a 1920s-era quote from Churchill calling the notion of dedicating gas taxes to roads ridiculous, saying it conflicts with both the authority of Parliament and common sense.

The upshot is exactly as you say: in the US, the fate of transit spending is tied to road spending. This is why the road to transit funding ratio is fixed by formula, and why APTA found itself lobbying against a climate change bill. Any attempt to fix the problem by merely changing the source of taxes from gas to VMT will miss the point, because there will still be this tie between road and transit funding. Better would be to sever them entirely. Make the gas tax a gas tax, toll the highways, and fund transit out of either the general budget (it already gets raided all the time, so it can’t really get worse) or a separate dedicated source.

The Senate Finance Committee has the opportunity to break the connection between the gas tax and transportation funding in the upcoming reauthorization. The last few extensions of SAFETEA-LU added funds from general revenues, so there’s a precedent. Proceeds from the gas tax alone won’t fund everything that the administration wants in a transportation bill. So a reasonably ambitious bill will have to be at least partially funded from general revenues. Once that happens, the politics of the split between transit and roads will change.

Correct me if I’m wrong but isn’t the French autoroute system almost entirely tolled and new projects taken on by private companies? Which is to say that it’s an entirely user-fee, profitable system outside of the cities, where tolling was historically impractical. The lack of French general fund spending on highways does mean a lack of investment in them, as France has an excellent and comprehensive system of freeways.

My father’s (Scottish) take on high petrol taxes was that car ownership was restricted to the wealthy in Western Europe until the 1970s, allowing petrol taxes to be seen as luxury taxes.

On the question of funding for transit, how do you maintain hope that efficient transit systems will be built if they’re not market-driven, whether by fares or property taxes on those whose property values increase? It seems that the best new systems in the world are in Asia, where farebox recovery is typically over 100%. The best old systems likewise where almost entirely built when they could still turn a profit, at least operationally. When funding is entirely political then the decisions are entirely political, such as the New Jersey Riverline, or my own town’s Princeton ‘Dinky’ branch line, which spends over $10/passenger to move them five miles.

My personal dream is that systems like the New York Subway could be privatized, have their costs slashed sufficiently to become profitable and then rise to the level of service and quality found in places like Hong Kong. Is that impossible?

Thanks for your interesting articles Yonah, even if I struggle with their fiscal ideology.

Yes, that’s impossible.

The only way to privatize transportation is to privatize *all of it*, at which point it faces a “rights thicket” problem.

Transportation can only be built with eminent domain powers, and as long as that’s the case, democracy demands that transportation be under the control of government. It always was political, from the days when every turnpike needed an Act of Parliament onward, and it always will be. By its very nature, transportation corridors have to pass through the land of many many people, and they therefore must be political creatures.

Now, as public enterprises, could systems like the NY Subway make a profit? Sure! As long as every single local road in NYC was operated as a toll road for cars, too. Free, government-subsidized competition forces prices of rail transport down.

We have one other very serious problem in the US, though. In many other countries, nobody has a problem with a public-owned company making a profit. Here, if it’s *loss-making* people demand that the government run it, but the moment the government makes it profitable they demand that it be “privatized” so that government “doesn’t compete with the private sector” (their words!). Apparently they hate to have competition. Examples include Conrail. Canada has the same problem, witness Canadian National. This is known as “lemon socialism” — privatize the profits, make the public bear the losses.

As long as this double standard is applied, where government enterprises are NOT ALLOWED to be profitable, where right-wing rent-seekers demand that they be sold off whenever they become profitable, we’re never going to have profitable public enterprises in the US. Whereas in Europe and Hong Kong and Japan (and obviously China), they have lots of ’em.

There’s no reason public enterprises can’t be efficient and profitable, except that crooked men always demand that they be sold off to private profiteers every time they become profitable.

Sidestepping the main argument, I have a caution:

So long as we argue about “American highway spending,” every rural and small-city citizen will feel directly and needlessly threatened. You need to talk about “American urban highway spending,” where that’s understood to include freeways connecting big cities and competing with HSR, but not the secondary highways that are inseparable from any kind of economic development in smaller cities, especially for agriculture and tourism.

I think that the right’s ability to wrap “highway spending” in the mantle of lifeline-for-the-heartland is part of why this debate remains confused in a way that benefits urban freeways. As with many issues, hard-working rural people become mascots for a highway coalition that has no intention of really serving their needs.

Think about how you could frame the idea so that a centrist, noncrazy rural politician like Montana senator Jon Tester could appreciate it. Montana gains nothing from more freeways around big cities in Texas, after all.

Yes, I know that rural roads can lead to unsustainable quasi-urban development, “hobby farms” and the like. (Montana knows all about that, too.) But that can be a separate debate in the rural/exurban context, more effectively addressed at local levels.

Dedicating formula grants out of gas tax proceeds to maintenance of existing highways (interstate, national, state, county and township) protects spending on maintaining rural highways from being raided to boost outer suburban developer profits.

Which is, of course, at one and the same time the political opportunity it represents, and the political problem with it.

The basic problem is that the biggest waste comes precisely in places like Montana. The suburban freeways in Houston and Dallas are huge money-losers, but by a smaller amount relative to their usage than I-90 through Montana.

An even more basic problem is that many of the suburbs of Houston and Dallas culturally identify and are identified with Montana more than with central cities. If the point is to be culturally sensitive, then merely saying that rural roads with grade crossings are fine but suburban freeways aren’t simply does not cut it. There is some Interior West liberal populism that pushes back against Texas oil and gas interests and is nominally in favor of environmentalism, but it’s not and never been pro-transit or anti-freeway.

Another big Freeway money dump would be the 120 mile long Coalfields Expressway a four lane semi freeway that can’t deiced if it’s a four lane wide highway or a interstate. It’s going though some of the most rural parts of Vriginia and West Vriginia and though some of the most mountous so that means it’s going to cost mondo money to build and to operate.

They could have better spent the same amount of money on the Coalfeilds Expressway by upgrading existing two lane roads by eather widening them or removing and bypassing the existing deadman’s curves and hairpin turns on them. Or add more passing lanes to them not blast out a whole new expressway though the middle of no where.

What happened to our rural rail lines? Seriously, rural areas have the same problem with overspending on highways and underspending on trains that everyone else in the US does. In Ithaca, I *should* be able to hop a train to New York, Chicago, Philadelphia, or Toronto, but I have to *drive 60 miles* to get to the nearest train station.

The Lehigh Valley RR isn’t going to be reincarnated and start running the Black Diamond through Ithaca between places that have …. population….

Ithaca’s the *only* growing metropolitan area in Upstate New York. And with the ludicrous number of college students (more than *20,000*) the intercity travel demand is higher than you’d expect by the raw population. (The metro area is over 100,000 already, to the disappointment of the “Keep Ithaca Small” group.)

Now is the time to get intercity rail service back in, before it becomes ludicrously expensive to acquire right-of-way. Unfortunately, we’re so far from the nearest functioning line that most remotely plausible proposals depend on someone else getting their act together first. (Even Cortland can’t plausibly try to get service until either Syracuse builds the Park Street Bridge or Binghamton gets service, and Binghamton is pinning its hopes on Scranton… and Ithaca is harder to reach than Cortland. A Cortland station would be a major improvement already, though.)

Define upstate. To some people that’s anything north of 96th Street on the West Side and 86th St on the East Side.

In some minds Orange County is Upstate. Orange County is growing very fast. On the other hand the growth is coming from people who commute to Manhattan. Saratoga County is definitely Upstate, It’s growing twice as fast as metro Ithaca.

There’s nothing between Ithaca and anyplace else. It’s unfortunate it’s off in the middle of nowhere. There’s never going to be another Lehigh Valley railroad hoping to compete with the Delaware Lackawanna and Western or the Erie for traffic to Buffalo. Take the hourly bus to someplace that’s not in the middle of nowhere.

Upstate defined by metropolitan areas, so anything which is a NYC suburb doesn’t count.

Saratoga County was bleeding population from Albany last I checked, so the consolidated metro area wasn’t growing. But perhaps my numbers were out of date.

Albany county grew between 2000 and 2010. I’m not in the mood to go rummaging around in the Census’ website to find information about Albany itself. Declining population in the core of a metro areas can only happen in metro areas that have a core.

“Take the hourly bus to someplace that’s not in the middle of nowhere.”

And if there were an hourly bus to *Syracuse*, that would be a more reasonable way to get to points north. But there isn’t; there isn’t even a bus at all. Trips northward are by car, period.

The bus goes to Binghamton and then New York City, taking 4-5 hours minimum. I, like many people, do not like spending 4-5 hours on a bus. The fact that the buses are as busy as they are indicates underserved demand.

Despite its drastic shrinkage over time, Binghamton is big enough to deserve rail service to NYC, and Scranton certainly is. (They are both arguably in the “middle of nowhere”, but then so is the whole of France apart from Paris.) But neither has service.

No surprise that Binghamton and Scranton suffer when Columbus doesn’t have service, Cincinnati has service three days a week on an unhelpful route, and Bethelehem/Allentown has nothing. No *surprise*, but it is a *disgrace*.

The bus goes to Binghamton and then New York City, taking 4-5 hours minimum. I, like many people, do not like spending 4-5 hours on a bus.

Then move someplace that has enough people to justify having a train pass through. Until then you are SOL.

Under FRA rules and American operating practices, it’s insane to try to serve Ithaca.

Under modern rules and operating practices, an hourly shuttle from Ithaca to Syracuse through Cortland making the trip in 1:20 is easy and unambitious (and could even tram-train to Ithaca proper instead of dead-ending on the Cornell campus), and a shuttle making the trip to campus in 0:52 is ambitious but possible.

There isn’t enough demand to entice one of the low cost bus carriers to the route and you wanna run trains?

Sure, why not? The low cost bus carriers aren’t serving a lot of reasonable city pairs: just try getting to New Haven or Providence on Megabus.

Why would you want to use Megabus when there’s cheap rail service? New York to New Haven on Metro North and Boston to Providence on the MBTA?
…. unless you think there’s overwhelming demand for New Haven to Providence. The only people who pine for New Haven to Providence on Megabus are Yalies hoping to nail someone from Brown or vice versa. Not a big market.

Seattle is a good example of shifting transportation funding from roads to transit. The graph ( shows how road funding will go from a 70/30 split to 30/70 in just a couple of decades. That’s the good news.
Unfortunately, mode shift from cars to transit remains about the same as before, around 10% transit on all trips, with cars doing the heavy lifting through at least 2040 and beyond.
Seattle’s entry into the light rail arena has been a real disappointment with lavish facilities, massive subways, hundreds of millions per mile to build and few riders to show for several billion dollars in investment so far. Net subsidies for riders has more than doubled on the bus routes replaced by rail.
This is not the way you coerce the majority of taxpayers to give their road dollars to the transit planners.

Seattle’s doing fine. The problem with their light rail is that they built the weaker extension first, and the trunk route (to the University) won’t be done for several years. Net subsidy per rider will be far lower than it was for the buses, the moment U-Link opens.

You’re blowing more smoke than an Baldwin 2-6-2. U-link only adds 3 miles of extension north and two more stations at a cost of 2 Bil.
The current system is costing over $7.00 per rider, and double that if you include depreciation and interest on the debt run up. Ridership is only 60% of that projected in their FFGA to the feds. Yes, the next two stops are the low lying fruit, but even if they get all 70,000 new riders from two stops (more smoke AND mirrors), they are still way more than the buses being replaced.
Meanwhile, Metro, the bus side of the house is delivering rides at under $4 per rider, which includes depreciation.
“Seattle’s doing fine?”.
As Donald would say – You’re fired!

You may not have noticed that people actually want to go to and come from the University. Smoke blowing indeed.

The entire set of proposals were based on doing the north-of-downtown part first, and they never really redid the ridership numbers when they decided to do them in the “wrong order”. Just wait and watch, you’ll find out that you’re blowing smoke.

I will admit that Sound Transit’s ridership numbers originally presented for Central Link were a bit dishonest, as the model had *not* been properly reworked for the absence of University Link (or even of North Link, really). I actually spotted that at the time, but why spoil a good long-term plan by pointing out that it was staged in the wrong order (which everyone knew anyway)?

“Net subsidy per rider will be far lower than it was for the buses, the moment U-Link opens.”
As long as we’re talking about being honest with the numbers, a quick check of Sound Transit’s website for U-Link is only forecasting 39,000 boardings in the year 2030 for both new stops.
Discount that back by 2%/yr gives about 30,000 the day it opens (a huge maybe).
$63,000 per boarding makes the investment dubious at best. Factor in much higher operating costs and the subsidy remains much higher than the buses it is supposed to replace.
That’s the whole point I was making. Getting people to fork over more dough for transit that doesn’t make itself more efficient or reduce congestion is a tough sell.
Transit should be careful about how many ‘self-inflicted’ wounds are permissible before public support is frittered away.

One thing I would like to see more emphasis on is the fact that the “highways are paid by user fees” argument is a myth. The federal transportation trust fund has been bailed out by the general fund, as have state funds, and local roads are paid for (as mentioned) out of property and sales taxes, often as not. We’re already subsidizing roads, so why is it a constant refrain that transit can’t exist without subsidies? I think severing the connection between fuel taxes and transportation spending would be a good thing. We also need to look at the reality that fuel taxes are going to continue to bring in less revenue as cars get more efficient. That’s particularly true as the fuel tax is denominated in nominal dollars, rather than as a percentage or even real dollars.

There’s also the problem that American cities are so sprawl-y that it’s not possible to just drastically slash road spending. We still need to maintain existing roads. Building new ones I can do without. But many of the roads in major cities are in horrible shape right now. (I noticed this up close yesterday while dodging massive potholes as I biked across Minneapolis. Yowch.) We actually do need to spend a bit more on road maintenance in many places. That money can and should come at the expense of new road projects. But with an ascendant GOP in many states, that money is likely to come at the expense of transit as well. It certainly is in Minnesota.

With that in mind, I agree with Yonah. More spending overall is necessary both politically and in many cases is also good policy. Making roads suck isn’t going to force people onto transit infrastructure that doesn’t exist or is woefully inadequate. We need to focus on new forms of funding, funding the road/bridge maintenance that we badly need, expanding transit, and avoiding new road projects.

At least that’s how I see it.

Right now many of the roads that have been already built are going to stay with us now in the year 2050 and there are not going to be very many new ones built or road widening projects in that most of the land right now in the US is already taken and construction projects are already way out of control right now.

To add: thanks to population policies that I strongly disagree with, the US is projected to add 100 million new people by about 2050, often in suburbia. That means that we’ll have to spend money on new roads for these 100 million. (Seems better to me to have no population growth and focusing on improving existing transit links to already-built areas, but that’s just me.)

They’ll need transport ~ whether they need the same spending on roads as we presently spend in suburbia depends on whether we continue to mandate maximum travel sprawl development, or permit development to cluster around locations to access dedicated transport corridors.

Oil and gas prices are going to shift a lot of that saying in that if by 2050 oil is $250 a barrel and gas is $12 dollars a gallon while at the same time hyrogen for gas is $8 to $15 a gallon a lot of people are going to think about living in the city. Also right now many of the US cities that have good ecomemies are building many new apartments and high rises right now. Also right now we are in the middle of a building boom of new light rail which are in away the streetcars grandchilden taking back streets and cities that where once theirs. We also have several major subway lines across the US and Canada being worked on.

The Suburbia model really is a question of how much gas costs or if it will be around by the year 2050? Also a lot of the new people moving in to this county and the existing people and their kids are mainly moving to southern and western cities in the sun belt right at this time. So what ever happens in those cities at the moment are what will happen to the bulk of the people in the US in 2050.

Also note that the US imposes one particular suburban model, via zoning regulations and through cross-subsidies for sprawl development, which is far from the most common suburban type form when it is not imposed as a matter of policy.

A more normal suburban form is to be oriented to a local center, which offers the connections to neighboring centers and to the urban central place for the area. Given that we already have 30mph neighborhood electric vehicles that work just fine over five to ten mile distances, small suburban villages connected into an effective regional transport corridor could each be the central jumping off point for a radius of five to ten miles around.

The most straightforward way to do that might be to provide transit connections to a commercial district or professional park while removing the subsidies for sprawl development and the single-use, sprawl zoning mandates, allowing infill residential development to be established to take advantage of that transit connection.

While that implies a substantial amount of “build to allow infill” of transit systems that will not look good to the beancounters until after the infill has happened … its a lot less expensive than building all new cities to provide places to live for the over 150m Americans living in suburban areas.

Ooh, you support education of women, sex education, and easy access to contraception? Those are the policies proven to reduce population.

If so, are you *sure* you’re a Republican? ‘Cause I agree with you! (Or are you a different Chris?)

Maybe a good discussion point is what is happening in Missouri, specifically the state DOT. It traditionally has a low state gas tax and has done pretty good job leveraging funds. However, It will make drastic cuts, reduction in faciliites which were probably overbuilt to begin with and will not pursue any new highway construction for at least four to five years as it focuses solely on maintanence of its current highway network.

The reality is a lot of states will be in that situation very shortly as their is not political will to raise gas taxes. The plus side, it will greatly curtail the ever expanding number of lane miles.

^Missouri is a bad example for transit. Bi-State (dba Metro) only gets about one percent of its annual operating budget subsidized by the State of Missouri, and that’s essentially a wash when paying back excise and payroll taxes.

So as you can see, a state can cut its new road projects, yet still provide little for transit.

There is another rationale for using highway user fees to fund transit beyond covering externalities. This is the Mohring effect, which describes the increasing returns of providing additional transit service. As a transit agency adds service, the Marginal Social Cost decreases. Why? Because an increase in service benefits all users not simply the additional user it is meant to satisfy. (Not all goods see this effect: 10 bus riders enjoying headways of 10 minutes vs 11 bus riders enjoying headways of 9 minutes; 10 apple eaters enjoying 10 apples vs 11 apple eaters enjoying 11 apples). The problem for transit agencies is that the optimal price of providing service (equal to MSC) decreases as service increases. As a result, the optimal price can fall below the AVERAGE cost of production (but not of course the marginal cost of production). At this point, the transit agency is losing money. As a result, it won’t provide the optimal level of service unless it is subsidized. Where do these subsidies come from? Well, efficient tax policy suggests that you implement a tax on a good that has the highest inelasticity with respect to price (so that the tax will cause a minimal distortion). Automobile taxation fits the bill nicely.

Leave a Reply to Ocean Railroader Cancel reply