Part of the Bay Area News Group

today’s free lecture: fare idea falls flat

By enelson
Tuesday, September 19th, 2006 at 6:45 pm in BART, Buses, Fare systems, Funding, transit equity, Transit vs. driving.

I really never thought that a story about one BART director’s idea to switch to a flat fare of perhaps $2.40 or $2.50 would generate so much interest, or end up at the top of the Trib’s Sunday front page. But what do I know?

Now the PhD’s are weighing in.

Aaron Golub, a researcher at the UC Berkeley California Transportation Center, sent me a copy of the open letter to the BART board he penned after hearing the flat-fare idea trumpeted on the radio Sunday.

I won’t reproduce the whole thing here, but he makes some interesting observations, such as how subsidizing longer-distance fares helped get more people to move to the suburbs a century ago in Boston and New York.

Ahh, let’s imagine a time when you had to pay people to move out of the city:

This expansion helped to depopulate the overcrowded urban cores and help settle the country’s first suburbs. More recently, the movement from zone and time of day fares to flat flares has helped to simplify payment options for riders, and reduce disputes between drivers and riders over proper fares. Even for systems where flat fares have been used for decades, flat fares bring questionable benefits to riders and operators and have troubling implications as well.

Point No. 1 Golub makes is that fares need to reflect costs better than they do now, and flattening them would make inequities and budgetary problems even worse:

The costs of operating BART vary greatly by time of day, by distance and by route. On the operating side, electricity, labor, wear and tear and maintenance all induce costs by distance traveled. Long trips cost more to produce than short ones. On the capital side, expensive tunneling in the East Bay and Transbay Tube means trips which use these facilities are using more expensive infrastructure than other trips. The large number of trains purchased and maintained to be used only during peak hours also exacts large storage and maintenance costs on BART, meaning that trips made during the peak hour cost BART more than trips made in the off-peak. In sum — BART costs vary significantly by time of day, by distance and by route, and fares should be structured to reflect this. Indeed, current fare structures do attempt to account for distance and route cost differences, though arguably, the function of distance is not steep enough.
The problems of not linking fares to cost are many. A rider making a Saturday trip from Fruitvale to Downtown Oakland subsidizes the long commute-hour trips from the far East Bay. This has important social equity implications, to be discussed later, but is also inefficient. By not varying fares with cost, expensive trips are subsized, meaning that more of these long trips are taken than would be if fares were more accurate. On the other hand — cheap trips are overpriced, and therefore underused. This exacts a cost on BART by being forced to provide more trips at a fare discount (long trips, peak hour, etc) and fewer trips where riders overpay (short trips, off-peak). BART loses fare revenues by not charging more accurate fares and this will only be exacerbated by moving to a flat fare.

Golub also believes ridership is likely to decline with a flat fare, and I have to give him credit for talking about my favorite economic concept, price inelasticity (I get shivers whenever I hear it):

For riders making peak-hour commute type trips to downtown jobs, parking rates, parking availability, traffic congestion and gas prices are all onerous enough to make these riders fairly committed to using BART on most mornings. BART fares are a very small part of the calculus in the commute costs — BART could probably increase fares quite a bit from the far East Bay before seeing any appreciable ridership changes. In economics, this behavior is called price-inelastic: behavior does not respond very much to changes in prices. By far, most of long trips from the far reaches of the East Bay are these fairly inelastic commute trips. On the other hand, riders in the urban core (Between say Berkeley/Rockridge to Coliseum to 24th street/Mission) make more off-peak and discretionary trips which are more price-elastic (respond more strongly to fare changes). Because of the inelasticity of the long commute trips, the substantial fare reduction from implementing flat fares will likely not increase ridership significantly. These riders might make the to-and-from commute trips on BART, but are unlikely to make other discretionary trips by BART.

You following this? There will be a quiz later.

Then Golub tackles a subject I’ve seen grownups kick sand into each other’s eyes over:

Furthermore, in only a few outlying parking lots is there additional capacity to handle more automobile access to BART. Indeed, in many stations, access problems are likely a bigger barrier to BART ridership than fares. Solving these access problems could probably lead to a larger ridership increase than fare reductions.

Some people just HATE providing parking for suburban commuters, expecting that they’ll walk a half-mile or so to catch a WHEELS or County Connection bus that will take them 30 minutes to get to BART when they could have driven there in 15.

And people out in the far ‘burbs making long trips are in the minority, so to speak, Golub notes:

It should be noted that one subset of the long trips — those from the East Bay to SFO or OAK – could rise following a fare reduction. These trips are such a small share of total BART ridership that they are unlikely to have an appreciable impact. It would also be questionable public policy to introduce such drastic changes to exact a benefit for a small segment of the riders.

And the majority will suffer:

In the urban core, fares will rise for a large number of riders. These riders, because of the availability of other cheaper travel options such as MUNI and AC Transit, and because they are making shorter, more discretionary trips, could leave BART for other options. The question here then is one a balance between the slight increases in ridership for the long trips, with the quite likely higher loss of ridership in the urban core. The higher elasticity for those trips with a fare increase should overwhelm the numbers of trips increased and lead to a reduction in overall ridership. Much more careful modeling of this question should be performed before proceeding with a decision on this matter.

And speaking of minorities and the disadvantaged, there are social equity issues to be pondered, such as how much we chose to cater to “choice” riders:

The riders making the overpriced (currently, and under a flat fare, even more so) trips are more often low-income and people of color. The riders making peak hour commute trips are so-called “choice riders” — wealthy and overwhelmingly white. Low-income riders mostly use the Richmond-Fremont line between Bay Fair and El Cerrito — the cheapest part of the system to operate — and are more likely to use the system off-peak and weekends than other riders. It is highly likely that a flat fare would exacerbate the already questionable social equity impacts of the current fare structure.

Then we fall deeper into the quagmire that is transit equity:

In 2005, AC Transit evaluated its fare proposals using detailed ridership information, and included a “Title VI” analysis of social equity impacts of the proposals. I urge BART to take the same measures to insure that board members and the riding public, to whom they are responsible, are fully informed of these impacts. In a time when MTC and the LA MTA and other operators are under increased scrutiny for purportedly inequitable funding policies, BART cannot ignore this issue.

There are also benefits to having a flat fare, but Golub implies that BART riders don’t much care about those issues, such as simplifying the fare structure. In our nation’s capital, for instance, fares are based not only on distance, but also according to peak-commuting hours.

And that brings us to the conclusion (when the bell rings, you may proceed to your next class):

BART should move to, not away from, fares which better reflect costs — fares which vary by distance, route and time of day. As smart media becomes adopted over the next decade, many operators will be making this move to better pricing and numerous discount options. BARTs recent move to un-bundle the charge for parking from the BART fare is a great step in the right direction. I write this letter as a regular BART rider, a transportation researcher and an advocate for better transportation options throughout the bay area, of which BART is an important part.

[You can leave a response, or trackback from your own site.]

14 Responses to “today’s free lecture: fare idea falls flat”

  1. Joel B. Says:

    That’s a lot to read. A lot, not even my comments are (usually) that long. The problem with wanting to make fares fair is that well guess what, a lot of those suburban commuter will just drive. People like to believe that BART is price inelastic, but um…apparently no one was paying attention during the spare the air days. Because if they were, they might see that BART ridership might be more elastic than BART likes to believe. Heck, the fact that ridership has expanded fairly with higher gas prices suggests that there are a lot of riders who’d be willing to ride at a lower fare than BART currently prices rides at. People even from Pittsburg/Bay Point make the choice to take BART or not, and when living in Walnut Creek I knew plenty of people who lived close to BART and worked close to BART, would have saved money by taking BART, and still chose not to based on time savings. Now, maybe if it had been cheaper sure, but not at the current price was it worth it.

    The probably with basing fares on cost, is that believe it or not, BART is not a genuine variable cost system. It’s actually much more a fixed cost system, once the tracks are down it doesn’t matter how much it cost to lay them, BART is going to be running on those tracks. See for example the SFO extension, BART still running trains, even though most people see that Caltrain is the better commute option. The real market solution, would be to lower long distance trip costs (to encourage longridership) and to raise short distance trip costs (because the revenue per trip generated is less, and who’s really going to go to the store on BART?) Is this fair? Well in some ways no, but who cares. Let me let everyone in on a little secret I was told long ago…Life isn’t fair. Craziness huh! Instead BART should try to maximize net revenues through encouraging long distance ridership over short distance there’s more fare money in it. (I am not advocating a flat fare, but a flatter fare.) Not to mention the fact that the “mileage” is based on BART mileage instead of “best vehicular mileage” so it costs an arm and a leg to get to SFO from D/P when heck, most sane people would just cross the San Mateo Bridge, and cover a lot fewer miles than BART does, but there the BART rider goes taking an hour and 30 minutes to take what is often a 30 minute drive.

    It goes unmentioned of course that the trains stop at every station, so certain stops have stations every 2 minutes, where some of those evil suburbanites who have their fare subsided have to take an hour to travel not all that many miles (it’s always important to remember even in traffic it’s hard to beat a car timewise). So is this “fair” no but again, life isn’t fair. We play the hand with the cards we’re dealt and make the best of it. That involves not worrying so much about fairness, but maximizing the overall utility and revenue of the system.

  2. Reedman Says:

    “Fair” and “Fare”

    I heartily agree with the premise that transit fares should reflect the cost of providing
    the service. The fundamental questions are: How do you apportion the capital investment?, and, Why government?. The first question is a transportation planning ‘third rail’ (touch it, and you are dead). Rail transit can never pay its capital costs because the real estate it uses is so expensive, oddly shaped [go try and buy a strip of land 100 feet wide and 10 miles long without eminent domain…],underutilized, and the fixed installation cost is so high. So, rail looks good when you ignore investment and focus on marginal cost (i.e. the price of adding 50 more commuters to rail is trivial). The only way rail transit gets built is by bond issues which are retired out of general revenues. Buses are the opposite. The highways are already there. The cost of buying a bus is trivial compared to the operating expense, which is substantial. The second question is actually the most basic. If riders of public transit had to pay 100% of providing their ride through the farebox (and that 100% reflected both the capital as well as operating costs), then private industry would enter the market. The huge subsidies to the operators of public transit guarantees two things: civil service job security and no competition to those operators.

  3. Frequent Amtrak Rider Says:

    “If riders of public transit had to pay 100% of providing their ride through the farebox (and that 100% reflected both the capital as well as operating costs), then private industry would enter the market. The huge subsidies to the operators of public transit guarantees two things: civil service job security and no competition to those operators.”

    If riders of public transit had to pay 100% of their ride, they wouldn’t take public transit. Note Even airline industry passengers don’t pay 100% of their trip. Neither do drivers on public highways and public streets and roads. If the private sector entered into public transit, they would slash service by eliminating the least productive transit routes. This raises public policy questions about access to jobs by people who are transit dependent. It also raises issues about traffic congestion and air quality.

  4. Ed Says:

    BART changing to a flat fare system has one significant possibility for increasing ridership. Currently, BART does not offer monthly or weekly passes for the following reason, “BART’s fare structure is built on a mileage-based formula, therefore weekly or monthly passes for BART fare are not available.” By changing to flat fares BART would be able to offer the sale of montly passes.
    Having lived in the Boston area, which has a largely flat fare fee structure with outlying areas being more expensive, I have at times purchashed monthly passes. I noticed in months that I had passes, I used the public transit system more extensively then during the months where I did not have the passes.
    Offering monthly passes has advantages for BART. By selling the passes up front, they are able the money up front. Also as a significant portion of the ridership are regular commuters, this would help to even out their revenue stream and make it more predicatble for BART.

  5. Paul Says:

    Selling unlimited-use monthly passes would almost certainly reduce BART’s farebox recovery ratio.

    BART has historically achieved a farebox ratio of about 50%, which is excellent for the San Francisco Bay Area. BART beats Caltrain, Muni, VTA, and most of the bus operators in this regard.

    Though I’m not a fan of monthly passes, because this pricing arrangement results in lower fares for frequent riders even though this group is less price-sensitive and in higher fares for occasional riders even though this group is more price-sensitive, a flat fare is not a prerequisite for a monthly pass. Caltrain has six fare zones and thus offers six different monthly pass prices. BART could sell monthly passes keyed to a specific trip value, and send you to the “Add Fare” machine if you happened to take a trip that exceeded the trip value you had chosen.

  6. Railer Says:

    You have no idea how much improvement BART can use until you ride something better. That better is the CTA in Chicago. It’s amazing.

    Flat fares, many people to help answer questions and keep the system clean and safe, passes for frequent riders; even for *gasp* tourists. You can actually understand the recorded announcements of upcoming station names and other CTA info. Brilliant!

    Chicago wants people out of their cars, not polluting the environment and that’s what they get. Their system covers a massive amount of ground, is simple to use, though not simple to cheat (seen anyone jump a wall, or pass through an open door into BART? I have many times). All of the fare cards are now tranferrable to the entire bus and lightrail system as well. Free, comprehensive maps abound. Bikes are encouraged greatly (nice and flat, Chicago). A couple lines run 24 hours too. It’s almost like they’re thinking in that city.

    Lower the fare and more people will ride. Period. Though the fares are lower, the increase in ridership, and reduction of cheats, will make up for what’s lost.

    I’m not a spokesperson for Chicago, but after you visit, you know that we in California do not always live on the cutting edge. Even the mayor of SF is starting to adopt a lot of the programs Chicago has in place. Check out Chicago. It truly is one of the greatest cities in this country.

  7. Reedman Says:

    In California, motorists don’t pay 100% of the cost of highways. They pay more
    than 100%.

    The gas tax for many years was put into the General Fund to pay for prison guards and
    welfare checks. Prop 42 slowed this practice, and forthcoming Prop 1A wants to ensure that gas taxes used for non-transportation purposes in an emergency get paid back into the transportation fund. [It would get tedious to get into Prop 87’s plan to tax oil production in California, or the handling of excise taxes versus sales taxes. ]

  8. Aaron Golub Says:

    Well, I guess since I started this thread I feel the urge to respond to these excellent comments. The comments bring up several issues which I glossed over in my original letter to the BART board. For one, as Frequent Amtrak and Reedman mentioned – transit riders don’t come close to paying their full (capital+operating) costs. This is completely true – BART riders pay about 60% of their operating costs and next to zero of their capital costs – all trips on BART are subsidized (as are all trips by automobile, plane, etc – it is explicit policy in the U.S. to subsidize travel). The question Reedman raises about how to even apportion capital costs is a difficult one. In a truly private system, like airport shuttles, the fare covers both capital and operating completely, plus profit. But society needs transit and auto travel for more than just the travel itself – it assists with the general economic activity and vibrancy of the region. More mobile/accessible societies can produce and consume more. Consumers, workers and industry benefit from this, not just the riders themselves. My argument was a bit academic – fares should vary to reflect total social costs (capital + operating) so that the system is used more efficiently from society’s point of view (expensive trips are not subsized, etc.) In actuality – nearly all capital funding come from State and Federal grants and Bonds, while most operating funds are from fares and local taxes, etc. A more simple (and more local) solution could be then to match fares to operating/maintenance costs only, and to disregard the capital aspects (tunneling, transbay tube, etc). On a related note, Joel B said: “BART is not a genuine variable cost system,” but I am not sure what he is referring to. BART spends about 400 million dollars a year to operate – about 4 dollars per passenger trip. Some of the most far-flung trips cost 20 or 30 dollars for BART to produce, on the margin. He also suggests moving away from cost based pricing to even more subsidization, as a “real market solution” to encourage longer distance ridership and to make up for low revenues from short trips. Better markets function from more accurate pricing and information, not less. Ironically, Joel’s goal of raising revenues for BART would only be compromised. Revenues will fall if fares are lowered in an inelastic market. Think about it: few new trips are generated (low elasticity) but fares are lowered. To raise more revenues in an inelastic market one would raise fares (what I am suggesting for the outlying stations). And, to be more clear – the market for BART is highly inelastic (though more elastic in the “urban core” market). The Spare the Air day experiment Joel mentions is a perfect example – ridership rose roughly 8% after a 100% fare discount! And a good portion of that travel was discretionary. That is a perfect example of the inelasticity of BART ridership. To meet Joel’s goal of milking the system of revenues (so as not to mince words), one would raise revenues across the board – not lower them. But, that wouldn’t be fair – bringing me to my next point – fairness. I agree with Joel here – nothing in life is fair. Fares haven’t been fair in the past, and there’s no reason they have to be in the future. Fares reflect a balance between some of this economics we are discussing, and the politics of the riders and the board. That is – fares aren’t “fair” because certain riders’ interests aren’t represented on the Board. Currently – about 65% of BART boardings are in the 16 “urban core” stations, with an average of 11,000 boardings per day per station. Only 10% of boardings come from the seven stations east of the hills (beyond Orinda), with an average of 3800 boardings per day per station. Why these more important core riders pay “unfair” fares, while the relatively few long-distance riders get more subsidized fares likely reflects an imbalance in the BART board or a lack of assertiveness on the part of riders, or both. I haven’t studied the dynamics of Board representation, so I can’t comment on this. I would hope that by having this discussion – both here in this blog, and in day to day life – a “more fair” fare can be found because more people will be better informed. That discount passes are more easily issued in conjunction with a flat fare, my only comment would be that fare media are going to be much more advanced in the near future, so that a wide variety of discounts will be possible – along with time of day and distance pricing (!). Yes – Chicago, New York, Boston etc are “better” systems – but those systems and land uses have grown up together over 100 years. BART was an attempt at limited coverage of areas left abandoned by the Key System and other rail systems in the inner East Bay, while at the same time provide coverage to far-flung suburbs – arguably a difficult task. From a performance standpoint – a much denser rail network should have been built west of the hills covering the old system, because land uses matched that network, while east of the hills should have been more of a commuter-type system with 30-minute headways, limited weekend service, etc. The result would have been much more transit ridership. Hindsight is 20-20. On the plus side – BART is key for making the morning commute work – about 45% of commute trips to SF from the East Bay are by BART. And as citizens who pay for it – we need to continue to work to make BART better – better access, “better” fares, etc.

  9. Tax attacks Says:

    Reedman is talking about the *sales tax* on gasoline, which is the same thing
    as the sales tax on everything else we buy. Those taxes (until prop 42) always went
    into the general fund.

    We don’t earmark the sales tax on tennis balls to build tennis courts.
    We don’t earmark the sales tax on restaurant meals to build restaurants.

    Somehow, people got the idea that the sales tax on gasoline was special,
    probably because all taxes on gasoline are lumped together in the total price.

    Even if you include the sales tax on gasoline, we don’t get all our road funding
    from gas taxes. In many counties, we have transportation sales taxes that
    devote a half cent or more of the tax on *everything* toward transportation
    projects. Buy a scarf, you’re funding transportation.

    Then we have transportation bonds, which don’t have a specific tax to pay for them.
    There’s no tax fairy to pay for those, they’re paid back out of other revenues, not
    gas taxes.

    Bottom line is that transportation in general, and roads in particualr, are not fully funded
    by the gas tax, with or without the sales tax on gasoline.

    And this is before you start to consider other related expenses, such as traffic cops.

    Remember, the farebox recovery of most roads (barring bridges and tollways) is zero.
    The tax on the gas you buy in Oakland doesn’t go to the nearby streets.

  10. Steven Hauser Says:

    Bus fares as barriers to use A complex, hard to understand and use transit fare structure is a barrier to transit use, he won’t be happy until we are all injected with a chip that reads various fare charges by tunnel,
    time of day, location, destination and weather and automatically
    deducts it from our bank account while giving Big Brother a complete
    GIS map of our presense so we can be safe from “terrorism”.

  11. Steve Taylor Says:

    I’m often reading comments from people who insist that transit should pay for itself, that they don’t believe in giving a subsidy to any form of transportation. The truth is that there is no form of mechanized transportation in use today that pays for itself, so why should transit be treated any differently?

    Let’s look at the major forms of transportation:

    Air Travel: Airports built, expanded and maintained using local, state and federal tax dollars. The air traffic control system and the TSA paid for by federal (tax) money.

    Rail Freight: The land the tracks are on was given to the railroads by local, state and federal governments in the 19th and 20th centuries, along with large sections of land on either side of the tracks that the railroads were free to use or sell as they saw fit. Imagine how much it would cost the railroads if they had to pay back the cost of all of the land they were given. There is also government paid over/underpasses and grade crossings to enable more efficient movement of trains as well as the government paid for roads to and from rail facilities.

    Shipping: Useless without port facilities usually paid for by state and local governments as well as government paid for roads to move goods to and from the ports. Also supported by government paid Navy, Coast Guard and lighthouse facilities as well as Army Corps of Engineers dredged ports and shipping channels.

    Roads: Gas (user) taxes and tolls don’t even begin to approach the actual cost of road and highways, let alone the indirect costs of traffic, pollution, injuries and deaths. There are not only the building costs, but maintenance, cleaning and patrolling which over the lifetime of the road cost more than the initial construction costs. Check the budget of cities, counties and states and they will all have budgets for roads that are not paid for by gas taxes but by general funds. If gas taxes paid for all of the costs of roads you wouldn’t need tolls or bond issues.

    If you eliminated all of the various types of subsidies every type of transportation receives in this country all types of travel would soon grind to a halt.

    So again, if no other mode of travel exists without some form of government subsidy why should transit be held to a different standard and be forced to pay for itself?

  12. Capricious Commuter Says:

    Well, Steve, if things continue to move in the direction our new US Transpo secretary is pushing them, people who commute by car to work will get to know just how far those subsidies don’t go. Secy. Mary Peters is a big fan of congestion pricing, or paying tolls generally to drive in high-traffic areas, then paying higher tolls to drive when everybody else is driving. Nice thing about tolls for policymakers is that it seems both Republicans and Democrats have made a pact not to call them “taxes.” And these tolls aren’t even called “tolls.” So far, however, I have yet to see congestion pricing proposed at a level that comes close to what people pay in transit fares to cross a body of water or jurisdictional boundary. The new San Francisco proposal that USDoT is offering to help pay for for is likely to be no more than $1.50 on top of the Golden Gate toll, which means that’s the max someone could save by driving off-peak. Hardly worth the delay of going off-peak or taking a ferry if you’re driving from your $900,000 cottage in Marin to your $250,000-a-year job in the Financial District.

  13. anon Says:

    Steve Taylor: many international container ports are in fact operated privately and without public subsidy, like in London (well, near London), Belgium, Singapore and Hong Kong. London’s airports are also privately-run.

  14. Steve Taylor Says:


    The operative word in your comment is “run”. While it is true many container ports and airports are privately run, the question is, were they built with private money? I do not know of any major airport in the world that was built with private, not government money. I do not know of any air traffic control system in the world that was not built and is not supported by government aid. Nor do I know of any container ports that were built without government aid of some sort (subsidies, tax breaks, etc.) for the facilities themselves or are not aided by government paid coast guards and port and channel dredging (Army Corps of Engineers anyone?). And even if they were (which I doubt), they would be useless without the government paid for roads that are used to move their cargoes from the port facilities to their final destinations. And what about the people who work at these facilities and without which nothing would move? Are you telling me that they don’t use some sort of government subsidized transportation (roads and transit) to get to their jobs?

    Roads, airports and container facilities can be privatized but I doubt there any trips by people, goods or the people who move them that do not utilize some sort of government supported or subsidized transportation, be it sidewalks, roads, highways, buses, trains, airplanes or ships.

Leave a Reply