« quote of the week | Main | unhelpful word watch: captive rider »


Feed You can follow this conversation by subscribing to the comment feed for this post.


Sorry, but I can't see this helping. No doubt there are many transit agencies in the US that take their operations seriously, and I have no doubt that you are probably running a good agency. But plainly speaking, there are some that are a complete mess.

As it stands, there isn't really any public transportation in the country that has a farebox recovery ratio above 80%. The one city that should be able to get over 100% is running at 52.4%

As it stands, there isn't an agency in the US that isn't in "emergency" mode because of their failure to self-fund operations. How easy would it be for a local government to just cut funding a little more, and then claim for a bailout because they don't have the funding (never did) to keep it running?

Transit agencies in the US simply do not have the discipline to make the hard decisions like cutting underutilized routes, laying off administrators that are redundant, and keeping union demands on the sane side.

The lack of discipline means that this money will not be used to maintain routes and service levels, but rather will be used to maintain the status quo of poor management and deficient accountability. $9 billion would never be enough, because the incentives to abuse it are too great for any city official to ignore.


Any good metric to measure transit agency effectiveness? FRR is probably a bad one; as it depends highly on the land-use patterns (and competing auto infrastructure) of a city or region--and many social service (or "coverage") routes are never going to have a high FRR, as the busses run empty most of the time on those--they serve a purpose other than efficient transit.

Here are a few ideas, though:

1) OVERALL cost per route-mile; with overhead items (such as administration) included in the numerator. Many operational cost metrics attempt to measure marginal cost/rm, which is useful for determining how much is spent or saved by adding or removing service; but an overall cost/rm is useful for noting how much stuff is spent on items NOT directly related to moving people. This won't always be directly comparable between agencies, as labor costs will vary from market to market, and some agencies play a larger role in planning and infrastructure design than others.

2) Labor cost/rm, compared to median salaries for similarily-skilled positions in the market. Basically-does the agency bargain well with its transit unions, or are people being paid $100k a year to open the doors?

Measuring the management effectiveness of transit agencies is probably a useful exercise. Of course, many agencies are held captive by larger political organizations which may undermine their effectiveness in various ways, so simply blaming the agency sometimes will be aiming at the wrong target.

Cap'n Transit

Danny is wrong; if he'd checked the data, he'd know that there are agencies that have a recovery ratio above 80%.

I think this is useful for telling us what would be needed to keep transit agencies running at their December 2009 levels for the next 2.5 years, at least.


Not only am I aware of the profitable lincoln tunnel buses, but I am also aware of the profitable long island luxury buses and the chinatown jitneys. I am aware of a few profitable tourist train operations, as well as a few bus operations that maintain profits even in the middle of the boondocks of eastern idaho, wyoming and montana.

But the difference between us is that I have already realized that these agencies are all privately owned and would never qualify for a bailout from an emergency fund like this. This idea has only one purpose: To keep mismanaged agencies from ever having to deal with the refiners fire.

This isn't directly about farebox recovery ratios. Transit agencies are mismanaged. Period. They are over administrated, pay above market wages, and their service structures are designed politically for vote winning rather than efficiency and effectiveness.

This idea is one giant perverse incentive waiting to happen. Cities that are running into budget problems will willingly cut their transit budgets first (if they don't already do so), for the simple reason that they will qualify for a bailout (excuse me...emergency fund). Transit directors will use this as a nice little slush fund to keep them from firing their duplicative administrative buddies. I can guarantee you, you can give $9B away, but it wont keep transit agencies any more solvent than they currently are.

Cap'n Transit

Danny, the difference between us is that I don't automatically jump to the conclusion that a low farebox recovery ratio is a sign of mismanagement.

I have not endorsed this fund; I was just responding to your comment that "there isn't really any public transportation in the country that has a farebox recovery ratio above 80%." I assumed by "public transportation" you meant transportation open to the public, not transportation operated by a public agency.

I don't know what your 52.4% figure is referring to; NYC Transit had a farebox ratio of 67% for both 2007 and 2008.

Finally, Bridgeport has a fairly high ratio for a municipal bus agency: 33.4%, which ranks it 40th among bus agencies and the highest in Connecticut.

Jarrett at HumanTransit.org

Farebox return ratio (FRR) isn't the relevant metric here. Farebox return always invites comparisons to more "profitable" businesses, often concealing the ambient subsidies on which such businesses rely, and fails to account for the low-productivity services that agencies run for social service "coverage" purposes.

Danny does speak for a lot of people who wonder whether transit agencies are able and willing to push back against labor and other intertial forces. There are also legitimate concerns about the distorting effect that sudden infusions of cash can have on the expectations of unions in particular. I think Ron is trying to address this by his proposed ways of ensuring that the money turns into service hours.

I do think that the reigning imperative at the moment is to bring down unemployment, and that's achieved by holding the line on pay/benefit expectations so as to hire (or not lay off) as many people as possible.


I am as distressed by the transit cuts as anyone, and the situation in California is particularly desperate. However, as a political proposal this is a non-starter. At first glance it seems a splendid idea, but it reflects a misunderstanding of the bareknuckled political realities.

Let's say I am a state, going through hard times (or claiming to). I need to cut $100 million from my annual budget, and I plan to go after transit, education, and roads. I am a conservative state that doesn't like transit, so my forumula is $50M from transit, $25M from education, and $25M from roads. Each of these cuts will be painful and anger various political constituencies.

Then I remember that under the federal Emergency Operations Fund for transit, the government will make up $50M worth of transit cuts for me. No such fund exists for education or roads. Great! I will just cut all $100M from transit, the feds will give me $50M to make up half the cuts, and I won't have to cut anything from education or roads, which spares me the ire of those politically powerful teachers' unions and drivers.

The net effect is that transit fares just as badly as it would have without federal intervention. And the delicious irony is that the emergency fund for transit has done a magnificent job of saving education and roads from being cut. It has also saved the political husk of my state's governor, so perhaps a more apt name would be the "emergency gubernatorial re-election campaign fund"!

If you think there is a way to stop state governments from playing these games, then count me a great admirer of your optimism.

The reason transit is so easily cut is because the political constituency for transit is not powerful. Transit in America is for the poor, the young, and minorities. In other words, for those whose voice in political affairs is marginal.

The only way to prevent transit cuts is to make the transit constituency powerful. And you do that by getting the middle class out of their cars. This is accomplished through punitive measures such as a national gasoline sales tax, emissions standards designed to raise the cost of cars, user fees for federal highways, and so on.

Transit and automobiles are fundamentally at war. And there is only one way you win a war: you destroy the opposition. That is how it's done. Not with this manner of gimmickry that can be so easily gamed.


An FRR comparison can be useful to compare transit to highways. Assume capital is public in both cases (roads, railroads). Transit operations are public with an FRR of, say, 75%. Highway operations are private (car companies), but due to various direct and indirect government subsidies they probably have an "FRR" below 100% as well.

Now, my question is: what will it take to have transit run as a public-private operation, just like highways? For example, what if the only subsidy private bus operators needed was free access to toll roads? That would make it much more palatable to critics of transit subsidies, while in reality this subsidy could be substantial.

Alon Levy

Could there be a way to distribute federal transit aid based on a set of operating standards? This is how it was done with federal-aid highways - there were minimum requirements for width, pavement, speed, etc., which were increased as time went on. For transit, those standards could be made to encourage lower operating costs and higher ridership, e.g. proof of payment fare collection and limits on the number of administrators.


I liked EngineerScotty's ideas of relevant metrics for understanding if a system is mismanaged or not. Those should already be in use, and would be a great way to maintain accountability for our current systems.

But it still doesn't address the perverse incentive that both I and Pantheon mentioned. Really, the known existence of a transit emergency fund just means that transit will be where all the state and city agencies make their cuts first.

Boris is right that roads have low "FRR"s too. And that is probably the most major impediment to transit profitability...but it is not an impediment to transit efficiency. He might be right that the best solution would be PPPs, but PPPs also need to be closely watched because PPPs can easily turn into PPGs (public to private givaways).

I think Jarrett should take a trip down to a Central American capitol. I don't know about Nicaragua or Costa Rica, but in Honduras, El Salvador, and Guatemala you have an almost completely privatized system of public transportation.

In Honduras, the routes are determined by government agencies, and then licenses are sold to run on those routes. The licenses dictate the fare structure which everybody charges. These licenses are transferable, and there exists a second-hand market for them. The funny thing is, the second hand licenses often sell for above the original license cost, and the license values adjust almost hand in hand with inflation. That means one thing: These routes are profitable, and they are desirable.

There are no timetables: Buses run so frequently that a timetable would be a waste of time. Buses compete for your service: They still let you off wherever they want, but if they see one bus stopping for passengers at one stop, they will pass it trying to get the waiting passengers at the next stop. This makes for an incredibly fast trip.

The fares for intercity buses amount to approximately US$0.15. Think about that: These are profitable businesses that have fares about 7% of the cost of the typical $2.25 US passenger fares that belong to incredibly unprofitable transit agencies. Sure, there are things there that we can't get here like extremely low wages, and a poor carless population. But good profits at 7% of the fare cost does mean one thing: We have a lot of room to improve here.

How much would I pay for frequent service within walking distance of my home? I would easily pay $3-5 for a fare. Even at that price, it is cheaper than owning a car even in the most car friendly cities. When I see profitable bus companies in such abundance that I can have 20-30 second headways for a fare of $0.15, it makes me wonder what America is doing so wrong when they charge many multiples more. Maybe PPP is the solution to that.



I think a lot of people misunderstand the reason why transit profitability is so important. It is not because profits are important. It is this simple relationship:

If a transit agency has a farebox recovery of 99.99999999%, then the way to maximize efficiency is to cut service. But if a transit agency has a farebox recovery of 100.00000001%, then the way to maximize efficiency is to increase service.

Alon Levy

Danny, PPPs have always been PPGs. The countries that use them the most, the UK and US, not coincidentally have the highest transit construction costs in the developed world.

The reason Latin American transit costs so little is that operating costs are mostly labor, so they scale with local incomes. In fact one of the reasons Latin American cities often prefer BRT over rail is that rail construction costs are more technology dominated, making rail less affordable in developing countries.

What you say about cutting and increasing service to increase efficiency just isn't true. If a transit agency has an operating ratio of even 90%, then the ratio excluding administrative overhead is probably above 100%, which makes individual routes on average profitable. Conversely, even at 400% operating ratio, new routes require extra capital construction, which may push down the recovery ratio including depreciation below 100%; at any rate, new routes they may be marginal or compete with existing routes, reducing efficiency.

Just because Wal-Mart is profitable doesn't mean it should open a supercenter at every street intersection; just because a transit system is profitable doesn't mean it should expand to include more routes.


[i]Danny, PPPs have always been PPGs. The countries that use them the most, the UK and US, not coincidentally have the highest transit construction costs in the developed world.[/i]

Not true, there have been plenty of successful PPPs in their 200 odd year of existence. The New York City Subway system would never have existed as we know it without PPPs.

[i]The reason Latin American transit costs so little is that operating costs are mostly labor, so they scale with local incomes. In fact one of the reasons Latin American cities often prefer BRT over rail is that rail construction costs are more technology dominated, making rail less affordable in developing countries.[/i]

Also completely false. Rail construction technology has advanced, but the rails themselves have stayed mostly the same, and as such can be done by manual labor instead of high technology equipment. An example of this is Richard Posner III's Railroad Development Corporation, which has successfully constructed or reconstructed rail lines in many third world countries using the same technology that was available at the turn of the 20th century.

The true reason for a lack of rail in Latin America has a lot more to do with the lack of capital and poor property rights(which exacerbates the lack of capital). Buses, being on the low end of capital needs, are much easier to launch.

[i]What you say about cutting and increasing service to increase efficiency just isn't true. If a transit agency has an operating ratio of even 90%, then the ratio excluding administrative overhead is probably above 100%, which makes individual routes on average profitable. Conversely, even at 400% operating ratio, new routes require extra capital construction, which may push down the recovery ratio including depreciation below 100%; at any rate, new routes they may be marginal or compete with existing routes, reducing efficiency.[/i]

Operating Ratios and Farebox Recovery Ratios are not the same thing. Operating Ratio includes all revenue such as advertising, subsidies, and fares. Farebox recovery only includes fares. Since fare revenues are the only revenues where operating costs are proportional, this means that FRR is essentially a "contribution margin". Because of its contribution-margin-like status, it helps to determine operating decisions much more than operating ratios do. ORs are more for investors than managers.

Even when a system as a whole is operating at a loss, if the contribution margin is positive, then it is rational to increase service until the marginal increase in service costs the same as the marginal increase in revenue (where contribution margin is zero).

Alon Levy

I was talking about farebox operationg ratios. Forgive me for not typing a three-word phrase over and over.

Fare revenues aren't the only revenues that are proportional to ridership. Advertising is proportional to ridership as well - the rates scale with the number of eyeballs. Subsidies are excluded from farebox operating ratios for obvious reasons.

In fact, advertising may be more proportional to train/bus-km than ridership. The extra routes that an operator would add, or conversely cut, are the more marginal ones, which would get less revenue than the average. Posters on trains have rates that scale with ridership, but bus and bus stop advertising scales with bus-km and the number of bus stops.

The NYCT construction was rife with corruption and took decades from planning to execution, in part due to concerns about the corruption. But no, it wasn't a PPP in the modern sense - it was built publicly and leased to a private operator. The modern PPP is a creature of Thatcherist Britain.


I don't even know why I bother responding to you. FRRs are not the same as ORs no matter how much you try to say they are.

And the subway system was built by the operators. The contracts were for construction and operating leases. Funding came through city issued bonds which were paid back in full with interest over the lease period. In fact, of all the things that NYC has done over time with private businesses, the subway system was probably the least corrupt of them all. It wasn't until the IND system was built that corruption became a concern.

PPPs have been around for a long time, and have been used successfully for hundreds, if not thousands of purposes. Just because you can think of a few anglo examples of corrupt ones doesn't mean a thing for the whole class of partnership. Your ideological absolutism does nothing to advance the cause of transit nor urbanism.

Alon Levy

I'm talking specifically about farebox operating ratio, which is the headline figure quoted for transit agencies. Sometimes people use other terms; the one that they end up referring to is the FOR.

Corruption was a serious concern with subway construction. Immediately after the first line was built, the reformers were adamant that the city stop colluding with Parsons and the IRT. That's where the Dual Contracts with the BMT and eventually the IND came from. The issue of public versus private operation was there from the start.

And if anti-PPP ideological absolutism is what gets Continental Europe rail investments for half the costs of Britain and the US, then count me in for absolutism.


I would love to see your cherry picked analysis of how continental Europe gets rail investments for half the cost.


I don't even know why I bother responding to you... I would love to see your cherry picked analysis...


Methinks some of your comments are starting to cross the line. Disagree without being disagreeable, please.

Alon Levy

It's not my analysis.

And by the way - "half" is a lowball. Subway construction costs in Continental Europe are usually about $250 million per route-km (unless you're in Spain, and then they're $80 million/km). This corresponds both to recent extensions and to budgets for future extensions. But in London, the cost is almost $500 million/km, and Crossrail is clocking at $1 billion/km, the most expensive project in the world outside New York. New York's two projects are $1.3 and $1.7 billion/km.

Links available on request...


Your link explains exactly why passenger rail construction is more expensive: The FRA makes it so. That isn't a failure of PPPs.

Furthermore, high construction costs for subways in NYC or London aren't failures of PPPs either, nor are they the failure of government construction/operation (which is the method both cities are using). They are a reality for building subways within rather than out.

When you build a subway outwards, you have no subway lines crossing your path and you rarely have sewer systems or water systems crossing your path either. Systems where lines are being built within an existing subway network (like the 2nd ave subway) are a huge pain to work with. You have hundreds of roadblocks (sewer, water, steam, other subway lines, power lines, communications cables).

The complexity of the crossways interfering with the subway path typically mean that building a new line through an existing Manhattan or downtown London will force you lower into the ground. Lower into the ground means more expensive stations, harder soils and bedrock, and tunneling under the water table. This is insanely expensive, and 100% of the source of the cost disadvantage borne by the NYC and London transit construction, as well as nearly every other city where you are building in rather than out.

Alon Levy

No, the link talks about consultants when comparing the UK to the rest of Europe. PPPs are part of the same structure leading to cost overruns.

The peer cities of New York and London are building in rather than out, too. Paris Metro Line 14 was built in the urban core only, as a relief line to existing overcrowded lines, like Second Avenue Subway. Unlike SAS, it had to be built very deep to cross under multiple subway tunnels. It still cost $250 million/km in 2009 dollars. SAS is slated to cost $1.7 billion/km.


Sure, I guess if you want to compare a 9km line with 2 new stations through soil to a 13km line with 10 new stations through bedrock, then yeah, the paris line (which is not deeper than SAS btw) was cheaper to build on a per km basis. But that is not an intellectually honest comparison.

Its all beside the point, because Crossroads, Line14, and SAS are all fully public ventures. There is no SSS involved in any of them, and so zero percent of the cost difference is due to public private partnerships.

Alon Levy

The Paris line didn't have 2 new station. It had 9. Some may have had transfers to other stations, but RATP still had to dig out shells for them. And yes, it is deeper - it goes down to about 20 meters if I'm not mistaken, versus 10 for SAS.

Length isn't relevant - I'm talking about cost per route-km. But the cost in question for SAS is for Phase 1, which is 3 km with 3 new stations.

The soil/bedrock thing is just wrong. Yes, Paris has softer soil. No, it doesn't matter - it has catacombs, so underground construction has to be just as careful to avoid causing buildings to fall down. And we could argue it the other way - SAS Phase 1 doesn't cross under anything, and doesn't cross water whereas Line 14 does.

SAS is public, but it's built within a culture of consultant creep, single-bid contractors, and PPPs, instead of in-house expertise.


Paris had two new stations. The rest were dug as an extension to an existing station. SAS had 10 new stations. Six were extensions to existing stations. All of the stations in phase I construction are brand new, with the exception of 125th.

The soil/bedrock thing isn't wrong. Paris had soil, with rock aggregates. SAS is mostly bedrock, with clay and other hard rock. Since they are using TBMs, collapse doesn't matter, because TBMs create their support as they move. But ground material does greatly affect cost, because the rate at which a TBM can drill varies greatly by the type of material it is drilling through. Rates vary between 6-7 feet per hour to as little as 3-6 inches per hour for the hardest materials. Some areas heavy in bedrock were so difficult to drill through that they changed the plan from TBM to cut and cover.

SAS depth varies between about 15 meters to about 23 meters, depending on cross pathway conditions as well as station connections. There might be some pathways which are shallower, but I haven't read about them yet. There are several lines with which it has to deal with cross paths, 12 to be exact if you include both termini. There is one more crossing line planned. It also crosses the path of the under-construction section two of water tunnel #3. Even given that the first phase doesn't cross very many pathways, the line depth still has to accommodate phase II and phase III construction.

You have yet to provide proof of any consultancy, single-bid contractors, or PPPs in the SAS. You have also yet to provide proof for a lack of consultants, single-bid contractors, or PPPs in france. You have also yet to prove that the cost differences are due to these factors.


I'm done here. I'm not going to argue with someone who uses easily debunked made up information about less than a handful of cherry picked infrastructure projects which are not even PPPs to try to support an unsupportable claim that PPP "culture" results in higher construction costs. I'm just baffled at the level of ignorant fabrications that you can come up with.

Alon Levy

Danny, you're still talking about the full SAS, not Phase 1, which stretches from 63rd to 96th and doesn't intersect any existing track. 63rd is an existing station, but the rest are new. On Line 14, all stations are new - they weren't extensions of existing stations like 63rd on SAS, but involved new tracks and platforms.

Stop calling people ignorant. People who live in glass houses shouldn't throw the first stone.


Silly Ron, operations will never get money. Where's the money in that?

Every big system in the country is cutting service but plowing forward with expansion. Dulles, 2nd Ave, Expo Line, etc etc

Expansion gives money to engineering and construction firms, and is great for labor, electrical, plumbing, carpenter unions etc. So many people to employ! So many parts to buy! So many places to profit.

Operations goes to....drivers. And that's it. No money for Bombardier, no money for Clark Construction, no money for the unions, except for the drivers union. Riders will benefit? Sure, but riders don't lobby.

Jarrett at HumanTransit.org

Ron Kilcoyne responds as follows:


I have read all of the comments and am a little frustrated that for the most part they don’t really deal with the message I was trying to convey. My message was a “call to action”.

Pantheon called it a political non starter. It is a political non starter only if the people concerned about transit cuts don’t bother to contact their legislators. Don’t worry about what others will do –it doesn’t take a lot of personal phone calls, fax’s or e-mails to have an impact. It is important that we state what our needs are. If we get nothing – nothing ventured, nothing gained. But the more of us who do contact our Senators the better chance we will get something.

The two substantive critiques – local government will cut transit funding if they know they will be receiving federal dollars for transit and that dollars will go to salaries instead of service are addressed in the proposal.

To the former point I propose that the funds will only go to restore or preserve service in which was cut or announced before the bill is introduced. In other words the cuts were implemented or planned when the local governments could not anticipate the federal government would come to the rescue.

On the latter point an agency would have to demonstrate that it operating at least as many additional hours that the funding they receive could buy. For example if an agency is receiving $500,000 per year and their incremental cost of adding an hour of service is $50 per hour, they must add 10,000 hours of service per year.

There are poorly run transit agencies, but most agencies are lean. And faced with declining revenues most agencies have made cuts in administrative costs. This proposal is intended for service only.


In response to Ron Kilcoyne,

I was perhaps a little clumsy in my use of language. By political non starter, what I meant was that it will accomplish nothing even if it is enacted. Realizing this, the federal government will not even consider it.

I am not convinced that the provision of applying funds only to previously announced cuts will prevent this bill from encouraging future cuts. People remember precedents, and they act on them. Here are some examples of what I mean:

1. A child throws a temper tantrum and the parent gives in. The child learns his tantrums will be rewarded, even if the parent says this is a "one-time only" concession.

2. President Reagan gave a one-time only amnesty to illegal immigrants in the 1980's. This action is widely believed to have exacerbated the problem of illegal immigration, as a flood of subsequent illegals acted on their belief that another president would repeat it.

3. An economist I was reading said one problematic aspect of the bank bailout of 2008 is that, absent tighter regulations, it could cause banks to engage in even riskier behaviour in the future. This is because the precedent has been established that they are "too big to fail", hence risk for them is a no-lose proposition: if they win big they get the profits, if they lose big the taxpayers shoulder the losses. This economist was therefore endorsing the new tax on lending as a punitive measure.

In politics, there is no such thing as "one-time only". Everything is a precedent. For this reason, a kindhearted proposal like yours is often ill-advised. And a hardhearted refusal makes for wiser policy in the long run.

My solution for transit is the same one that some have proposed for banks - don't just bail it out, change the underlying fundamentals. It's a longer, harder road. But I promise you, it's worth the journey.

The comments to this entry are closed.

the firm

Jarrett is now in ...

Related Posts Plugin for WordPress, Blogger...
Related Posts Plugin for WordPress, Blogger...