Thursday, August 02, 2007

transit metrics again and again

Last week the County Controller released yet another audit/study of the Port Authority. The Press Release is online, but not the audit itself. I have a hard copy scanned here (large file: >24mb) for those who are interested. I will add it to my very abbreviated transit bibliography.

The audit looks mostly like some benchmarking of the Port Authority with some other large transit agencies. The data is mostly stuff you can pull yourself from the National Transit Database (NTB). The NTB really is a wealth of information compiled in one place, really a model for how public data should be made available. Check out their page dedicated just to their databases. For studying the Port Authority, I suggest you want to look at their profiles of the top 50 transit agencies.

Here is my only comment on the audit itself. It points out that the the Port Authority could be operated at a much more efficient manner, efficiency being defined solely by cost it looks like... PAT compares poorly in some sense for things like riders per hour, or revenues per hour. What does not get mentioned is what I have tried to explain in the past, that Allegheny County ranks high in terms of how much use it gets in the county. At least it did before these last cuts.

The problem is, you just can't have it both ways. If we want to have a transit system that ranks high in terms of usage, you almost have to have a system that provides more service over lesser utilized routes. Those are the routes that are the most cost-inefficient. If we want PAT to be like other transit systems, you are really saying you also want the utilization of our public transit system to be like these other regions as well.. and for most other systems that means less service and much less use than in Allegheny County.

Let's take one example... the Controllers Audit benchmarks the Port Authority with 11 other systems, one of which is Dallas. In Dallas the cost per rider is indeed lower than here ($3.80 vs $3.96 which is a 4% savings) yet the public transit utilization among commuters in Dallas County is 66% lower than here (3.55% of commuters there use public transit there compared to our 10.5% ). That seems to say that the Port Authority is doing a pretty decent job of providing a lot more service with only a minimally higher cost. But since the benchmarking never looks at comparable utilization metrics, the reader never considers that logic.

In many ways public transit has been a success in the region, providing a service chosen by more riders day in and day out than most comparable regions. If you just want to benchmark costs alone.. guess what: you could make the Port Authority a whole lot more efficient by cutting back to a small fraction of the routes in the system. The top 10 routes might have a fighting chance of making a profit.... but is that the point? Is that what we are striving for? That is pretty much the argument this type of benchmarking is making implicitly.

Add it all up and what do you get. Public transit seems to be in a perpetual fight for its existence these days. This round, the opponents of public transit won almost before the battle began by shaping the debate as only about costs. Once reducing costs became the be all goal, there was only one endgame. In military speak, you always try to shape the battlespace before you fight. In this case, one side succeeded, the other didn't even try.

Benchmarking has a role, but lets benchmark the right metrics. Output and usage are as important as just benchmarking costs which is all anyone seems to want to do these days. There are other important factors here.. Funny how nobody has gone out of their way to connect the dots between the report recently put out on the regions racial disparities, and how transit cuts affect those populations. Maybe there is a benchmark we should come up with to see how well low-income populations are served in different regions and how the Port Authority compares. We talk about all these issues as if they were completely independent of each other when nothing could be further from the truth.


Blogger Schultz said...

One word solution for this mess:


Thursday, August 02, 2007 4:24:00 PM  
Blogger Mark Rauterkus said...

I'd not say that everything needs to be with the "P" word... Rather, two words:

Allow privatization.

Allow private carriers.

Allow bus companies that drive from Murrysville to pick up in Monroeville and Churchill and Turtle Creek and so on. That isn't even permitted now.

Allow competition.

Thursday, August 02, 2007 6:47:00 PM  
Blogger C. Briem said...

privatization. take a little tactical advice that is so self-evident I can say it without taking sides... if your goal is convince Democrats to vote for a Republican you don't want to go around saying things like that. What are you thinking?

More seriously. My only question about privatization of public transit is how do you get around the cherry picking problem?

Thursday, August 02, 2007 10:29:00 PM  
Anonymous Anonymous said...

Privatization is an ideal target, but it has to be phased in slowly so that large service gaps are avoided. In the interim, the system does have to become more efficient.

Do any of you three ride public transit regularly? If you do, you will also have noticed that buses are frequently empty, even something like the 67A leaving downtown after 5:00pm does not fill up until Oakland. That is because there are also the 67F, 67J,67E, and 67C leaving near the same time so there is an over-supply of bus space. That is money wasted. The redundancy of 67s can be eliminated and demand better matched to capacity by implementing a "trunk and feeder system," which would convert the different 67s to one 67 between downtown and Wilkinsburg. From there people would transfer to the a,f,c,j,e. Commute times to would increase to the end of feeder routes because of transfers and slower service rates, but they would decrease, perhaps dramatically, on trunk routes. Further, I think everyone would agree that slower service to the burbs is better than no service at all.

Also, the Port Authority (I have talked to Bland and his outgoing operations officer, Bland is not at fault here) are somewhat averse to increasing service rates (read: speeding buses up on their routes by eliminating some stops in Oakland, etc.) because of fears of "bunching" downtown. What they do not give enough weight to, and Queuing theory explains, as service rate increases, the number of servers (i.e. buses) may decrease. This lowers overhead costs dramatically. Further, service rate increases are easier to achieve on trunk and feeder system because redundant buses are removed and stop gunking up the routes.

In short, there is plenty of room to improve the current system. From what I have heard, the Port Authority is going to begin phasing in the trunk and feeder system in 2008. This strategy has the support of the Allegheny Council, Harrisburg, etc. We should push it forward with vigor to make it happen quickly and with minimal screw-ups. It would be the first stable step in the direction of privatization.

Friday, August 03, 2007 10:07:00 AM  
Blogger Jonathan Potts said...

Am I mistaken in thinking that the private bus companies that used to operate in the region were begging the government to get them out of the business? I can certainly see outsourcing things like maintenance, but I find it hard to imagine that a truly public transit system can operate profitably.

That doesn't mean there aren't a lot of inefficiences that couldn't be eliminated without adversely affecting service. There are too many stops in many neighborhoods, and too many redundant routes. But schulz, take Chris' advice on this one--you are not going to win friends in this city, at least not at this point in time, by throwing around the word "privatization"

Here are my own thoughts on outsourcing government functions to the private sector:

Friday, August 03, 2007 10:40:00 AM  
Blogger C. Briem said...

a lot of that makes sense for sure.. I would point out the difference between perception and reality in some cases. You point out the inefficiEncy of the 67A, which by PAT's own metricts ranks as 7/10 in terms of cost per rider which makes it one of the more efficient (i.e. lowest costs per rider) in the system for weekday ridership. Thats not to say I disagree with the need to reform routes, but thats type of route inefficency is not what is driving PAT costs in the big picture.

as I have pointed out before.. the whole spine line concept was not just about putting in rapid transit between downtown and oakland but really using it to implement some part of trunk capacity between Downtown and what would have become a transit hub in or near Oakland.

Friday, August 03, 2007 10:41:00 AM  
Anonymous Anonymous said...

You are right about efficiency not driving the costs (nice pun, btw), but from what I have gleaned, it has to be the first step before any legacy cost bailout can be negotiated.

Friday, August 03, 2007 10:49:00 AM  
Blogger Schultz said...


What is up your a$$? Seriously. I do think privatization is the key, or, like MR suggested, allowing for competition. You are correct about me not winning over friends - but I am not running for public office. "Right sizing", "privatization", or other terms that have been thrown around indicate people will lose jobs.

I do not advocate privatization of everything - and certainly not all of public transit, but we do need a radical change in how things are done otherwise we're just going to keep facing increasing costs, reduced routes, and many more PO'd riders.

Sorry for being real about my feelings on this Chris, I'm not a politician so obviously it is difficult for me to feed BS to everyone.

Friday, August 03, 2007 11:03:00 AM  
Blogger Jason Wilburn said...

hey how about work rules? nobody in the previous posts talked about dealing with the labor situation.

Allowing competition would be great but unrealistic in the current political environment I would take a flexible workforce that allows for smaller buses on less utilized routes.

Friday, August 03, 2007 11:43:00 AM  
Blogger Adam said...

How about the fact that no transportation system in this country works without public subsidy? The only question is how much service are we willing to pay for? (I.e. how big a subsidy?)

There is no such thing as "private" transit. Let's say we "privatize" some or all of local public transit. How big a license fee do we charge to offset the cost to government of road maintenance, safety inspections, police protection, etc? Enough to recoup costs?

If the license fee actually does recoup costs, then it will likely be so high that the private companies can't make a profit and will leave the marketplace.

So if it doesn't recoup costs, then government is subsidizing the "private" transit companies.

Sure, allow competition. But let's not pretend that the private bus companies won't be operating without "subsidy."

Friday, August 03, 2007 12:01:00 PM  
Anonymous Anonymous said...

We wouldn't have to subsidize the Port Authority as much if there weren't so many maintenance people being so lazy and wasteful. I ride the T everyday and see them sitting in their trucks with the engines running reading the paper. Sometimes there are up to 4 people in the trucks just sitting there. I used to think that they had people standing by in case there is a problem during rush hour. However, I noticed a lot of them hiding the trucks behind buildings, or up the street away from the stations. That tells me that they know they are doing something that they shouldn't be doing. Plus, it isn't always during rush hour. I often see these guys hanging around late at night in their trucks(10 or 11pm, sometimes later) when I am out walking my dog. The news media would have a field day if they would go out and film these bums. Meanwhile, station platforms are crumbling, trash is piling up and the weeds are 4 feet high along the tracks. Most of us would be unemployed if we acted like this on the job.

Friday, August 03, 2007 12:51:00 PM  
Blogger C. Briem said...

well.. there you have it. lazy maintenance workers are the source of all problems... no. wait. It's the workers period. no. no. it's the long squiggly line. no no no it's.... these are soundbites not solutions.

all of these ideas still presume there is some way to magically save money without decreasing ridership. There is no doubt that lots of efficiencies could be put in place to save money: fewer routes, less maintenance, all sorts of things. but how many efficiencies could be put in place without decreasing ridership. Some I bet, but not anywhere near as people want to think there are out there. I bet there are some pretty clear patterns between the cost-efficiencies of public transit systems and the depth of utilization. It just isn't reasonable to expect PAT to find some efficiency that no other public transit system in the world has yet to find.

Saturday, August 04, 2007 10:28:00 PM  
Anonymous Anonymous said...


Efficiency gains are not the beginning and the end of saving the transit system, they are simply the beginning. There are significant savings to be had by making the system better utilized. And these savings are not "the fix" but necessary for the Port Authority to secure the political allies (including transit riders at the grass roots, who are currently pissed off at PAT, the Allegheny council, whose business members respect efficient operations, and Harrisburg, which would like to throw as little money at our transit as possible and would, thus, like to see a credible and substantial commitment to eliminating waste).

On ridership impact and addressing the first constituency--
Many of the inefficiencies in the system impact the rider directly: frequency of arrival, punctuality, and "bunching" are the primary. So, a switch to the trunk and feeder, spine and whatever, can have positive effects, from a user perspective, in all of these whole drastically reducing bus hours.

I have a rough version of a system wide transit overhaul (in map form) that was derived from pre-15%-cut PAT maps that shows no decrease in geographical coverage, increases in arrival frequencies on major routes, and a hypothesized 25% decrease in total bus-hours. They are the same types of changes that you see in the airline industry that have drastically cut their costs. Unlike planes, though, transferring between buses does not carry a huge user cost.

On transit systems of the world and what is actually possible. Check out Curitiba Brazil, it is amazing. It is about the same size as the Pittsburgh metro region and is, apparently, the most livable city in Brazil. 75% of all commutes in that city are made on public transit.

Monday, August 06, 2007 12:52:00 PM  
Blogger C. Briem said...

I've addressed Curitiba before as many readers here know so I will not repeat myself. The main innovation there that could be replicated is the use of busways which Pittsburgh is recognized as having implemented more than most other regions in the US. Beyond that the analogy really breaks up, its almost as if the argument is that Pittsburgh should try to emulate itself.

Monday, August 06, 2007 1:45:00 PM  
Anonymous Anonymous said...

Busways are not the main innovation Curitiba has to offer; long-term strategic, idealistic planning and civic culture that led to such an unbelievable system is what we could truly use.

Starting to aim for something like Curitiba is the only way to have something comparable in 30+ years. The first step of achieving efficient operations remains the same.

Monday, August 06, 2007 3:09:00 PM  

Post a Comment

<< Home