Thursday, July 06, 2006

Followup to "Transportation lessons from Houston"

After posting part 2 of "Transportation lessons from Houston" last week, I received the following email and information from David Wolff, Chairman of the Board for Metro, which raised some fair points about the impact of the Main St. rail line on overall transit ridership.


As you know, I receive, read and respect your communications.

I believe that you made a significant misstatement yesterday (“overall ridership has dropped precipitously”) concerning METRO’s overall ridership. Please see the attached information from Jim Archer which indicates that we are anticipating the highest ridership in our history this year.

There was a slight drop in FY 05 because we eliminated some severely underperforming routes – we were running about 11% of our service miles carrying about 1½% of our riders on those routes. I trust you would agree with that decision although it was not a politically popular one and was a decision that no recent METRO board had been willing to address.

As I write this it has occurred to me that perhaps the numbers at which you were looking included charter operations. I think we have been reducing these as they are not our real mission.

I would very much appreciate your getting the appropriate facts ( I am sure Jim Archer would be happy to talk further with you) and, if a correction is in order, I hope you will disseminate it.

Best regards,

Thank you,

David Wolff

The table below presents METRO’s fixed-route ridership since FY03.
  • Fixed-route bus ridership experienced a significant decrease in FY05 as a result of : 1) the bus / rail interface (May / June 2004); and 2) the discontinuation of nearly 40 weekday, Saturday and Sunday poorly performing bus routes (Fall 2004, Spring 2005).


(In 000s)





% Change

















Through the first eight months of FY06, METRO has achieved the largest fixed-route ridership increase in its history with 6 million more customer boardings than for the same eight month period in FY05. Fixed-route ridership for FY06 is projected to be the highest in METRO’s history based on this trend

Jim Archer

Service Evaluation

OK, now on to my response, which I'm basing on numbers in addition to those above, but which Blogger won't let me post here, because they're from various spreadsheets and pdfs. I'll be the first to say that teasing apart the numbers is a difficult and complex task, and not one I have an appetite for. With that caveat, here are my limited observations:
  • I agree completely that Metro needed to rationalize its route system to eliminate the least productive ones.
  • That said, it's hard to tell how much of the 2005 drop is from the new forced bus-rail transfers and how much is from dropping the unproductive routes. In Jim Archer's own words above, the total was a "significant decrease."
  • A big part of the complication is that Metro counts boardings, not trips. So if a route used to be on a single bus, that would be one boarding, but if it now connects to the rail, and then you transfer to another bus, that's three boardings where there was one before. So ridership can decrease while boardings increase (or at least don't drop as much).
  • Metro's own numbers show total transit passenger miles dropped from 620 million to 584 million from 2004 to 2005, a drop of 6%. Keep in mind this drop is during a period of substantially rising gas prices, which should be pushing transit passenger-miles up, not down.
  • The next cause-effect kink that muddies the numbers is Katrina, which brought tens of thousands of transit-dependent people to Houston in Sept-Oct 2005, which would seem to explain why Metro is on track for a record 2006. Monthly numbers I have confirm a large spike after September last year, which means that 6% drop in the previous point might have been worse without Katrina.
Summing up: using the phrase “overall ridership has dropped precipitously” was probably a bad choice of words on my part. But clearly there was a drop, and by Jim Archer's own admission, it was partially caused by the bus-rail interface changes (i.e. new forced transfers), which was the point I was trying to make. I will soften the language by changing the word "precipitously" to "noticeably" in the original post, but I think the overall point still holds up: when transit agencies invest large sums in capital-intensive rail, they feel the need to justify the investment with high ridership numbers, often accomplished by linking in bus routes that were previously direct or continuous, forcing riders to transfer and making their trips slower and less convenient, which in-turn reduces overall transit system ridership, even if the individual rail line looks good. This is one of my essential arguments against commuter rail. Conversely, in the core, a short-distance urban LRT/BRT network like Metro is building has other benefits (that I won't rehash here) that I think - in net - can outweigh the costs and additional inconvenience to some riders when the bus network gets tied in.

I hope that clarifies things. Have a great weekend everybody.


At 8:19 PM, July 06, 2006, Anonymous Anonymous said...

Buses run rings around rail!

At 8:46 AM, July 07, 2006, Blogger Richard Layman said...

Buses run rings around rail in everything but ridership.... and actually, buses don't run rings around rail at least compared to places with transit-complementary development, which is not how anyone would define Houston.

(I like Houston and the energy of its people--I spent a fair amount of time there from 1991-1995 involved in an entrepreneurial project.)

It'll take some time before land use and development patterns can catch up with the developments in transit.

However, being from Detroit originally, I think I can make this statement. My joke is that Detroit today is exactly what the auto industry intended to have happen to cities. Houston, being the poster child of the oil economy, could be a similar case.

The development pattern is based on continued supply of cheap gasoline, which allows people to make disconnected decisions about where to live, work, and what activities to pursue, as long as people value their time at almost next to nothing.

If cheap gasoline dries up, then this automobile-centric development paradigm is no longer tenable.

Because Houston's development pattern is polycentric, it's difficult to make transit work efficiently.

At 9:31 AM, July 07, 2006, Blogger Tory Gattis said...

Some good thoughts, Richard. Agreed that the land use patterns take time to develop around rail. I do believe car transportation will remain - if not cheap - at least affordable to the vast majority of Americans, no matter what direction personal vehicle and energy technology evolves.

Disagree on people valuing their time at next to nothing. People weigh off their time vs. the cost and size of their home and the quality of life and schools for their family. It's also worth noting that the longest commutes in the nation are in heavy transit cities like NYC and Chicago. Transit generally takes quite a bit more time to get just about anywhere vs. a car and a freeway - even at rush hour. I'm always stunned how long it takes to get even a couple miles in Manhattan by taxi or subway/walking.

Houston is very polycentric, which is rail transit unfriendly, but can work reasonably well with HOV express commuter buses.

At 10:32 AM, July 07, 2006, Anonymous Anonymous said...

Tory, I don't think you really answered David Wolff's numbers. He said Metro is on track for six million more boardings between this year and last year. The drop in boardings between 2004 and 2005, which you called "noticeable," was only one million. Also keep in mind that Metro started doing those bus-to-train transfers in 2004, so that factor should be constant betwen 2004 and 2005. And I don't think the Katrina refugees would alone make THAT much of a difference.

I read your commuter rail article, and I do not understand your opposition to it. You say it will cost "billions." No it won't - not to try it out on existing lines - and you know better. You are using rhetorical tactics to scare people away from something they haven't tried and don't understand. Yes, a person would have to walk or take a bus upon getting downtown in a commuter train, but isn't that also true of the HOV network? Do the HOV buses drop people off in front of their buildings?

You also misinform people by saying that the longest commute times in the nation are in heavy transit cities like New York and Chicago. First, let us keep in mind that New York has a metro area of over 20 million, and Chicago has one of over 9 million, vs. our 5 million. The fact that their commute times are even in the same ballpark as ours is, in my mind, a huge stat in FAVOR of mass transit. If you weigh in metro population, you will find that the cities with the worse commute times are places like Houston, Phoenix, and Dallas - places with weak mass transit systems.

Case in point - in Chicago they recently began the massive project of rebuilding the Dan Ryan Expressway, the city's 18-lane southern artery. Has this caused massive traffic headaches, the way freeway construction in Houston does? No, people are just switching to commuter rail - ridership has spiked on the southern route, and the diminished road traffic has been easily manageable. It's nice that Chicagoans have that option, isn't it?

I thought for a long time that this was a sensible, non-partisan blog, but it is starting to seem very lonestartimes-ish.

At 1:08 PM, July 07, 2006, Anonymous Anonymous said...

"Richard Layman said...
Buses run rings around rail in everything but ridership.... and actually, buses don't run rings around rail at least compared to places with transit-complementary development, which is not how anyone would define Houston."

That should read ...with TAXPAYER-SUBSIDIZED AND/OD DIRECTLY FUNDED transit-complementary development...

Further, urban rail in Houston is too expensive as well as unsafe, unreliable, and underutilized.

At 2:08 PM, July 07, 2006, Blogger Tory Gattis said...

Mike: Evidently, the bus-rail transfer switch was mid-2004, and then Katrina was Sept 2005, so we got 12 months of new data spread over two years, so comparing all of 2004 to all of 2005 is only partially helpful. And Katrina makes a *huge* difference, because those 100,000 people bused to Houston were the ones *without* cars to get them out of NOLA on their own. Considering that transit is already less than 5% of trips in a city of 2m, I would not be surprised if Katrina added 1/3 or more to the transit-dependent population.

Don't forget the problem with counting boardings. I think the passenger-miles number is more indicative.

The HOV buses do circulate around to different buildings/stops. And billions is not incorrect when you look at the number of commuter lines considered, freight re-routings (huge investments in themselves), new tracks alongside existing ones in many cases, plus stops and trains.

Car-based LA has faster average commutes than Chicago, with a substantially higher population. Houston metro is larger than transit-based Boston or DC, but has lower commute times.

Chicago construction: Houston could easily provide the same option with more express buses in the HOV lane, as they are doing with the current I10 construction.

At 4:54 PM, July 07, 2006, Anonymous Anonymous said...

METRO admits that the level of full-fare paying, bus-riding commuters has increased significantly. There are still declines on many of the routes system-wide, I speculate as a result of METRO's pro-urban rail bias that negatively impacts the poor, minority, elderly asn handicapped bus transit dependent riders throughout the service area.

At 11:18 PM, July 07, 2006, Anonymous Anonymous said...


These commute times that you speak of... are they for car, or train commuters, or an average of the two? When I lived in Chicago and took the train into the city, I don't recall it taking very long. Of course the car commute was horrible, and everyone knew it - that is why we all preferred the train.

I read that it would cost less than $200 million to run commuter trains along Hempstead Hwy. To me, this doesn't seem like too much to try a new transportation option. Are you against even trying this option?

At 9:13 AM, July 08, 2006, Blogger Tory Gattis said...

Average commute times across all modes reported to the Census American Community Survey.

I think the $200m is a lowball (like not including costs to accomodate freight rerouting), but I'm opposed even at that price, because, regardless, once the money is sunk, Metro must do everything possible to make it work, and that means canceling any potential competing HOV express bus service and forcing those people onto the train, even if it substantially lengthens their commutes. The "coolness" factor will wear off quickly, and then people will be stuck with it and frustrated, making it harder for core employers to attract and retain employees, and eventually driving them to move to a nice, cheap suburban office park.

At 3:59 PM, July 11, 2006, Anonymous Anonymous said...

I don't think it's possible to make a blanket statement on whether commuter rail (or any other transportation technology) is worthwhile. It depends on the corridor. And two important considerations are:

(1) Is spare capacity available on an existing rail line? If there is, commuter rail can be very inexpensive compared to other options. If there isn't, it can be very expensive.

(2) Is there an existing high-quality transit service in the corridor? If the current transit service is buses stuck in congestion on freeways or surface streets, commuter rail can increase transit ridership. If the current transit service is fast buses on dedicated lanes, most of the people who will ride transit already are.

In most corridors with sucessful commuter rail startups, the answer to both questions is "No." But in Houston most of our rail lines are already congested witrh freight and we've already made a large investment in HOV lanes and park-and-rides for buses to provide high-quality commuter transit service in most corridors.

At 4:59 PM, July 11, 2006, Anonymous Anonymous said...


I meant: "The answer to the first question is "yes" and the answer to the second question is "no."


Post a Comment

<< Home