Phoenix Spent $1.4 Billion To Cannibalize Buses

I have written many times about my problems with Phoenix light rail -- examples are here and here.  We paid $1.4 billion in initial capital costs, plus tens of millions a year in operating losses that must be subsidized by taxpayers, for a line that carries a tiny tiny percentage of Phoenix commuters.  Capital costs equate to something like $75,000 per daily round trip rider  -- If we had simply bought every daily rider a Prius, we would have save a billion dollars.

But, as with most things the government does, it is worse than I thought.  Over the last several years, I have been treating these daily light rail riders as if they are incremental users of the area's transit system.  In fact, they are not, by Valley Metro's (our regional transit authority) own numbers.  Here is the key chart, from their web site.

ridership report chart graphic

Compare 2009 to 2012.  Between those years, light rail ridership increased by just a hair under 8 million.  In the same time period, bus ridership fell by just a hair over 8 million.  So all new light rail ridership is just cannibalizing buses.  We have spent $1.4 billion dollars to shift people to a far more expensive transit platform, which does not offer any faster service along its route (the light rail has to fight through traffic lights on the surface streets same as buses).

This is a pattern seen in most cities that adopt light rail.  Over time, total ridership is flat or falls despite rising rail ridership, because rail is so expensive that it's operation forces transit authorities to cut back on bus service to balance their budgets.  Since the cost per rider is so much higher for light rail than buses, a dollar shifted from buses to light rail results in a net reduction in ridership.

Postscript:  Looking at the chart, light rail has achieved something that Valley Metro has not seen in decades -- a three year period with a decline in total ridership.  Sure, I know there was a recession, but going into the recession the Valley Metro folks were arguing that a poor economy and rising gas prices should boost their ridership.




  1. mahtso:

    One quibble: I believe the traffic lights are timed or rigged to give some priority to the (Phoenix) light rail. I don't know if that makes the rail faster than the bus. But I believe it adds to the congestion at certain intersections. (Relieving traffic congestion is one of the purported benefits of mass transit.)

  2. marque2:

    It adds to congestion because they had to remove two traffic lanes, and the number of people who travel by rail does not equal the number of people displaced from those lanes. Warren wrote about that last year some time.

  3. Anonymous Mike:

    The project was sold to locals as a way of increasing overall mass transit ridership. However given that the planned tracks were basically serving lines already serviced by busses (the initial tracks roughly followed the old "Red" line), that argument was a nonstarter because no one could explain why people who wouldn't take the bus would now take a slightly faster train that ran to very limited areas. Well actually there was an explanation and that was called snobbery

    Actually that was how the whole light rail system was sold - by snobbery. Snobbery in that it would make us a world class city, snobbery in that it was so much better to ride a souped-up trolley rather than a bus, snobbery and fear that we would be left behind North Haverbrook if we didn't spend gobs of money on it

    Actually the best reason ever explained to me - besides some coming from those who drank deeply of Richard Florida's Kool-Aid, was that the reason to build it was that it was the only thing the thing the feds were funding.

    It's too bad the light rail doesn't run past the Ed Pastor Transit Center, that would be only fitting

  4. perlhaqr:

    I think that either you have misinterpreted the chart or I'm misreading it.

    But it seems like you'd have to compare the green line from 2008 to the purple lines from 2009 - 2012 to get the accurate picture.

    i.e.: in 2008, ~62M rides occurred on buses. In 2009, ~65.6M bus rides and ~5.6M rail rides occurred, totalling ~71M rides. In 2010, ~55.6M bus rides, ~12M rail rides, ~67.7M total rides. Down from the previous year, but still a higher aggregate total than 2008.

    Not that I think it was worth $1.4B to build a light rail system, the per rider subsidy is still clearly insane, and the lack of flexibility is idiotic. But I'm not sure this chart shows what you're purporting it to show.

  5. herdgadfly:

    The gray line is the total of the green line and the blue line in each year. Not very complicated. Warren's point is that 2009 and 2012 had approximately the same number of total riders and the gain in light rail came at the expense of bus rides. All that massive investment in light rail and no appreciable increase in additional public transportation riders. Why spend the money in a country that is attached at the hip to the automobile?

  6. Hasdrubal:

    If you run a simple linear regression, the slope of the trend line from 1997-2008 is 2.778 while the slope of the entire range is 2.7112. That means, over the entire 16 years they averaged adding 2.7 million riders a year. Adding light rail did not change this trend significantly, though it did pull it down just a little bit.

    So Warren's point stands: Even though there are more total transit riders now, there are pretty much as many riders as you would expect to have if you just had busses, based on prior trends.

    (You can do the math in excel if you like, just enter the dates in one column, the ridership numbers in the next, [I divided everything by 1 million so 2012 had 71.043949 in my chart, it makes the final numbers a little cleaner, just remember to multiply your beta by 1 million when interpreting the results] highlight them, click insert > graph, make a scatterplot, then once the graph is created go to chart > add trendline, choose linear and under options choose "Display equation on chart." I imagine you can do the same in Google Docs or Open Office for free.)

  7. LoneSnark:

    As it seems the light rail runs along an existing road, there was probably a bus line running down that road. As such, all the riders of the new light rail line were already riding that bus line.

  8. John Washington:

    Nice work!

    Even former Scottsdale mayor Mary Manross said it: Light rail is NOT about transportation. It's purpose is facilitating development. As in "transit-oriented" high density development. As it making money for property owners and developers along the light rail routes.

    To the other comments regarding congestion, Valley Metro's own estimates published during the sales campaign showed that vehicle emissions near the rail lines were going to increase. This is due to increased automobile wait times at the intersections, particularly for left-turning vehicles.

  9. John Washington:

    That's an excellent point, and one that I tried to raise many times during consideration of a light rail line for Scottsdale. Unfortunately, the shameless light rail promoters were successful at ignoring it.

  10. Matthew Slyfield:

    "It's purpose is facilitating development. As in "transit-oriented" high
    density development. As in making money for property owners and
    developers along the light rail routes."

    If it actually accomplished any of that they should see higher levels of ridership not cannibalized from bus transit.

  11. irandom:

    We need Sarbanes-Oxley for public agency reports. Lock-up a few of them and things might change. Especially those stupid this will cause x economic development.

  12. mahtso:

    That the removal of lanes would increase congestion is almost a no-brainer. My point is that the congestion is increased at intersections where the same number of lanes exist. More clearly (perhaps): the cross streets also have increased congestion.

  13. skhpcola:

    Richard Florida is a world-class ephtard. He favors big cities and central planning, like every wannabe thug dictator in history...better to control the filthy masses. That he has any influence is a pitiful indicator of the tyrannical desires of the people that we elect to run things. In a perfect world, people like him would be poor and die in railcar accidents while they try to get to the next dying city. Filthy bastards.

  14. marque2:

    Not necessarily. They pretended that the lanes removed would not add to congestion because so many more folks would be taking the train. If they take away 6000 cars per hour and 7000 more folks take the train then you have break even. But it seems like they didn't get the extra ridership

  15. Craig Marks:

    Fares have also gone up dramatically. One-way fares were $1.25 from 1994 until 2009, when they went up to $1.75. They went up again this year to $2.00. That's a 60% increase in just four years! Of course this hurts the poorest, with no other transportation options, the most.

  16. Mr Lizard:

    They are trying to pass this nonsense off in Tampa bay. We beat them once at the polls, hopefully hillsborough and pinellas counties will vote it down again. It seems like light rail is just a plan B option for certain special interests in a down economy.

  17. Harry:

    I am not sure when I learned the first story about defaulted NYNH&H bonds. I was around for the assumption of northeast passenger rail transportation by AMTRAC maybe ten years later.

    Now, if rail travel in the Boston to Washington corridor cannot be a very profitable business in a free world, and I know about a lot of the details to the contrary, then how can light rail travel in Phoenix ever be anything be other than a Grand Canyon money pit?

    Engineering question: would a trillion one dollar bills, crumpled in the same way as they are given back in change from Starbucks, fill up the Grand Canyon, or just a portion of the North End? Could it obliterate Los Angeles? Would it take five trillion?

  18. perlhaqr:

    Yeah, I got that part, I was looking at the increase in ridership from 2007 - 2008 - 2009. But Hasdrubal's point about how that's just a part of the overall linear trend of ridership increase fixes my confusion. It's not that there was no more ridership after the light rail came in, it's that there was no more ridership than there would have been otherwise, even if they hadn't built it.

    And, yeah. Waste of bloody money. But hey, light rail is sexy, but ew, busses are so ghetto. Seattle is about to do the same damned thing to themselves, it looks like.

  19. perlhaqr:

    Ahhhhhh, ok. I get it now. Thanks for the explanation! 🙂

  20. Alik:

    Economic profit prevent this most sustainable transportation model to be operated in Greater Phoenix area. Despite every one knows the petroleum projection only for 25 year, but people naively believe scientist will find other alternative. However, first and second energy law tell us: Even scientist find other alternative energy, but still need some other non-renewable resource to create the alternative energy. So if human find other alternative, then we are depended on another non-renewable resources. Even the solar energy need very expensive operation to produce silica ingot and the process is very harmful for environment. Every one kilogram silica production needs 80 kilograms Chlorine or Florine. The minimum dosage of the important carcinogen Helotrimethane which important component are Chlorine and Florine is only 0.8 micrograms/liter.

  21. Sweet63:

    It always comes up among the progressives in my town. LR is a cure all, a magic bullet for whatever ails us. Fortunately I don't think we have the population to warrant serious consideration. But LR is always something that *should be done* in a Decent Society.