TRANSIT SPIN:Flacks con

The trick to having your project declared a “success” in the media is to do deep cuts in “projections” or “forecasts” in the run-up to opening day. Call ‘em “conservative”, presently an in-word especially among liberals. That’s the secret of instant “good media” as demonstrated in several rail projects in the west. In Los Angeles for example the Green line opened with forecasts of 10,000 weekday riders and then when they got 15,000, they were all set up to declare the project a “great success.” No matter that a single lane of a freeway would carry the same number of people at a fraction the cost. No matter that the project had been “sold” to the public and to officials some years earlier when the project was being approved by the legislature with a number of 100,000 weekday riders. Short memories they have, those guys. Besides they wouldn’t want to be seen as party poopers?

Portland OR did the same thing. They lowered forecasts by two-thirds before opening as compared with projections that had been used back at decision time with legislators. They still call it a success even though ridership (27k trips/day) is a third short of the decisiontime forecasts (42k trips/d), because of course they are above the opening time projections.

But can we believe their numbers even?

Portland’s Tri-Met transit agency has just been caught in the classic embezzler’s practice of keeping a second set of books on its business. Jonathan Richmond of Harvard’s Kennedy School reveals in a recent paper “New Rail Transit Investments” for the Taubman Center for State and Local Government that Portland’s Tri-Met has been feeding the Fed Transit Admin, and hence the public, with one set of data on the ridership of its MAX light rail while keeping a separate internal set of ridership numbers. For public consumption it presents a base year number 6% below the presumably accurate number that it keeps internally and by now it submits to the feds a number that is 5% higher than the actual numbers as recorded internally. So it claims a 37% increase while its internal data show a 21% increase over 15 years. Richmond devotes several pages to the fraud which he exposes after extensive communciation back and forth with Tri-Met. The agency is reduced in the end to saying Richmond is biased because he is critical while “Portland’s accomplishments” have been hailed by the NEW YORK TIMES and other national publications. And note the effort to invoke communal responsibility for a folly heavily promoted by the transit monopoly, using gimcrack projections and only narrowly approved by voters. Their reasoning: “We gulled the NEW YORK TIMES so it must be OK.”

Portland OR is frequently cited outside it as some kind of model for enlightened planning, but inside it the establishment is widely regarded as wacky, and worse. Voters there have rejected moves to extend the light rail even though the feds will pay most of the capital cost.

More transfers more trips

Deliberate fraud by transit hucksters is a crude crime but perhaps more effective in legitimizing the ripoff of taxpayers for transit is confusion over what statistics measure and mean. Take the issue of measuring transit’s performance by citing unlinked trip numbers by looking at a local example in the Washington area. Most express commuter bus service from southern Fairfax Co VA used to travel I-95, and then inside the Beltway it went on the Shirley Highway (I-395) HOV lanes for direct trips to for the Pentagon and Washington DC. But with the opening of Blue line rail service from Springfield just outside the Beltway mid-97 many of those buses became feeders to rail. They now stop half way dumping their passengers off at the Springfield metro stop. And guess what? According to the statistics there was quite a healthy surge in transit usage, because transit usage in the US is commonly measured solely by boardings or ‘unlinked’ trips. Suddenly the same Mavis Maviotes going from her home in Lorton VA to her job at the Pentagon who previously was 2 trips per day becomes 4 trips/ day. Contrary to the impression of growing popular patronage statistics Maviotes now pays more, has a longer commute and costs the taxpayer more in subsidies, but her doubled tripmaking enables them to say that transit is a great success.

Such deceit is the staple of the propaganda by which American voters are persuaded to support one disastrous new rail system after another. Salt Lake City, Minneapolis, Hudson Co New Jersey, Orange County CA, Minnneapolis MN, one after another the snakeoil salesmen of commuter rail sell their fraudulent white elephants. In almost every case the new rail lines substitute for more economical and flexible bus, increase steeply the overall costs of transit and usually reduce its convenience by multiplying the number of transfers required.

Volpe Center

Don Pickerell of the Volpe Center did pioneering work in 1990 documenting the systematic exaggeration of ridership and gross underestimation of cost that rail sponsors used to get their projects launched. A light rail for Buffalo for example was forecast in 1971 to have a 1995 weekday ridership of 160k. In fact it has 26k, less than a quarter of that forecast. Vorhees one of the most creative of the consultants involved in this scandalous business managed to get almost everything badly wrong — the speed of the trolleys, relative costs of bus and trolleys and the capital cost — which they put at $241m and which turned out to be $536m. The Los Angeles blue line which consultants said would cover 67% of its operating costs out of farebox revenue in fact is covering 13% of its operating costs. Miami’s rail line was sold on forecasts of 239k riders weekdays and they get 47k. Even ones that seem rather close to projection often turn out to be contrived. In Sacramento where ridership on a new trolley is 28k/day compared to 50k forecast, about half the trips turn out to be short inner city trips in a “free” trips zone downtown that was never planned for. If you can’t sell it, give it away!

Richmond paper from Harvard

Richmond’s paper takes the analysis of this late 20th century American urban transport fiasco an important step further. He examines in some detail the promise and the performance of a dozen new light rail projects and a couple of busways. The general pattern of light rail he says has been to eliminate direct bus routes and institute a bus/car feeder/transfer/rail line-haul arrangement. In justifying the enormous capital cost of rail (up to $100k/daily rider) it is claimed the productivity of rail is higher.

Driver productivity for sure, because of the higher ratio of passengers to drivers, but Richmond finds that in general the greater non-driver costs of rail offset driver economies.

Moreover the railophiles will make apples and oranges comparisons. Routes are cherrypicked for rail in the sense that the most densely trafficked bus corridors are chosen for conversion to rail, and it is invalid to compare productivity of rail corridors with bus corridors when bus is left with the less trafficked. Moreover the comparison of a line-haul rail with bus invariably attributes to bus the very high cost bus feeder routes. Since bus feeders are only needed by rail, their costs should be attributed to rail — not to bus, Richmond says, a cost allocation that changes the picture completely.

Richmond finds “busway costs come in well below light rail” and says that they provide “greater value overall.” The Ottawa transit agency modelled the use of feeders and line-haul for its busway and compared that with direct service — buses which did local pickup and setdown at the ends and made use of the line-haul busway in between. It also studied ‘interlining’ in which a bus finishing a journey heads to another nearby line rather than returning on the same line. The results were strikingly in favor of direct service option with many fewer transfers and in favor of interlining. But light rail has no such flexibility. It has to be arranged for simple back and forth line haul trips. All this makes for high operating costs.

Unfortunately Richmond does not attempt to combine operating and capital to produce overall costs. But since bus is more productive in both capital and operating costs on comparable routes, the verdict is clear — urban rail is very wasteful vs bus.

On the matter of rail’s promise to alleviate highway congestion or improve air quality Richmond writes that “in no case” has this occured. He concludes that the best strategy for increasing transit use will generally be to build bus service, and to keep bus fares down. He is only impressed by one new light rail (San Diego’s initial Blue line). He is full of praise for Houston’s management of busways, in part because they are used very productively by accomodating HOV as well as buses. (Jonathan_Richmond@harvard.edu)