Thirty-five years ago, San Diego kicked off the light-rail fad when it opened the San Diego Trolley, the nation’s first modern light-rail line. The city paid $18.1 million for the right of way and $87.5 million to build 13.5 miles of rail line. Two years later, they double-tracked the line bringing the total cost, including right of way, to $137.35 million, or just slightly more than $10 million a mile. In today’s dollars, that would be $23 million a mile.
Now San Diego is planning a new light-rail line that will cost a mere $2.17 billion for 10.9 miles of line, or slightly less than $200 million a mile–and that’s only if there are no cost overruns. That’s more than eight times the cost per mile of the first line. Ridership is likely to be no greater and probably less than the first line. Despite the high cost, the Federal Transit Administration has agreed to fund half the cost.
What is it about transit planners that they see nothing wrong with this kind of cost escalation? Transit advocates claim that transportation spending has multiplier effects that generate net benefits for the economy. But considerable academic research suggests that the government multiple is negative (or, depending on your formula, less than one), meaning every dollar of government spending translates to less than a dollar of additional gross national product (or national wealth). The truth about the multiplier effect probably depends on what the government spends its money on, but billions spent on low-capacity rail lines almost certainly have negative multipliers.
Seattle’s regional transit agency, Sound Transit, wants voters to approve a tax increase so it can spend another $54 billion on new light-rail lines. The agency’s first light-rail line went 86 percent over its original projections, but the agency assures the public that it has realized that voters are so innumerate that it no longer needs to low-ball the cost estimates in order to get tax increases approved.
To promote its plan, the agency has hired Peter “Paint Is Cheaper Than Rails” Rogoff to run the agency and get federal grants. Rogoff argued in 2010 that buses can attract as many riders as trains, and that “Bus Rapid Transit is a fine fit for a lot more communities than are seriously considering it.” Of course, he must believe that rail makes more sense than buses for Seattle, or he wouldn’t have taken this $298,000 per year job (a $118,000 increase over his previous job), right?
Seattle’s first light-rail line cost $3.1 billion in 1995 dollars, or $4.8 billion in today’s dollars for about 20 miles, for an average cost of $240 million a mile. According to the Census Bureau’s American Community Survey, out of nearly 1.6 million commuters, a respectable 160,000 took the bus to work in the Seattle urban area in 2014 but fewer than 3,000 took light rail while another 7,500 took commuter rail or streetcars to work. It’s possible that some survey respondents were confused and marked streetcar or commuter rail when they meant light rail, but it is still an insignificant number.
Judge Richard Leon apparently invalidated Maryland’s Purple Line project just in time to prevent the Federal Transit Administration from giving the Maryland Transit Authority nearly a billion dollars for construction. Naturally, rail supporters are outraged by his decision.
The Washingtonian, for example, calls the decision “ridiculous” because it was based on declining Metro rail ridership and “the Purple Line will not be part of Metro.” But the article admits that 27 percent of the projected Purple Line riders will be transfers to or from Metro, so if Metro ridership declines, the Purple Line’s will as well.
The National Environmental Policy Act is supposed to be a procedural law, the Washingtonian complains, so why is Judge Leon allowing substantive issues such as ridership influence his decision? The author of this piece is obviously not an attorney and probably didn’t even read Judge Leon’s decision, or he would understand that getting the numbers right, as the judge demands, is procedural. He might also be able to read between the lines of the decision and see that Leon realizes this project is a turkey, and is using ridership as just the most obvious reason to overturn the decision to build the low-capacity rail line.
A federal judge has vacated the decision to build the Purple light-rail line in suburban Washington, DC, effectively delaying and possibly halting the project. Judge Richard Leon’s decision said that “recent revelations regarding the Washington Metropolitan Area Transit Authority’s ridership and safety concerns” had persuaded him that projected ridership numbers for the Purple Line were overly optimistic.
Some Purple Line stations were planned next to Metro Rail stations and projections for light-rail ridership depended partly on the continued growth of Metro Rail ridership. But, largely due to safety issues, Metro Rail ridership has declined every year since 2009, noted Judge Leon, and this meant that the Purple Line ridership numbers, which were calculated in 2009, would probably be lower as well.
“WMATA and the FTA’s cavalier attitude toward these recent developments raises troubling concerns about their competence as stewards of nearly a billion dollars of the federal taxpayers’ funds,” wrote the judge. The Purple Line would be built and operated by the Maryland Transit Authority, not WMATA, but, the Judge Leon noted, this doesn’t mean that declining Metro ridership wouldn’t have an impact on Purple Line ridership.
The Antiplanner will be in Littleton, Colorado tonight talking about housing issues. The event is open to the public and starts at 7 pm at the South Fellowship Church, 6560 South Broadway. If you are in the Denver area, I hope to see you there.
In the meantime, interesting news from Sacramento: the regional transit district is considering shutting down one of its light-rail lines for lack of ridership. As the Antiplanner noted two months ago, the agency has lost more than 26 percent of its transit riders in the past six years and has raised fares by 10 percent to make up for the lost revenue.
The light-rail line that it is considering shutting down is only 1.1 miles long–so it is more like a streetcar line–and it attracts just 400 riders per day. Despite this poor record, Sacramento still wants to build a 3.3-mile streetcar line.
“We’re driving less than other metros,” says Portland Metro, “and we’re driving less than 20 years ago.” These are the kinds of lies that are so typical of Portland planners: easy to check, but since few will bother, they get away with it. First, although Portlanders do drive a little less than residents of other large urban areas, they also drove a little less than residents of those same urban areas 20 years ago, so that’s no change.
More important, Portlanders are in fact driving more than they were 20 years ago. Metro’s source for its data is the Texas Transportation Institute (TTI) urban mobility study. But those data only count miles of driving on freeways and other arterials. Because Portland hasn’t built any new freeways or arterials in 20 years, people are forced to drive more on collectors and other roads. When all driving is counted, according to the Federal Highway Administration (which is the source of TTI’s data), Portlanders drove an average of 22.9 miles per person per day in 2014, up 13 percent from 20.2 miles per person per day in 1994.
Metro uses lies like these to implicitly claim that all of its spending on light rail is worthwhile. In fact, light rail is just not very important to Portland travel. As the Antiplanner showed last week, transit carries just 2.6 percent of motorized travel in the Portland urban areas, and since light rail is less than 40 percent of transit, that means light rail is less than 1 percent of all motorized travel.
In a move that surprised no one, the staff of TriMet, Portland’s transit agency, wants to build light rail instead of bus-rapid transit between Portland and Sherwood. Since the Obama administration no longer requires transit agencies to do a rigorous alternatives analysis, this decision was based on subjective criteria and erroneous assumptions, yet will probably not be challenged by either TriMet’s board or the federal government that will have to pay for most of the line.
TriMet’s last light-rail line cost about $168 million per mile. This proposal is for an 11.5-mile line that will cost at least $2 billion, or $174 million per mile. Of course, that cost is likely to go up. By comparison, Portland’s first light-rail line cost only about $28 million per mile in today’s dollars.
A state auditor says TriMet, Portland’s transit agency, is falling behind on light-rail maintenance. TriMet’s general manager says that the agency’s pension and health-care obligations are so great that it will have to cut all transit service by 70 percent by 2025 to meet those obligations. So naturally, it makes perfect sense to talk about spending $2 billion that the agency doesn’t have on another low-capacity rail line.
The Los Angeles Metropolitan Transportation Authority (Metro) wants voters to give it $120 billion over the next forty years so it can build more rail projects that are already obsolete. Among other projects, it proposes to build a nine-mile light-rail tunnel between west LA and the San Fernando Valley that it estimates will cost at least $8.5 billion, and probably much more. That’s a billion dollars a mile, which is neither a misprint nor an April Fool’s joke.
The plan, which will probably be on the November ballot, includes some new roads as well as trains. But Metro proposes to spend twice as much on new transit construction as on new road construction, plus lots more on transit operations. As little as 19 percent of the funds would be spent on highway projects.
In 2008, Metro persuaded voters to dedicate a half-cent sales tax to transit for 30 years, which is estimated to bring in $34 billion. Now it wants to double that tax and extend it to 2057, which is estimated to bring in $120 billion on top of the $34 billion it is already getting.
Sacramento’s Regional Transit District (RT for short) is facing an existential crisis. The region’s transit ridership fell by 22 percent between 2009 and 2014, and preliminary information indicates another 6 to 7 percent decline is likely in 2015. The agency’s January, 2016 performance report shows a 9 percent decline from January 2015.
A light-rail train trundles its way through downtown Sacramento. Flickr photo by PaulKimo9.
Some of this downward spiral is due to low gas prices, but much of it is due to an 18 percent reduction in bus service and a 7 percent reduction in light-rail service between 2009 and 2014. Declining tax revenues after the 2008 financial crisis forced these service cutbacks. In turn, reduced ridership means reduced fare revenues, and RT has responded by raising fares, which is not likely to do ridership any good. RT is also thinking about asking voters for a tax increase, but with just 2.7 percent of the area’s commuters taking transit to work, support for the transit system may be slim.
Portland’s first light-rail line turns 30 years old this year, which is about the expected lifespan of a rail line. Not by coincidence, the system was highly unreliable last year, being “plagued with delays and disruptions” and having terrible on-time performance.
The line between Portland and Gresham originally cost more than $200 million to build, which in today’s dollars is around twice that. It is likely it will cost roughly that amount of money to restore it to like-new condition.
But Portland has a choice. Instead of sinking a bunch of money into an already-obsolete transit system, it could scrap it and replace it with buses. Before building the rail line, the parallel freeway had HOV lanes; restoring those lanes (or turning them to HOT lanes) would give the buses an uncontested route to fallow. We know that the buses would be faster than the rail, because the rail line was slower than the buses it replaced.