Despite SEPTA’s budget cliff, Roosevelt Boulevard subway backers seek path to Northeast Philly extension

Unless Pennsylvania lawmakers hurry up and pass a budget that includes a solution for transportation funding, SEPTA will begin its first round of service cuts in late August. And if a deal to help stabilize the agency's operating budget isn't in place by the end of the year, the Philadelphia region is staring down a public transit system that will operate with an unfathomable 45% reduction in service.

In times like these, it may seem naive to even consider the multibillion dollar vision to extend SEPTA's subway system into Northeast Philadelphia. Legitimate momentum for the idea last surfaced in the early 2000s before fading from relevance, but there has been a serious revival of interest and advocacy in recent years. The city and PennDOT have been evaluating various options to rectify the transportation nightmare of Roosevelt Boulevard, including cost and ridership assessments for building out the Northeast Philly subway.

MORE: SEPTA says bolstering enforcement efforts have helped reduce serious crime

The problem is that in addition to SEPTA being starved today, the proposed timeline for a Roosevelt Boulevard subway would not have construction begin until 2040.

"Predicting the future is very hard, especially in the timeframes we're talking about," said Eric Goldwyn, a researcher for New York University's Transit Costs Project. "Once you get to over five years, it's like, who the hell knows? You're trying to put guardrails and parameters around a very unknown future. Maybe COVID happens, or some other thing happens."

Goldwyn's research team analyzes the cost drivers of transit projects in cities worldwide, publishing in-depth reports on how to complete them with realistic budgets and forward-thinking land use. They spent much of this year examining the Roosevelt Boulevard subway proposal on a granular level to compare findings with projections PennDOT released late last year.

"Transit is full of zombie projects that just sort of stick around," Goldwyn said. "From a ridership standpoint, this is viable from our perspective."

TCP's new report on the Roosevelt Boulevard subway has a few key takeaways, but the biggest is that the project should be smaller than the proposed 12-station concept — shown in the map below — that extends to the Far Northeast in Neshaminy. Instead, the TCP report suggests building the subway out to the proposed Rhawn Station.

Provided Image/Transit Costs Project

The map above shows the proposed framework for an extension of SEPTA's subway system into Northeast Philadelphia along Roosevelt Boulevard. The long-discussed project could include connections to the new line from both the Broad Street and Market-Frankford lines.

TCP's analysis found that about 90% of projected new ridership for the 12-station subway plan — roughly 51,000 people daily — would come from the first seven additional stations. Beyond that point, the cost per rider becomes less favorable in places where the sites of proposed stations are not as accessible on foot.

Jay Arzu, a Ph.D. candidate at the University of Pennsylvania and one of the most vocal advocates for the Roosevelt Boulevard subway, praised TCP's work and said cutting the project in half would make it much more affordable in a shorter timeframe. He lamented that SEPTA's operational challenges create a sense of defeatism toward long-term infrastructure investments.

"That's the problem with our region," Arzu said. "Whenever the going gets tough, we want to stop planning. That's not the way competitive regions move forward."

At what cost would extending the subway be justified?

Figuring out how many people would use the new subway line — and how many more might ride it if the project spurs new development — is critical to determining an acceptable cost. Virtually all large transit projects rely on federal funding for up to 50% of their costs, Goldwyn said, and a sweet spot in the United States in recent years has been projects that land between $50,000 and $70,000 per rider. Examples include Boston's Green Line Extension and Seattle's East Link.

PennDOT's current estimates for the Roosevelt Boulevard subway put the full 12-station project anywhere from $11 billion to more than $16 billion, which includes all planning and development-related costs. It also includes a 40% contingency for all kinds of unexpected costs.

On the low end, TCP's estimates put the cost per rider at about $148,000. Goldwyn said shortening the length of the project would give it a chance to fall in the range of $3 billion — especially if cost-saving engineering methods are used — but only if construction begins well before 2040 to stay ahead of inflation.

PennDOT's projections for the full 12-station subway proposal are a bit more optimistic than TCP's — about 62,000 new daily riders — but the agency similarly concludes those gains would fall off after a certain point along the route. The biggest discrepancy is that PennDOT's study estimates about 10,000 daily riders at Welsh-Grant Station, while TCP projects only about 800.

Goldwyn chalked this up to his team using a machine-learning model that takes a different approach to extrapolating ridership figures from SEPTA's existing subway stations. He's not sure which estimate is most accurate. Arzu cautioned that Rhawn Station poses problems as a terminus because it lacks surrounding space for a subway maintenance yard and transit center, which would be necessary for connecting buses to the new subway route.

Although PennDOT envisions having standardized stations along the potential route, the agency also discussed tailoring them to their surrounding communities during meetings last year.

"At a station like Welsh-Grant, there could be a parking deck," Arzu said. "But at Cottman, you would want to see developers build (housing) adjacent to the station. Each one is going to be different."

PennDOT to narrow down options this fall

Since PennDOT already is evaluating spending billions to improve Roosevelt Boulevard — with alternatives including a a light rail line and rapid bus service — Goldwyn said it would be "irresponsible" not to give the subway serious consideration for its overall cost and ridership benefits.

Still, he acknowledged the timing of TCP's report coming out is "terrible in a lot of ways" because of SEPTA's immediate crisis and what it means for the region.

"It is one of the most beleaguered agencies in the country, but they do a great job," he said. "It's just sort of like no one is trying to help them. Figuring out the operating challenges should be the primary focus of SEPTA and the state legislature. But in terms of growing your city and increasing your tax base — if you want to attract people from somewhere else — investing in infrastructure is a good way to do that."

Based on last year's analysis and community meetings, PennDOT is expected in September to narrow down its list of preferred alternatives for improving Roosevelt Boulevard. Arzu said he believes PennDOT will combine various features of the six initial alternatives and boil them down into three options that will be discussed during a fresh round of community meetings.

PennDOT did not immediately respond to a request for comment about the agency's timeline.

If one of the subway options is chosen, Arzu believes the required environmental assessment, engineering and design work could be completed by 2029. He called the project's ridership numbers a "shoo-in" for federal funding once it's shovel-ready, and he said elected leaders from both parties would be wise to embrace a transit project that creates jobs and clears congestion from major roadways.

Arzu's critics have told him it's a ridiculous time to invest in a Roosevelt Boulevard subway when SEPTA can barely stay afloat. Arzu said he's not downplaying what SEPTA is up against in Harrisburg right now or the broader climate of federal funding getting chopped left and right. He just thinks the message needs to be hammered home to policymakers that public transportation is an essential driver of economic growth, one worthy of bold investments.

"A project like this should be able to move in three to four years," he said.