Two-thirds of Metrorail Trips Cross a Boundary

February 3rd, 2015

A large majority of trips on Metrorail cross jurisdictional boundaries, illustrating that Metro is indeed a regional service.

We’ve mentioned before how the station improvements in Metro 2025 will benefit riders from all jurisdictions.  In fact, Dupont Circle is the only station identified in Metro 2025 with a majority of users living in DC.   We thought we’d take another look at ridership that crosses jurisdictional boundaries.  The table below illustrates the percent of trips, by jurisdiction of origin, that cross into another jurisdiction on Metrorail, sliced by Weekday AM Peak, Weekday PM Peak and Weekend.  Data is from October 2014 and includes the new Silver Line stations.

A few things pop out:

  • Regardless of period or day type, over 65% of Metrorail trips cross from one jurisdiction to another.
  • A vast majority of non-DC AM Peak trips end in another jurisdiction, from 85% in Montgomery County to 97% in Fairfax County.
  • The number of trips leaving DC in the AM Peak — 28% — is actually quite high and illustrates a strong (and growing) reverse commute market.
  • The 60% of trips leaving DC in the PM peak is understandable, as workers in the region’s core return to their home jurisdictions after work.
  • This means 40% of trips originating in the District in the PM Peak are destined for other District stations.
  • The weekend numbers are interesting, with nearly half of the trips originating in the District destined for other jurisdictions.

Metrorail customers have many opportunities to stay within their state on a cross-jurisdictional trip.  How does this data roll-up to the state level? It turns out that it’s not that different.  In Maryland, 87% of AM Peak travelers leave the state on Metrorail, higher in Prince George’s County (90%) than in Montgomery County (84%). Across the river in Virginia, interstate travel accounts for 76% of Metrorail trips, ranging from 71% in Alexandria to 80% in Arlington.  Things look a bit different in the PM Peak, where between 54 to 70% of trips end in a state different from the state of origin.



As we look to funding Metro 2025 and Momentum, it’s important to understand how the system is used.  We will perform more analyses like these in the near future. In the mean time, feel free to download this data and see what else you can find.


Related Posts:

  1. MDworkingVA
    February 4th, 2015 at 15:33 | #1

    This is very interesting, but it would be even more helpful to know the origin-destination jurisdictions of the riders? It’s reasonable to assume most are into DC in the AM, but there are likely many from Maryland to Virginia (and visa-versa) as well as intrastate but cross-jurisdictional.

  2. Michael
    February 4th, 2015 at 16:52 | #2

    This isn’t exactly what you’re looking for, but we do have jurisdiction (state*) of ridership by station already for the Metro 2025 stations online:

    *State, Commonwealth and District

  3. Stephen
    February 5th, 2015 at 12:58 | #3

    I would be interested in a year-over-year comparison between the October 2014 and October 2013 data. However, I understand that October 2013 presents an unusual case, given the two-week federal government shutdown in that month. With this release of October 2014 data, has Metro determined an appropriate, other time period comparison. Or, does Metro have some sort of estimated ridership for October 2013 that tries to “correct” for the shutdown for planning purposes?

  4. JPC
    February 6th, 2015 at 09:32 | #4

    I kind of don’t like much of the commenteary on these stats for one simple reason: They seem to distort the issue, because AM Peak is when are kids traveling to school, which is(almost) always intra-DC travel. So realistically, the numbers for “cross jurisdictional” traffic are probably pretty similar for AM and PM peak

  5. Dave
    February 6th, 2015 at 09:39 | #5

    Um… this isn’t exactly a shocking finding.

    1. The system forces everyone to transfer to another line in either the District or in Arlington… so “Metrorail customers” DON’T actually “have many opportunities to stay within their state on a cross-jurisdictional trip.” Try going from Greenbelt to Largo and stay within Prince George’s on rail. Or Huntington to Tysons Corner and stay with Fairfax. So the system is designed in a way that everyone EXCEPT district (and maybe Arlington) residents will have to enter another jurisdiction to complete their journey).

    2. The price of Metrorail is so high especially out in the suburbs where stops are further apart, so it’s usually cheaper to take the bus for any point A to point B journey. Combined with the problem identified in (1) above, it’s usually MORE DIRECT to take the bus as well if you’re staying within a single jurisdiction. So this tends to encourage people to use Metrorail only when the trip is really worth the money to do so AND only when the trip cannot be reasonably completed by bus, e.g. there would be a lot of local-bus-to-local-bus transfers or the bus journey would take way too long or both. (Remember: Because each bus system tends to stay within its own jurisdiction, Metrorail is often the ONLY service that DOES cross jurisdictional boundaries, especially during off-peak hours.)

    If Metrorail ends up being the ONLY reliable way to cross boundaries… it’s not at all shocking to find out that people use it heavily whenever they have to cross boundaries… any more than it would be shocking to realize that if an airline routes all of its flights through a hub airport and it’s the only airline to use that airport 24/7/365, then people living in that hub use that airline for almost 100% of their travel too. Cross jurisdictional travel IS the reason for Metrorail’s existence.

  6. JimT
    February 6th, 2015 at 10:37 | #6

    Does riding from Greenbelt to Suitland count as crossing a state line in this analysis?

  7. Low Headways
    February 6th, 2015 at 11:34 | #7


    Unless, for some reason, you tap out when you transfer, the only data recorded is the origin and the destination. So in your Greenbelt to Largo example, the SmarTrip card taps in at Greenbelt and out at Largo, and it has no idea where you’ve been between. Anything in DC or Virginia are where people have actually exited the station.

  8. Michael
    February 6th, 2015 at 11:55 | #8

    I don’t think it has anything to do with where school kids travel on Metro but instead the fact that DC is the regional employment center. DC’s 28% in the AM is reverse commuters. DC’s 60% in the PM is people leaving their jobs in DC and traveling back to their homes. Again, this is by station location, not jurisdiction of residence.

  9. Steve
    February 6th, 2015 at 12:16 | #9

    Another way to look at the data is the failure of suburban residents to use Metro for commutation within their state and the failure of zoning and planning to create opportunities for suburban employment at suburban Metrorail stations.

    Not enough folks are boarding the Red line at Shady Grove for instance and getting off to work at NIH or downtown Bethesda. People would rather drive on I-66 to jobs in Arlington than park and ride WMATA.

    If more folks used Metrorail for intrastate commutation trips WMATA could fill seats twice and be more productive.

  10. Steve
    February 6th, 2015 at 12:19 | #10

    High weekend cross jurisdictional travel shows the regionalism of Metrorail but it also means that suburban residents only use Metrorail on weekends to ride into DC for sporting and cultural events. They apparently aren’t using WMATA for intrastate errands or shopping.

  11. MDworkingVA
    February 6th, 2015 at 13:32 | #11

    I did see that and definitely think that is helpful information. However, I guess what I’m wondering is more so how the core would be potentially impacted by a line that did not travel through downtown – like an extension of the MTA Purple Line to Tysons or Fairfax, catering to the existing Beltway traffic, or a line connecting Rosslyn to the Red Line in NW, allowing more Arlington to Bethesda/Rockville traffic. Or perhaps southern Green Line Traffic to Alexandria/Springfield. Right now, for good reason, the system requires all riders to go through Downtown DC (excepting transfers at Ft. Totten or trips from Rosslyn to Pentagon) even though significant traffic crosses the Potomac each morning between Montgomery and Fairfax. The Silver Line will help, but such data could provide very helpful in future planning efforts.

  12. MDworkingVA
    February 6th, 2015 at 13:40 | #12

    Or maybe the problem is the development patterns around Metro and the comparative access and travel times, not the riders. Zoners and planners don’t build buildings, developers do. And I do believe that, despite the anecdote, people who live within walking distance of Shady Grove would drive to NIH. The problem is that once you’re already driving, it requires a significant incentive to transfer modes (either heavy traffic or steep costs). Travel time would be much increased is a Gaithersburg resident drove to Shady Grove, parked, and trained to NIH. Wouldn’t be worth it.

  13. JimT
    February 6th, 2015 at 15:31 | #13

    @Low Headways
    People seem to be making the opposite assumption on GGW. Regardless of what the smarttrip data know, WMATA is aware that the Green Line crosses a state boundary and Michael knows that the Green Line crosses a state boundary. It is not hard to tabulate the data matrix with realistic assumptions, assuming that Michael of WMATA meant what the title of his post says.

    On the other hand, you may be right, in which case the title of this post is inaccurate.

    So WMATA, please tell us: is the title of this post accurate or are you rally calculating the portion of trips that start and end in the same jurisdiction?

  14. Michael
    February 6th, 2015 at 15:34 | #14

    Okay, sorry for the confusion. The title of the post is misleading: it’s really about same-jurisdiction entry and exit stations. Greenbelt to New Carrollton would not count as a “boundary cross” even though you have to enter and leave DC in order to make the trip.

  15. JimT
    February 6th, 2015 at 15:44 | #15

    @MDworkingVA What do NIH employees pay for parking these days?

  16. Michael
    February 6th, 2015 at 15:49 | #16

    I just called their parking office. Parking is FREE for NIH employees.

  17. MDworkingVA
    February 10th, 2015 at 09:34 | #17

    Interesting, but they would have to charge a LOT to get those people to Shady Grove. It’s $5.10 to park at Shady Grove. So if we consider the NIH employees that already travel past Shady Grove (because certainly no one in Chevy Chase or Greenbelt or Potomac is going to drive to Shady Grove first), and add in the time for parking and walking to and waiting for a train – assuming no delays, which is often not reasonable on the Red Line – plus a 15 minute train ride – plus the walk from the station to the NIH Main Building (over 1/2 mile), the trip would reasonably be an 35-45 minutes from entering the Shady Grove lot in one’s car to the entering the NIH building, at a cost of at least $10 round trip (including parking).

    Meanwhile it’s a 20-25 Minute Trip right now. Even if it was $10/day to park at NIH, no one is getting off of 270, parking, and taking the Metro. The solution is TOD – If someone can walk to the train, their incentive increase exponentially.

  18. Damon
    February 15th, 2015 at 13:05 | #18

    Yes, the data show that suburban residents “apparently aren’t using WMATA for intrastate errands or shopping”, but why would they do so when they can shop closer to home (but it is harder to find work near home)?

  19. Damon
    February 15th, 2015 at 13:09 | #19


    TOD is good but a partial solution. Need to 1) expand one-seat ‘frequent transit’ OD pairings through new transit investments (which would also mean expanding TOD possibilities); and, 2) charge for parking – many NIH staff can take the existing Ride On to work. They just choose not to because parking is free.

    Ironic that NIH has free parking.

Comments are closed.