The Washington, DC region is great >> and it can be greater.


Change the Metrorail formula to change incentives

The current Metrorail subsidy formula does not reward jurisdictions for increasing ridership at their stations. Revenues from increased ridership get spread out across the system, while costs are concentrated.

Photo by akirsa.

By changing how we calculate each jurisdiction's share, we can change the incentives. That will encourage jurisdictions to better use the land around their stations, and to start innovative programs to encourage transit ridership.

The current Metrorail subsidy formula takes the cost of operating the system, subtracts all Metrorail revenues from various sources, then divides the remaining subsidy into three pieces.

For one of those pieces, jurisdictions pay according to their population, adjusted by their density. Arlington, Montgomery, Falls Church, Alexandria, and the District pay more than their raw share of population, while Prince George's and Fairfax Counties and Fairfax City pay less.

For the second piece, jurisdictions pay according to the number of stations. Some stations count as shared, like Capitol Heights, Van Dorn Street, Friendship Heights, and Southern Avenue.

For the third piece, jurisdictions pay according to average weekday ridership. For example, DC had about 30% of the ridership, while Arlington had about 11% of ridership in 2005.

The first important thing to note from the subsidy formula is that the revenues are subtracted first, before subsidy is divided. The second thing is that increasing ridership increases a jurisdiction's costs, even as the increased ridership helps WMATA's bottom line.

Let's take an example of Arlington County. The county sponsors and operates Arlington County Commuter Services, which encourages people to take other modes for travel, including transit service. Arlington pays county employees to operate this service, which increases ridership on Metrorail.

These additional Metrorail passengers pay fares into the WMATA budget, which then get subtracted from the common subsidy pool. Some of the benefits of increased ridership are therefore shared with jurisdictions that are not paying for commuter services.

Additionally, since the commuter services have increased Arlington's ridership, when it comes time to allocate the remaining subsidy, Arlington's share gets bigger, not smaller. This is a backward incentive, causing jurisdictions that encourage ridership to pay more, and having the benefits of high ridership shared with jurisdictions that are cutting back on commuter assistance services.

Let's change the formula around. The costs of the system before subtracting revenues should be divided using the formula, then the revenues should be allocated to the jurisdictions. In that case, a jurisdiction like Arlington would benefit from the increase in their riders' fares even while paying more under the cost formula because of increased ridership.

For rail, parking and station advertising revenues would be allocated to the jurisdiction that owns the station. Advertising in railcars could be subtracted from the system costs as a whole. Fare revenues could be allocated base on the origin station, the destination station, or split 50/50 between the two.

With this new formula, a jurisdiction with a large parking lot near its stations would have a big incentive to redevelop the area, provide feeder bus service or otherwise encourage ridership, because they get to subtract the fare revenues from their assistance to WMATA.

The funding formula made sense when the system was just getting started. Practically all jurisdictions had low-density development around their stations, and ridership and cost recovery were low. Now that Metrorail cost recovery is high and could get higher, we should increase the incentive for jurisdictions to encourage ridership, not reduce it.

Michael Perkins serves on the Arlington County Transportation Commission, though the views expressed here are his own. He lives in Arlington with his wife and two children. 


Add a comment »

dang! great idea!

by Peter Smith on May 14, 2010 12:49 pm • linkreport

hmm, I remember arguing with you over this a few weeks ago. I was trying to make the point that places like Arlington (well, Arlington) underpays into metrorail.

minor points: when the system was built, downtown dc was already very dense. And I'm not sure what you mean by cost recovery.

Couple of other minor points: metro ridership residency is determined by survey, which isn't always right, and I forget how the station are counted (do places like rosslyn get counted twice). I know arlington cemetery isn't counted.

How would you allocate the revenue to each jurisdiction? the 50/50 split might result in a higher bill for places like Arlington and DC. (pentagon/federal workers)

The real laggard here in PG county, and I'm not sure that would result in much improvement in their ridership on rail. Most of the money PG county pays goes into bus, no?

by charlie on May 14, 2010 12:55 pm • linkreport

Great ideas. Too bad the implementation of great ideas has about the same occurrence as sightings of unicorns. This is Washington, after all.

by Jasper on May 14, 2010 12:58 pm • linkreport

@Charlie: Our comment exchanges inspired me to write the full post. It was an idea I had in my head for a couple of years now.

Downtown DC was already dense, but it has gotten denser, and more areas of DC have grown in density with the arrival of Metrorail. See Columbia Heights or the NY Ave Metrorail stations for examples.

Cost recovery in the transit business is usually a description of how much of a system's operating costs are provided by sources that are system generated or user fees. So fares, parking, advertising and leases are compared to system operating costs. At 100% cost recovery ratio, no subsidy is theoretically needed to run the system. Metrorail is between 70 and 90% cost recovery, which is phenomenal for a North American transit system.

Metro ridership surveys aren't 100% accurate, but they're statistically valid for the purposes of sharing revenues. The only way to call the surveys 100% accurate would be to have someone at every station every day logging zip codes for people, which isn't practical.

Some stations are counted twice, in that they are officially part of two jurisdictions under the formula.

For rail, you could allocate all the revenue for a trip to the originating jurisdiction or 50/50 to the origin and destination jurisdictions. Because the vast majority of trips end up being round-trips, the distinction is not terribly important.

Maryland pays in on PG's behalf. PG does end up paying quite a bit for rail service.

by Michael Perkins on May 14, 2010 1:12 pm • linkreport

And hey, you're right. Arlington Cemetery is not counted as a station. I learn something new every day.

by Michael Perkins on May 14, 2010 1:14 pm • linkreport

Arlington Cemetery is not counted as a station.

I guess I should start explaining to tourists that regretfully Arlington Cemetery is only for zombies.

Here's another fun fact about the cemetery and Arlington. Arlington has more dead bodies than living inhabitants. Technically, that makes it a ghost town. Yihaa!

by Jasper on May 14, 2010 2:08 pm • linkreport

Well, a broken clock is right twice a day and all that...

"Because the vast majority of trips end up being round-trips, the distinction is not terribly important."

Is that true? I can see it true for commuters but what about non-commuter use.

As I've said before, we need to split rail and bus. A 90% "cost-recovery" is impressive for rail. You could even privatize the system and given vouchers to taxpayers in return for the jurisdiction operating subsidies -- much like what the federal government does.

the point I was trying to make is high-use areas like Arlington drive up the capital expenses, which should change their capital contributions. I do see the argument, when applied to operating expenses, isn't so good.

by charlie on May 14, 2010 3:27 pm • linkreport

The comment about just counting the origin or both end points being equivalent is based on my reading of a TCRP report which analyzed bus line ridership statistics based on boarding and alighting counts for both directions, compared to using farebox boarding counts alone in both directions. The report found that the data was similar enough to qualify for reporting to the FTA and was found to be statistically significant at a high level (I think something like 95% confidence).

So if it's good enough for the TCRP and the FTA, I would say that for purposes of the funding formula, it would be close enough.

by Michael Perkins on May 14, 2010 9:05 pm • linkreport

This is a good post. The revenues can and should be parsed out in a fashion similar to the other formulas, rather than being split evenly. E.g., likely advertising space is worth more in the DC stations, than in a station like Falls Church.

by Richard Layman on May 18, 2010 9:53 am • linkreport

i'm interested in knowing the details of this formula. for a project, we are looking at ways cities in a county in Florida could contribute to its bus transit system and this may be an example. could you please share the actual formula? or point me to where i can get it? thanks!!

by jessica on Jul 20, 2010 4:32 pm • linkreport

@Jessica, The funding formula is described here:

Metro currently subtracts all system revenues from costs before allocating the remainder among the contributing jurisdictions according to the formula described in the presentation.

My change would be to divide the system costs among the jurisdictions according to the formula first, and then subtract the system revenues that come from each jurisdiction's riders (funds allocating according to boardings or alightings or a 50/50 mix). Parking and advertising revenues at rail stations would be allocated to the jurisdiction "owning" the station, and systemwide (railcar) advertising and lease revenues would be allocated to the system as a whole.

by Michael Perkins on Sep 13, 2010 8:34 am • linkreport

Add a Comment

Name: (will be displayed on the comments page)

Email: (must be your real address, but will be kept private)

URL: (optional, will be displayed)

You can use some HTML, like <blockquote>quoting another comment</blockquote>, <i>italics</i>, and <a href="http://url_here">hyperlinks</a>. More here.

Your comment:

By submitting a comment, you agree to abide by our comment policy.
Notify me of followup comments via email. (You can also subscribe without commenting.)
Save my name and email address on this computer so I don't have to enter it next time, and so I don't have to answer the anti-spam map challenge question in the future.


Support Us