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


Metro's future rides on Saturday night

The builders of our Metro envisaged a railroad that would take commuters from scattered residential neighborhoods to jobs in a small downtown. Non-work trips by tourists and residents were expected, to be sure, but they were an afterthought. And sixteen years ago, a commuter railroad is essentially what we had.

Photo by thisisbossi on Flickr.

In May 1995, nearly one trip of every ten carried a morning commuter to just five stations in what was then the downtown business districtDupont Circle, Farragut North and West, McPherson Square, and Metro Center.

An equal number of trips took those commuters home. Federal office buildings were, of course, major destinations as well. Once the workday was over, trains were rarely crowded.

Today, we have a changed system.

Not only are there almost 50% more riders each week, but they take a very different mix of trips. Commuters still flock to offices in the old downtown, but they are not Metro's future. In May 2010, fewer passengers exited Dupont Circle in the morning rush hour than in the same month of 1995. The two Farragut stations were up, but only by 5%. As a share of Metro's total weekly ridership, morning commutes into the old downtown plummeted from 9.8% to 7.2%.

The biggest source of ridership growth is the emergence of new commuting destinations as the land around stations has been redeveloped. A new downtown blossomed between Gallery Place and Union Station while the federal buildings south of the Mall produced more riders, and reverse commutes to the suburbs soared.

But the real key to understanding Metro's growth is found elsewhere: in the explosion in non-commuting trips by area residents. The best way to measure this is by looking at Saturday night travel. In 1995, 39,198 people entered the rail system after 7:00 Saturday night. By 2010, the number was 94,646. The increase of 142% dwarfs the 43% growth in weekday morning commutes.

Not only are more people riding on Saturday night, they are different people. The ten busiest Saturday-night stations are listed in the table. Back in 1995, tourists going home late were numerous enough to give Smithsonian tenth place on the list.

Today that station's ridership has risen from 1830 to 2457, but it ranks 25th and is barely ahead of Fort Totten. Columbia Heights and U Streetone not even open in 1995, and the other rarely usedhave surged high on the list.

These busy Saturday nights are the key to understanding Metro's ten years of growth. The root cause of ridership growth since the late 1990s has been increasing demand for an urban, transit-oriented lifestyle.

True, the majority of the added trips have been travel to work. But there has been little growth in Metro travel between existing homes and offices. Growth in Metro commuting has followed mixed-use development near the stations.

Transit-oriented development is, of course, market-driven. And what the market demands is the opportunity to take non-work trips on foot or by Metro. This can be seen clearly in Arlington's Orange Line corridor from Courthouse to Ballston. Morning rush-hour travel from these stations increased only 20% from 1995 to 2010, less than the growth in population. Saturday night travel rose 165%.

In the nineties, you got an apartment in Ballston for an easy commute into DC Now, you live there because you want to be in the city but are stuck with a job in Tysons.

Last week, some WMATA staff and Board members suggested eliminating late-night rail service on Fridays and Saturdays to save money and facilitate maintenance. That is a cut Metro cannot afford, because it undercuts the future growth of all-day ridership.

New development is what has brought new riders, not increased mode share in existing markets. To cut back on what makes transit-oriented development appealingmobility without an automobilewould be to destroy Metro's future.

Busiest Metro stations on Saturday night
(total entries and exits after 7:00 pm)

May 1995:

  1. Dupont Circle - 5,336
  2. Union Station - 5,169
  3. Foggy Bottom - 4,062
  4. Metro Center - 3,821
  5. Pentagon City - 3,399
  6. Stadium-Armory - 3,077
  7. Silver Spring - 2,501
  8. Woodley Park - 2,087
  9. Rosslyn - 1,905
  10. Smithsonian - 1,830
May 2010:
  1. Gallery Place - 14,060
  2. Dupont Circle - 11,119
  3. Foggy Bottom - 6,637
  4. U Street - 6,354
  5. Metro Center - 6,287
  6. Columbia Heights - 5,755
  7. Union Station - 5,669
  8. Pentagon City - 5,499
  9. Silver Spring - 4,803
  10. Woodley Park - 4,501
Ben Ross was president of the Action Committee for Transit for 15 years. His book about the politics of urbanism and transit, Dead End: Suburban Sprawl and the Rebirth of American Urbanism, is now available in paperback. 


Add a comment »

The $64 million question is this: what gets cut or changed within WMATA's budget so that late night service is preserved? As things stand now, Metro can't afford it, but neither can we the taxpayer. Something's gotta give, but nobody on this blog is giving details on what that should be.

by Froggie on Feb 14, 2011 10:12 am • linkreport

Excellent post, Ben.

by Matt Johnson on Feb 14, 2011 10:12 am • linkreport

I thought this was a really interesting post, particularly in light of the one arguing the other position on Unsuck DC Metro the other day. Both make really good points. I want Metro to stay open late for my own selfish purposes-I use it at least a couple of times a month during the late weekend hours-but the post on UDM had pretty much convinced me that it wasn't what was best for the system. This one made me much less sure about that.

by Kate on Feb 14, 2011 10:25 am • linkreport


You're asking a necessary question, but you're asking it at the wrong time.

We cannot determine how to close the budget gap without first agreeing on what kind of system we want to have.

This post notes that the days of metro as a commuter-only system are not just over in terms of mindset, but they are over in terms of empirical fact. Those that propose a complete abandonment of all late night service are not acknowledging this.

by Alex B. on Feb 14, 2011 10:25 am • linkreport

I still don't see VA or MD upping their contributions to support residents going into the district (8/10 of the "top 10 stations") to spend their play money. The district is in a severe budget hole and won't be expanding their contributions anytime soon either. Only the fed's have a pile of money to spend, and that's looking unlikely at this point.

I wholeheartedly disagree with your assertion that late night "can't" be cut because that is where the growth is. That's diminishing returns growth on an exponentially decaying population (as the hours progress). Growth should be focused on existing commuter communities under served by metro. Case #1: going after the existing slug line population with a service that's attractive to their needs. Taking over the federal agency express bus services that are already in existence.

The priority of the system is commuters. Late night reveler cash is marginally important, but not an economic priority in a down economy (other than DC's local economy).

There are going to be cuts. Either service is going to be cut or excess staff. I'd like to see overall staff cut, underused stations and bus routes shuttered in off peak times, and the savings split into increased salaries for the remaining employees and improved safety and automation fixes that will scale efficiencies quickly when the economy rebounds.

by Balanced Checkbook on Feb 14, 2011 10:36 am • linkreport

It's still not clear what the advantages of cutting late night services are. The CEO should answer two questions:

(1) What is the savings per quarter?

(2) Is it necessary to conduct maintenance?

If the savings are truly large, we should consider cutting the service. If it is necessary for technical reasons, we should cut the service.

Mr. Sarles should come out and say either "This is about saving money" or "This is a repair and maintenance issue."

However the Board and the Public (hey that's us!) should remember there are trade-offs to each decision. We really risk asking for contradictory things ("More government service! Lower taxes!) Eventually, something has to give. No one here has the responsibility that Sarles or the Board has. We deserve to have the trade-offs presented in public.

by WRD on Feb 14, 2011 10:40 am • linkreport

Very few people talk about Metrobus, specifically, how inefficient it is. I work downtown and see too many empty buses going up town, cross town or out of town. Same goes for routes that few people use but have a vocal minority.

by Randall M. on Feb 14, 2011 10:41 am • linkreport

probably an easier answer is a $1 surcharge for entering the "drunk" stations after midnight.

Those trains are full. WMATA should be making money on them.

It is the empty midday trains that need a price cut.

by charlie on Feb 14, 2011 10:43 am • linkreport

@Froggie"Something's gotta give, but nobody on this blog is giving details on what that should be.

@Alex B. We cannot determine how to close the budget gap without first agreeing on what kind of system we want to have.

If we can acknowledge that there is demand to expand the mission of Metro to be more than a 'commuter' system, then we need to consider a couple of things. First, how do we expande the mission of Metro in a way that its original mission doesn't get impinged on (assuming it's still needed.) I.e., Can this system which was built to be heavily used only twice a day, and close down early still be properly maintained during shorter 'down' hours given the increased usage of the system? (And could all our recent problems with the system have to do with the recent extension to nighttime hours ... without looking at whether maintenance could still be performed ... for a much greater use than the system was built to carry.)

Secondly, if we can get past the first hurdle of 'can this system really handle the increased types of usage?', then we need to have a business model that charges the correct amount for these other 'non-commuter' uses. I believe originally Metro was purposely priced lower than it really costs to operate because the objective was to entice people to use it to get to work rather than to drive in. If we're going to extend it's use to other purpoese, then maybe there isn't the same reason to subsidize these trips. The commuter subsidies were meant to solve an existing problem .. i.e., congestion during 2 heavy rush hour periods Monday through Friday. Subsidizing other purposes at other times may not make sense ... as well as not being feasible. A cab ride between DC and the inner suburbs can be as much as $25 or more. Currently we only charge a measily 10th of that for someone taking the Metro. There's a lot of middle ground in there for increasing the fare during non-rush hours to cover the real cost of expanding the mission of this rail that was built to only carry commuters. Would anyone really object to say something like a $10 fare from Downtown to Ballston .... at 2 ... or 3 or 4 in the morning? It's just slightly more than the cost of a drink. And given that it's not everyday that one would be incurring that, it's hardly a budget buster ...

by Lance on Feb 14, 2011 10:48 am • linkreport

WMATA's suggestion shows how little understanding, leadership, and common sense is employed by WMATA. We need a federal takeover. WMATA will only run itself and our region into the ground.

by Redline SOS on Feb 14, 2011 10:56 am • linkreport

Be careful what you wish for. A "federal takeover" would probably lead to an overriding emphasis on service geared towards M-F commuting.

by spookiness on Feb 14, 2011 11:03 am • linkreport

... and with the goons in charge now, they'd probably just prefer to shut the entire system down and use the stations and tunnels for fabulous indoor shooting ranges.

by spookiness on Feb 14, 2011 11:04 am • linkreport

@Redline SOS--

Shouting "FIX THIS" at politicians and Board members isn't a solution, it's a recipe for even larger problems.

What, specifically, would you do as CEO?

by WRD on Feb 14, 2011 11:05 am • linkreport

@ Redline - please identify one example where the fed run a local function better than a local concern?

I think WMATA is overly concerned about the politics of local government to cut back on services that are less efficient. According to the most recently submitted WMATA budget, the cost recovery rate for buses, the amount of revenue collected over the budgeted cost, is 23.4 percent. This means that for every dollar paid in fares, 76.5 percent is covered by a subsidy. Conversely, rail returns 71.0 percent from its fares.

If we really think about how buses don't work, we can perhaps see great savings in this area to support rail, that does work.

by Randall M. on Feb 14, 2011 11:07 am • linkreport

One also has to consider the drop in income from evening trips. My guess the 14,000 trips will cut in half on the weekends. I know the costs are still more then the income, but it means the potential savings will likly be less.

Again I have to ask the question. How do other systems which are open late make this work.

by Matt R on Feb 14, 2011 11:09 am • linkreport

@ Matt R.

For systems that have late night or 24 hour service, the it appears that major reason is because they have more than two tracks per line or they have system redundancy, several lines within close proximity. In some cities, they have a greater number of switches, allowing trains to move around areas under repair more easily.

by Randall M. on Feb 14, 2011 11:15 am • linkreport

I don't get it. The WMATA system isn't that much larger than 15 years ago (three new stations), and has a lot more riders. One would think even with the additional wear and tear on facilities that the deficit gap for running the system should be smaller since you have more people paying to use the same sized subway system. What gives?

by GerryS on Feb 14, 2011 11:15 am • linkreport

@ GerryS

0. 15 years ago, the metro system was newer and their were fewer people using it during fewer hours.

1. More people using the system means more electricity because of heavier trains.

2. More people means more trains.

3. More trains means more employees are needed to maintain and repair those trains.

4. Those people who repair the trains probably should be paid.

5. While the fares have increases as well as usage, it pales in comparison to the actual cost of every thing above.

by Randall M. on Feb 14, 2011 11:23 am • linkreport

I am not sure what this post was meant to do. You use post- 7pm stats to try to buoy support for keeping the metro open from midnight to 3am.

Closing metro at 7pm isn't on the table, so why is anyone even talking about it?

If you want to talk about late night service, then about it in context and use the time frame currently under debate, 12-3am.

The late train (per the washpost article this morning but something we all know anyway) is the bar train. A train for people to pass out, urinate and vomit on during their ride home. Plain and simple. It doesn't serve any other purpose, So we have ~13,000 people in 3 hours, or 4300 trips an hour in a regional system with more than 100 miles of track and 86 stations.

Keeping an entire regional system open for 3 hours for 13,000 people plainly put, is ridiculous. Folks can wax poetic about how it will destroy local businesses and plunge the DC mass transit system into the dark ages, but it simply doesn't pass the smell test. 13,000 people is nothing in a metro region of 5.4 million people and the economic effect will obviously be neligible, if any at all.

As all the interviewed people said in that article, cutting the late night trains aren't going to stop them going out.

by freely on Feb 14, 2011 11:25 am • linkreport

@Alex B:

Valid to a point. The days of Metro being a commuter-only system may be over....but the last time I checked, packed trains and platforms/broken down trains/mishaps/etc etc are being felt the most (and commented on the most) during the peak hours, not during the late night hours. Plus, maintaining/improving service during the late night hours isn't nearly going to have as much of an impact on the local economy as maintaining/improving service during the core hours.

I'm not saying late-night service should be cut...if there's a way to fund keeping it, all the better. What I'm saying is that the calls to keep late night service need to be mindful of the fiscal and usage realities. So far, I haven't seen that. Nor have I seen anyone come up with a realistic proposal that both solves the budget gap and allows late night service to remain.

I disagree with's *NOT* the wrong time to be asking that question. The sooner the question gets asked, the sooner solutions can be found, instead of through the last-minute ransackling of budget negotiations.

by Froggie on Feb 14, 2011 11:28 am • linkreport

What we need is data for trips after 12.

Right now the top 10 stations have about 66,000 trips after 7. Thats quite a bit and its only the top 10, I would not be suprised is the total was closer to 120,000 for the whole system.

by Matt R on Feb 14, 2011 11:33 am • linkreport


The ridership for the entire system per the Post article this morning is 13,400 trips total in the 3 hour period from 12:00 to 3:00am.

by freely on Feb 14, 2011 11:38 am • linkreport


There were a lot more than 3 added in 15 years.

Red Line
Glenmont 1998

Largo 2004
Morgan Boulevard 2004
Franconia–Springfield 2007

Georgia Avenue - Petworth 1999
Columbia Heights 1999
Branch Avenue 2001
Suitland 2001
Naylor Road 2001
Southern Avenue 2001
Congress Heights 2001

Also as big factors that were added since 1995: Nationals,DC United and Verizon Center

by RJ on Feb 14, 2011 11:39 am • linkreport


Your assumption, shared by most posters on this topic, is that the late train is the bar train. This assumption is linked to yet another assumption; that all people work 9 to 5.

by Ray on Feb 14, 2011 11:43 am • linkreport


I'm still not convinced that the late-night-service or maintenance tradeoff is actually real. I think asserting those two options as mutually exclusive is lazy, to be honest. The standard should be to provide as much service as possible. If you have to make changes to that to do maintenance, so be it - but simply arguing that we should cut all late night service to do maintenance is not a coherent position.

Regarding funding - this is a matter of priority. The conception of the system matters, because the current conception is one where Metro has been a key element in getting people to live car-free. It is a necessity.

The old conception of a commuter system that might help reduce traffic congestion, one that uses park and rides all the time, that's something that's nice to have but hardly a necessity.

The fundamental questions are these: what kind of city do we want to be? The follow-up is obvious; based on that vision, what kind of system do we need to have to achieve that goal?

Simply saying that cuts need to be made is a cop-out, it doesn't make you answer this question or address the implications of your answer.

I can tell you what my answer is. As Ben Ross' post notes, ridership has been surging in off-peak hours. This is part of a lifestyle now. It's re-shaped the physical nature of the city itself. Not too long ago, the people overwhelmingly embraced fare hikes over service cuts - further evidence of the indispensable nature of the system.

It's clear to me which way we're going. I want Metro officials to acknowledge that and discuss it before they make their decisions.

by Alex B. on Feb 14, 2011 11:43 am • linkreport

Agreed. Totally agreed. What are they thinking? Here's an EXCELLENT video story about this in the WaPo.

by nicolleindc on Feb 14, 2011 11:46 am • linkreport


It isn't an assumption. It is a fact. What do you think the 12-3 train is?

And yes, most people do work 9-5 and the vast majority of metro users are the 9-5 commuter crowd as is evident by all the facts within this posting.

by freely on Feb 14, 2011 11:48 am • linkreport

How about installing solar collectors along all above ground tracks (similar to Germany) to help reduce the electrical operating costs?

by William on Feb 14, 2011 11:52 am • linkreport


Where did you get the 13,000 figure? Also, why aren't you taking into account the fact that Metro runs fewer trains as the night goes on and ridership dwindles?

The cost of offering additional late-night service absolutely pales in comparison to the cost of adding additional rush-hour service, no matter how you spin it.

Extra late-night service requires additional operational expenditures. Employees need to be paid, and a few more trains need to run. Adding more rush-hour service requires purchasing extra trains and hiring more drivers (which are poorly utilized, as they only need to work a few hours a day to cover the rush hour), and making various infrastructure and signaling improvements to allow trains to run closer together.

by andrew on Feb 14, 2011 11:55 am • linkreport

Metro has to reengineer the bus lines to get more commute riders into the subway, instead of a "poor mans" alternative to the subway. Metro can't be the piggy bank for jurisdictions to pass on the costs of social policy that they should be paying for themselves.

MD needs to suck it up and fund a cross boarder connection across the Potomac for it's residents commuting to Tysons and Dulles. Increased ridership at peak times is the way to increase income with lower increased costs.

New bus lines need to reduce MetroAccess usage and to target new pockets of commuters that are currently underserved. That means buses running on shorter routes, more frequently.

Metro has to find a way to run their buses and trains in a *highly* predictable manner --shorter bus routes to reduce arrival variance.

by ab on Feb 14, 2011 11:55 am • linkreport

I'm with WRD - we need to see an objective analysis of what reducing this service would achieve and I don't think that GGW or Metro has presented a realistic, fact-based analysis yet of this proposal and its consequences.

by grumpy on Feb 14, 2011 11:57 am • linkreport

$3M dollars a year for 13,000 riders a night doesn't seem like a particularly good ROI to me.

by jcm on Feb 14, 2011 12:00 pm • linkreport

+1 Alex.

Framing this as "safety vs. late night" is highly misleading.

I can understand why WMATA employees don't want to work that late -- it is never fun. Overtime is another problem. But safety -- just close the damn Orange line for a weekend and fix the switches.

by charlie on Feb 14, 2011 12:06 pm • linkreport

@William Solar cells is a mice nuts solution to the electricity problem. More efficient/lower weight cars is where the beef is, but is largely constrained by passenger loads.

@Alex B. - The question about what the system "should be" is important, but the immediate follow on question (for anyone who's even remotely financially responsible) is "is it financially viable?". I'd love to dig a ground water look to get geothermal energy into my house, but on my salary, it's not going to happen. 100% of the people on this blog would love a late night service, but it has been instituted in a very irresponsible way by the WMATA board and if it comes down to the economics of late night users and daytime users, late night users lose hands down.

The train system has been starved of basic maintenance for 20+ years. What people /who understand the system/ are saying is that it can't be done in the 3am-6am window safely. You can barely get a tunnel project started in 3 hours much less do anything beyond caulk a few seems or replace a rail.

by eb on Feb 14, 2011 12:08 pm • linkreport


See my above wapo link to the 12-3 usage stats.

And Andrew, running fewer trains while keeping the system open doesn't matter. Why? Because the system is still open.

Employing many hundreds of station operators, drivers etc to keep an 86 station system open for 3 hours to service 13,000 people is an enormous waste of money.

There is no other way around it.

by freely on Feb 14, 2011 12:10 pm • linkreport


At least keep your units consistent. $3M a year for 1.4M riders a year (13400 x 52 x 2), or a cost of $2.14 per trip, which represents a farebox recovery of just over 50%. Still way better than what Metrobus manages during the day, and not horribly far below the 71% that Metrorail manages overall (and this isn't considering the fact that late-night service does not require any additional capital expenditures).

Sounds to me like you've got an axe to grind, rather than a coherent argument to make. Late-night riders are not receiving a ridiculous subsidy compared to the rest of the system, especially considering the reduced service that they have to contend with.

by andrew on Feb 14, 2011 12:11 pm • linkreport

@charlie. Come on man. If it was just a question of shutting down the system for a weekend and fixing every single problem, it would be done by now. What we're talking about is how does metro perform regular maintenance on a system going forward. Things break down at a fairly predictable rate and there aren't enough hours from 3am-6am in the year, to keep up with the breakage rate.

And I'm sure metro would be happy to shut down all weekend (including nights) to do the work. That's an even better solution for them.

by eb on Feb 14, 2011 12:12 pm • linkreport

@andrew You have to add the opportunity cost of every lost hour of maintenance. What does it cost the maintenance budget for every extra hour that they're open.

by eb on Feb 14, 2011 12:15 pm • linkreport

What would be really helpful is knowing what the actual hour by hour usage stats are from 12-3am.

Yes, we know that there are on average 13,400 trips taken from 12-3, but anecdotally (and I don't have the actual wmata numbers to prove) I would imagine that the bulk of those 13,000 riders are in the 12-2 time frame.

by freely on Feb 14, 2011 12:18 pm • linkreport


104 Fridays and Saturdays per year multiplied by 13,000 riders per late night = 1,352,000 riders for $3 million/year. Not altogether that bad... does the $3 million include fare recovery?

by Adam L on Feb 14, 2011 12:24 pm • linkreport


Actually, that's exactly why we want WMATA to come forward and tell us what sort of maintenance they're planning to do, and how closing early every single weekend will help them do it.

It's also why we're generally okay with occasional weekend closures. They get a lot of work done during those, and they help to minimize disruptions the rest of the time. An extra 48 hours is a lot more helpful than an extra 3.

by andrew on Feb 14, 2011 12:34 pm • linkreport

@ andrew I don't have any sort of axe to grind. I didn't claim that "Late-night riders are receiving a ridiculous subsidy compared to the rest of the system". I said it's a poor ROI. The system has a huge budget deficit. If you're looking for things to cut, late night service isn't a bad place to look. It's a nice sized chunk of money and it inconveniences relatively few people.

by jcm on Feb 14, 2011 12:43 pm • linkreport

Are you interested in voicing your opinion to the WMATA Board about whether or not to cut weekend late night service??

If so, attend the next Riders Advisory Council, it is being held on Wednesday, March 2nd at 6:30 pm.

Address: Metro Headquarters
600 5th Street, NW, Washington, DC.
All are welcome, this is open to the public.

They will be looking for feedback from riders in regards to the idea of cutting weekend late night hours.


Riders' Advisory Council
On September 25, 2005, Metro's Board of Directors established a Riders' Advisory Council. The Council advises the Board on issues affecting Metrobus, Metrorail and MetroAccess service.

The 21-member Council includes six individuals from Maryland, Virginia, and the District of Columbia, two at-large members, and the head of Accessibility Advisory Committee.

Council meetings are open to the public. Meetings are held the first Wednesday of each month from 6:30 to 8:30 p.m. in the lobby level Meeting Room at Metro's Headquarters located at 600 5th Street, NW, Washington, DC. Please call 202-962-2891 or email for more information.

If you have a comment or suggestion that you would like to share with the Council, you can email

To apply for a position on the Riders' Advisory Council, please complete and submit an online application form. Or if you prefer, you may print, complete and mail a paper application.

Please note that the Riders' Advisory Council is unable to address specific service complaints. If you have a complaint, commendation or suggestion for Metro, please contact Metro customer service.

by J on Feb 14, 2011 12:53 pm • linkreport

And what the market demands is the opportunity to take non-work trips on foot or by Metro.

not bike? :(

by Peter Smith on Feb 14, 2011 1:04 pm • linkreport

Let's play budget analyst for a minute here. This Post article says there are an average of 13,400 each of the two nights per week in question.

Round down to 13,000 for conservatism, so that 26,000 people per week. Multiply by 52 weeks and we get: 1,352,000 people riding after 12am and before 3am per year. At a total reported yearly savings of $3MM, that comes to roughly $2.22 in yearly savings per person per year.

That seems material to me, at least at first. For some perspective, Metro's total operating revenue (per their audited financials) is roughly $728MM and operating expense is roughly $804MM. The proposed gap between the two for the upcoming fiscal year is roughly $72MM. This plan would decrease the gap by 4%.

So those who argue this will save money have a point. It will save money, and enough money to matter, although only little bit. However there are obviously lots of things that save money and just because it saves money doesn't mean we should do it.

Here is where I would speculate on how much different levels of service cost. But I won't, because I don't know. Metro does (or should!).

Here is where Metro's CEO (or actually the CFO through the CEO) need to present options, if this is all about money.

There might be ways to ring out savings. Some of them have been thrown up as comments here, in fact. But at the end of the day, it's all speculation as we don't have the data.

Pressure should be on the CEO to show us why cutting the service is preferable to the alternatives. What would I have to pay to make it worth Metro to run the trains? Are their costs variable, in which case cutting an hour off service would result in substantial savings? Or are they fixed? (Cut all hours for any savings and an hour of cuts gets you no savings)

It's clear that lots of people, myself included, value service after 12am. It's not clear that we all value it enough to pay for 3am service. What are the trade offs here?

by WRD on Feb 14, 2011 1:05 pm • linkreport

Interesting to watch everyone still arguing over whether late night service should be cut given the budget shortfall. Provided the system can take the extra riders, making available late night service is easy ... Just hike the fares! $10 to anywhere in the system after 11 pm ... Problem solved.

by Lance on Feb 14, 2011 1:16 pm • linkreport

Sure, but we wouldn't even need a $10 solution. If we're looking to recover the costs -- add $2.22 to the fares, and call it even.

by Jacques on Feb 14, 2011 1:28 pm • linkreport


Just hike the fares! $10 to anywhere in the system after 11 pm ... Problem solved.

Umm..NO. As the fares go up, ridership goes down. Whether a fare increase will raise or lower Metro's revenue is a quantitative question. If their demand curve is inelastic, it will more likely increase revenue. If it's elastic, it will more likely not.

by WRD on Feb 14, 2011 1:29 pm • linkreport

I would also like to see the revenue/cost breakdown for this service too.

The money you will save by cutting service is equal to:
Cost of providing the service
Revenue gained from passengers using that service

BUT you also have to take into account the fact that you will save less due to the fact that some other (earlier) trips will also be lost. E.g. if I used metro to go out, but now my return train doesn't exist, maybe I choose to drive both ways instead of metro+cab. That means you lose my late night trip AND the revenue from my earlier trip.

by MLD on Feb 14, 2011 1:40 pm • linkreport

@WRD "Umm..NO. As the fares go up, ridership goes down."

Isn't that how most things work in life? And please, let's not hear about the $3/hr bar worker who won't be able to afford to get home. The bars (and all businesses) will pay their workers what they need to to get them to be able get to and from work ... or they be finding themselves with a shortage of workers ...

Of course, the bar owner could argue that he/she is at a disadvantage to the bar owner in Bethesda or Ballston where there is plenty of cheap parking available ... But THAT is a discussion for a different thread.

by Lance on Feb 14, 2011 1:47 pm • linkreport


Supply and demand control everything, not just "most things." Which is why we can't just say raise fares $2.22 on average and call it even, as Jacques said.

If you look at what I said, I am arguing that we should look at raising fares. But it might not solve the problem if fares are already at the revenue-maximizing level. If that is the case, increasing fares will make the $3MM loss worse not better! Further, the short term revenue maximizing fare probably isn't the long term one.

There are moving parts here and it's not a matter of addition and subtraction.

As usual, I'm making an economic efficiency argument, NOT an equity or values argument.

by WRD on Feb 14, 2011 1:55 pm • linkreport

@WRD, I understand there are many parts to the mix ... including, as I mentioned at the beginning of the thread, we need to be sure they system can be made into a system serving more than just commuters short of substantial capital investments. Back when CM Graham extended the hours I don't think much thought was given to whether the system could actually handle the extra load ... and whether that would have maintenance considerations and other considerations such as shortening the life span of the capital components. But IF that hoop can be jumped, then raising the fares to an optimal point is possible. And if it can't, than it's a given that more capital investment (i.e., 'more of a system') is needed.

by Lance on Feb 14, 2011 2:18 pm • linkreport

So 2.22 is to much to increase the evening fares, but this could be done in such a way to at the very least reduce the loss. Perhaps adding 50 cents twice over a 6 or 8 month period.

Thus the evening loss is reduced by about a million and hopefully the increase was gradual enough that ridership does not go down.

by Matt R on Feb 14, 2011 2:20 pm • linkreport

As CEO the first thing I would do is coordinate with BART and METRA and other dual tracked systems and go to Congress and say we need third and fourth tracks to maintain the current systems and expand to meet future population and ridership growth.

It creates jobs short and long term. It addresses the maintenance issue. And it addresses the capacity issue.

Is there current dollars for it? No. But seeing the job creation and increased tax revenue from income and long term growth near TOD's would make Congress look twice.

by Redline SOS on Feb 14, 2011 2:27 pm • linkreport

I'm still new here in Washington, and I like this blog mostly because of what it says on its masthead: "The Washington, DC area is great, but it could be greater." And the arguments that try to justify reducing Metro's hours, or that demonize the "drunks" who are riding the rails after midnight do not seem to be in the spirit of making DC greater. I thought that Ben did a great job of reminding us that transit is being used more and more for non-work trips, and that a system with longer opening hours is one that is desirable for those of us who want to live an urban lifestyle, whether we use the train late at night or not. For those of us who want to see Washington be truly greater, we should be fighting to keep the later opening hours rather than help to justify something lesser. Actually, the biggest questions I had about Metro when I moved here are: "Why is this the most expensive transit system to ride in the country?" and, "Why isn't the subway open later on weeknights and earlier on weekends?"

by DSDC on Feb 14, 2011 2:40 pm • linkreport

@Redline SOS--

That's great, but Congress would laugh you out of the room! The cost to upgrade the system would be measured in the tens of billions of dollars if it can be done at all.

No one has the appetite to pay for it. Not Maryland, not DC, not Virginia, and certainly not the Feds. In case you haven't noticed, they money isn't flowing as much any more and one House of Congress has proposed cutting the $150MM the Feds already offer up.

To be honest, I don't blame the politicians. If I were God of Money, I would tell Sarles to fix the agency using the money he has and come back for more once he's shown steps in the right direction. And if I was Congress, I would say to the jurisdictions: Pony up a dedicated source of funds. Then maybe we'll talk about expansion 5 or 10 years down the road. Maybe.

Maybe Sarles should do a study and put out a figure. Maybe he should go to Congress. But when he gets shot down by everybody, he's lost credibility with politicians and the Board, plus he's back to square one.

If this was SimCity, I'd agree with you--adding a 3rd track is a great idea. And hey, I might be wrong on the cost. But I doubt it. And none of that will fix any of the fixable problems Metro has now.

I think this shows how pie-in-the-sky these issues are for the average rider.

by WRD on Feb 14, 2011 2:48 pm • linkreport

Just a note, Metro is the second most expensive transit system in the country. BART takes first place.

by Matt Johnson on Feb 14, 2011 2:53 pm • linkreport

Ok, so I guess we can stop with this false argument about not matching the "world class" metro systems, considering we stay open later than all other major world metro's (NYC not included).

I copied this off a comment off wapo. I checked the times for the first two and they were correct.

London: 1:00
Paris: 1:15 Fri, 2:15 Sat
Berlin: 1
Tokyo: 1
Boston: 1
Chicago: 1:30
Los Angeles: 12 (a few other trains until 2)

So we already exceed the closing time frames in every single case, quite a bit compared to some enormous systems like London, Paris and Tokyo.

So unless you are accusing London, Tokyo, Berlin and Paris of being back water sleepy towns, I think this strawman has to stop.

Double the fares to cover the cost or shut it down. This is some pretty low hanging fruit.

by freely on Feb 14, 2011 2:53 pm • linkreport

Regarding, adding a higher fee to the 12-3a ride. There is such a thing as a "no-solution" where the fee vs. demand curves never reach the "cost to operate" minimum. In economics that's called bankruptcy. In the public sector that's called "complain for more money". Unfortunately, we don't know where we are without metro input, but there are no buckets of money sitting around waiting to be spent. The "should we be a better system" question is moot if there's not more money.

by EB on Feb 14, 2011 2:58 pm • linkreport

I suspect that the system has not been maintained properly because of a dearth of funding and good management, not because there aren't enough down hours.

Also, building on Lance's point, there seems to be no acknowledgment from Metro of the third option, in addition to maintaining or abolishing late-night service - why won't they put raising the fare on the table?

by Arnold on Feb 14, 2011 2:59 pm • linkreport

"And please, let's not hear about the $3/hr bar worker who won't be able to afford to get home. The bars (and all businesses) will pay their workers what they need to to get them to be able get to and from work ... or they be finding themselves with a shortage of workers ..."

Just so this can be clear (as the late night metro keeps being called the drunk train): It is not (just) bar employees that this will effect. Security guards, office cleaners, theater workers, fancy restaurant employees and fast food folks too, heck, reporters/editors at WaPo work til 2 am.

And... WHEN will they pay their workers more? I'd love to see that - restaurant workers still make minumum. Office cleaners make 8.50$ an hour. When will these businesses pony up more money? After they have burned thru the current employees who NEED these jobs, and who got these jobs in the past 15 years because they could travel to them on the existing transportation system.

I am with the "let's see some data Metro, before you start slashing our transit options." crowd.

Metro is not just a party bus. But I do believe it will go back to being the subsidized M-F crowd commuter route.

by greent on Feb 14, 2011 2:59 pm • linkreport


Chicago's two busiest and most crucial lines for the El (the Red and Blue lines) both run 24 hours a day.

by Alex B. on Feb 14, 2011 2:59 pm • linkreport

You're comparing everyday service in these world cities to 2 nights a week in DC. On Sunday-Thursday, WMATA starts closing before EVERY other heavy rail system in the United States. Every single one.

What you're arguing for, if I understand it correctly, is to make that true on Fridays and Saturdays, too.

by Matt Johnson on Feb 14, 2011 3:00 pm • linkreport

Why would $2.22 be too much to increase evening (did you mean late night?) fares? The alternative is not having any late night service.

by Lance on Feb 14, 2011 3:08 pm • linkreport

@WRD - You're right, it would cost a ton of money. But investing now is cheaper in the long run. Always is.

And given the long term projections, it has to be done. Capacity will be reached in the next decade. Then what?

Someone has to go to Congress and tell them the truth. Even if the answer for the next few years is "No", speaking the truth now will get us that much closer next year.

by Redline SOS on Feb 14, 2011 3:26 pm • linkreport

The number for Berlin is wrong too, the S-Bahn closes for only 90 minutes on the weekend.

by Phil on Feb 14, 2011 3:30 pm • linkreport


As others have pointed out, much of that information is incorrect or is only for certain lines. In addition, those cities have a myriad of other non-private-automobile transportation options, including extensive bus systems, trams, and light rail that provide near 24-hour service.

by Adam L on Feb 14, 2011 4:14 pm • linkreport

The DC area lacks a true multi-layered transit network. The choice between heavy rail or bus/streetcar (no, streetcar is not light rail in my book) ultimately won't cut it. And, streetcar is not a replacement for heavy rail or even light rail.

For instance, a light rail system (ie rapid rail with lower capacity) in the DC area could provide lower capacity transit needs for off peak times so that we wouldn't have to rely on our heavy-rail network to do the job. Also, the usefulness of transit is a consideration. I live in DC and have given our bus network a chance, but I keep going back to riding Metrorail, even for short distance trips. Why? Because our bus network is less attractive to me for crosstown trips. It takes too long and is too unreliable. We need a real multi-layered system consisting of bus, streetcar, light-rail, and heavy rail system. That need is largely met in other world cities where a true multi-layered transit system is of great use, and we need to get there too. With studies like the DC Alternative Analysis, we have further limited our choice by not seriously considering light rail in some of the corridors studied. The transit mode should fit the need. We have not done a great job of diversifying what's available to fill our transit needs in this region.

There is such a strong reaction to keeping Metrorail service open late at night because it is the only really useful mass-transit option available.

by JOT on Feb 14, 2011 4:38 pm • linkreport

Boston's T closing at 1 means that trains leave the ends of the lines at 1, so you can get a train at the central Cambridge/Boston stations between 1:30 and 1:45. Much more useful than Metro's Midnight from Metro Center. In Boston you can hit a movie at 22 and know you can get a train home on a Tuesday night. Not so here.

by dcseain on Feb 14, 2011 10:19 pm • linkreport

freely, you post misinformation and misinformation.

Yes, cities like london and paris close up at 1am every night. BUT, they have super extensive night bus networks. Essentially 24 hour service for the entire city and the RER suburban networks.

Heres a london night map.

Heres the Paris night bus network. It blows away most cities DAY networks, including DC.
Select Noctilien on the left.

You also "bet" that ridership drops off heavily after 2am.

Well, no shit. NOBODY wants to be on the last train, because if you miss it, you are royally screwed.

20 minute headways, meaning there may only be 2 more trains after 2am. Extend service to 3:30am....and youll note that the 2:45am gets more riders.

You also seem to forget that the majority of americans work in the service industry. You know those people who take your orders and you avoid making eye contact with? Or the maintenance/cleaning crews you never seven see? They're not going home at 2am because they're out partying, they're going home because thats when their second job lets out. Theyre the same folks youll find on the 8am train going to their other job. You cut the 2am trip....and they have to quit.

by JJJJJ on Feb 15, 2011 12:06 am • linkreport

dcseain, not exactly correct. Boston starts shutting down at 12:25 am, meaning the subways hit downtown at 12:45am. Boston does have a no-missed-connection policy, so every train waits downtown until all the lines have made it before leaving outbound. Once they arrive at their termini, the buses leave, so the last bus runs begin at around 1:30am.

by JJJJJ on Feb 15, 2011 12:08 am • linkreport

@Arnold Also, building on Lance's point, there seems to be no acknowledgment from Metro of the third option, in addition to maintaining or abolishing late-night service - why won't they put raising the fare on the table?

Yep, that's the elephant in the room, isn't here ... reflected in this discussion by a glaringly lack of discussion. It's like you have people coming to the trough unable to accept the fact that you want more service you gotta pay for it. Instead it's all about "why won't they give us what we want, other cities do that?" ... There's absolutely no sense of 'we want more, let us pay more'. And you know the reason why it's not up for discussion at Metro? All the politicians on their board know that there's a sense of entitlement associated with this service, and none of them wants to be the one that asks their constituents to wake up and face the reality ... that like anything in life, you get what you pay for.

by Lance on Feb 15, 2011 8:18 am • linkreport

Given that its possable that metro could lose its 150 million in fed funding I think this may be a mute point.
If that happens late night should be dropped period. In fact I would also drop weeknights to 11 if this happens.

It sucks, but its something to consider.

by Matt R on Feb 15, 2011 9:18 am • linkreport

Metro already charges more during these hours, so why not raise fares a bit during late-night? For most, it'll still be cheaper than a cab/parking, but still a viable alternative.

And I don't entirely buy the maintenance argument - it amounts to 6 hours a week, which is likely paid in overtime (late nights + Sundays= overtime).

I really wish Metro board members would actually ride the system more - it seems like they're completely out of touch with the average rider.

by John M on Feb 15, 2011 10:05 am • linkreport


The cynical part of me agrees with you. But it is quite early in the budget process. Metro still needs to present their budget to the jurisdictions, especially MTA, DDOT, and VDOT.

Then they have to worry about the federal funding and make sure their lobbyists earn their salaries. Likewise for the 4 Democratic Senators from VA and MD.

So it's still preliminary. Maybe the service will be cut, maybe not, but either way I don't think we can tell until at least the state governments weigh in.

by WRD on Feb 15, 2011 10:59 am • linkreport

Good post, but the first sentence has a "glaring" error. It should have been written to read thusly:

The builders of our Metro envisaged a railroad that would take commuters from scattered suburban residential neighborhoods to jobs in a small downtown and the federal district in the center city.

by Richard Layman on Feb 15, 2011 3:22 pm • linkreport

Richard - that's essentially what I wrote in my first draft. I then checked in Zachary Schrag's history and deleted "suburban." After much political struggle, the Green Line was added to the plan specifically to provide access to jobs for lower-income African-American city residents. The area around Columbia Heights was the focus of this debate.

But even when adding inner-city service, the focus was on travel to jobs.

by Ben Ross on Feb 15, 2011 3:30 pm • linkreport (from an online history presentation by Schrag)

by Richard Layman on Feb 17, 2011 11:37 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