HSR and Air Travel – partners and competitors

FRA long distance rail station – Wikimedia Commons image

Alon Levy wrote about the USDOT’s airfare database and the implications for potential high speed rail markets. Alon’s broad conclusions are that the regional markets suitable for HSR haven’t changed all that much in the last decade – the NEC is the most promising corridor, California is second, and others also make sense – but short of a true national network.

His post sparked me to ask – what are the best practices for integrating HSR with airports? High Speed Rail networks have different characteristics from airline networks. Some elements of the services are competitive, while others are complementary. A big part of the political sell for HSR in California was the counterfactual of required airport investment to facilitate travel in the state – but what’s the real impact on airports and airline service?

HSR Corridors vs Airline Hub Networks

Even modest rail systems can effectively compete against air travel in given corridors. Amtrak often brags about the barely-high-speed Acela earning an 80% share of the air/rail market (a notably smaller portion of the overall travel market) between NYC and DC. American and Delta still offer their Shuttle flights between DCA and LGA, but they don’t dominate they way they used to.

Amtrak has the advantage along a linear corridor, where one train making multiple stops serves multiple markets – If United Airlines wanted to match service along the NEC, they’d need a lot of feeder flights from DCA, BWI, and PHL into EWR – and they’d have to run them far more frequently to offer as many departure times.

Airline hubs are a different beast – the spokes all feed traffic into the hub. They’re not flying a route solely for the local traffic, they’re also feeding traffic into their hub to connect to other routes.

International Comparisons

Despite the track record for HSR in other countries, there are surprisingly few examples of excellent air-rail integration. Only a handful of big airports have quality connections to HSR networks: Frankfurt, Paris-CDG, and Amsterdam are most prominent. I found this paper exploring the history of both FRA and CDG:

In Frankfurt, the addition of HSR service allowed Lufthansa to decrease domestic flights and increase international flights without losing domestic passenger feed:

With the opening of the Frankfurt–Cologne and Frankfurt–Stuttgart HSR lines, Frankfurt Airport increased its catchment area by 10 million people who suddenly lived within 2 h of the airport…

Whereas Frankfurt Airport may be limited in terms of capacity, it has managed to support an increase in international passenger traffic and to maintain dominance in Germany as the major long-haul international airport serving the country.

Conversely, at CDG, Air France retained a small number of domestic flights to feed their long-haul flights, despite the TGV’s dominance in markets to/from Paris:

The other two routes, Paris–Lyon and Paris–Montpellier, experienced a relatively flat trend in passenger traffic, although there are competitive HSR alternatives that link these cities directly to CDG. In particular, the capacity on these routes has remained relatively stable, supporting research claims that airlines are likely to maintain a certain number of flights at their hub airports to maintain their network, even with the presence of fast and reliable HSR–air connectivity (7).

The paper’s conclusions about the characteristics of a successful HSR/airport integration:

Infrastructure. To provide feeder or transfer service between HSR and air transportation, the rail station should be located at the airport. If the HSR connection at the airport is constructed as a detour from the primary network patterns on the rail system, it is unlikely that the airport will be served with enough frequency.

Schedule and frequency. Rail operators and airlines often have the same goal of optimizing their networks, but they are separate networks. Coordinating timetables to ensure that rail service meets banks of connecting flights is an important consideration.

Market characteristics of the airport. In the two successful cases in this study, the primary airports with HSR links were the dominant international hubs of each country. For both CDG and Frankfurt International Airport, domestic traffic declined and international passenger traffic increased. Two key factors may have influenced this growth: partial alleviation of congestion at the airport by decreasing domestic flights and success of the HSR lines as feeder service for international flights.

In other words: lots of similar patterns to good transit planning on the rail side of the equation: Be on the way; frequency is freedom, etc. The key difference is about integration: is HSR service best thought of as a means of ground transportation – just extending an airport’s catchment area? Or is it best considered like a true connecting flight – integrated into an airline’s ticketing, loyalty programs, baggage handling, etc.

For airports and airlines, things are a bit more complicated. Both CDG and FRA are huge hubs, and the primary international connecting hubs for their respective countries.

In the US, the airport/airline situation is quite different. As a much larger country, no airline operates a single hub that dominates international traffic the way that CDG and FRA do in France and Germany. Far more medium-distance air travel is domestic travel; US visa rules and unfavorable geography make international-to-international connections less common – meaning domestic feed at big connecting hubs is even more important.

Future US HSR/Airport Links

Alon’s conclusions:

In the Midwest, the core lines remain strong, but more peripheral Midwestern lines, say a bypass around Chicago for cross-regional traffic or improved rail service due west toward Iowa, are probably no longer worth it. The Cleveland-Columbus-Cincinnati corridor may not be worth it to build as full HSR – instead it may be downgraded to an electrified passenger-primary corridor (as I understand it it already has very little freight).

Now, Alon was using this air travel data as a proxy for overall travel demand. Assuming the demand is still there (even if the recent growth has been mdoest) Chicago strikes me as a market with great potential as an air/rail hub – a huge hub airport, located near existing candidate HSR lines, and wouldn’t require a large detour to serve. The full Midwest HSR network centered on Chicago could extend ORD’s catchment area (connecting MSP, MSN, MKE, DTW, STL, CLE, and others) and potentially free up airport capacity for longer-haul, higher-value flying.

SFO is another airport that stands to benefit from HSR. SFO’s airfield is constrained and unable to expand; United’s trans-pacific hub depends on feeder flights often limited by poor visibility; the airport’s location is immediately adjacent to the planned HSR service.

Better planning for CAHSR would add even more value to the connection at SFO – particularly had planners used Altamont Pass, the airport would have a faster connection to Sacramento, increasing SFO’s catchment area.

The Northeast Corridor has lots of potential air-rail connections. The rail line itself passes plausibly close to lots of major airports: DCA, BWI, PHL, and EWR. DCA lacks international flights and is can’t grow; BWI’s hub carrier is almost entirely domestic and doesn’t have many international flights to feed.

Both PHL and EWR are interesting candidates for American and United, respectively. Newark in particular meets all of the criteria for success described in the paper. United already uses EWR’s proximity to code-share with Amtrak in lieu of flying short connecting flights from Philadelphia (though the business practices of Amtrak and United are quite different and can make for a challenging passenger experience) and United recently announced cutbacks to flights within the NEC to EWR – dropping the 4x daily BWI-EWR service due to airport constraints.

Philadelphia would require an airport station along the NEC that doesn’t currently exist. The airport’s proximity to New York and the larger travel market there means a bit of a precarious existence, but PHL’s hub carrier (American) has recently been shifting connecting traffic to PHL and away from their greater New York operations, which are split across LGA and JFK and increasingly focused on serving local traffic.

The potential is there for these airports – the big questions will be a) if HSR service ever exists (or improves), and b) what form the airline/railroad partnership takes – as a true connection, or as extended ground transportation?

Transportation and the Green New Deal

If you follow the people I follow on Twitter, the last few days have included lots of chatter (often pained) about the transportation elements – or lack thereof – in the Green New Deal. Given that transportation is the single largest source of US greenhouse gas emissions (and because the sector relies on direct use of fossil fuels more than, say, buildings), transportation ought to the focus of any climate policy.

So, why isn’t it? The Green New Deal was (and perhaps still is) more of a slogan than an actual policy proposal. Earlier, the think tank Data For Progress released their policy proposal.

The key transportation elements include fairly weak language about increasing “access to” transit and bike facilities, paired with much stronger language about electric cars:

  • 100% Zero Emission Passenger Vehicles by 2030
  • 100% Fossil-Free Transportation by 2050
  • Modernize Urban Mobility and Mass Transit
    The growth of cities, the rapid change in vehicle technology, and the need for low-carbon transportation means that the way in which we move ourselves and goods from one place to another is going to change forever. This transition needs to be executed thoughtfully to meet the needs of cities and the scale of change required. Large investments are needed to increase access to safe pedestrian and bicycle travel, low-carbon bus rapid transit, and electrified light rail.

The twitter commentary jumped all over this: why not address land use at all? Why such wishy-washy language on transit and bikes? Why not mention any host of technocratic ideas and policies that will be useful tools in decarbonizing transportation?

A few arguments in defense of this proposal:

  • These things are popular. Data for Progress has done a great deal of polling and message testing. They make a convincing case that these elements are not just effective, but popular. As important as land use is to addressing climate change, I can understand how it’s not the best item to lead with.
  • Lead with strong messaging. Support among technocrats and wonks is required to execute any idea, but the technocrats are often bad at messaging and not a natural fit to build a successful coalition (Jeff Tumlin has made this point regarding road tolling and congestion pricing)
  • It’s better than anything else on the table. The proposal, as it stands, is light years better than anything else anywhere close to the agenda of anyone in power.

Still, the critiques aren’t wrong, per se.

  • Nothing on land use. Slipping in a plank to abolish single-family zoning might be unpopular (Minneapolis’ recent planning efforts aside), particularly at the national scale.
  • Transportation should be a bigger focus: Alon Levy made a persuasive case here for why the GND must focus on transportation. And, naturally, lots of transportation elements are indeed quite popular – and could be framed to emphasize that popularity.
    • Stronger language for transit and safety mandates (something simple yet radical, like vision zero) could be a more popular way to frame the trade-offs required to meet these aggressive goals.
  • Nothing is free in this world. Leading with a popular message framework for something as big as a new New Deal is by definition incomplete; a first step. But there’s a risk of politicians skipping over the trade-offs required to implement the plan. This might be premature, but something that needs to stay on the radar.

Given how early in the process the GND concept is, we should all give the benefit of the doubt, particularly given Data for Progress’s efforts on polling and messaging. But decision-makers still have to grasp the trade-offs involve.

As an example, read Alissa Walker on California, electric cars, and the disconnect between the Air Resources Board and the state’s Transportation Commission. Both bodies are charged with addressing climate change, but they operate in silos. The Transportation Commission has assumed electric cars will do the trick, while the CARB has done the math, and shown conclusively that electric cars are not enough – the state needs to drive less.

This is the big concern: setting a big goal is vitally important, both because of the scale of the problem and because of the potential motivation for a radical change. But radical change will require trade-offs, and it doesn’t help to mislead the public about the nature of the trade-offs involved – just look at the omnishambles that is Brexit.

702,445 – DC’s population reaches heights not seen since 1975

In case you were wondering what the hex code color is for 702,445, here it is.

While the pace of DC’s population growth has slowed a bit in the last year, the city nonetheless officially surpassed a big milestone this week. According to the state-level estimates from the US Census Bureau, DC topped 700,000 residents for the first time since 1975.

Last year’s estimates meant the city was close to this mark; the city even celebrated the (estimated) birth of the 700,000th resident back in February 2018.

Milestones like this are a good time to step back and look at the broader context:

  • 700,000 is still 200,000 below DC’s all time peak population
  • The current level is more like 100,000 below DC’s sustained peak population level (absent war-time restrictions)
  • The pace of growth is impressive, but still slower than historic rates when DC had greenfield growth opportunities within the District limits
  • Population growth will continue due to the number of units already under construction, with approximately 15,600 units under construction right now.

On the other hand, it’s worth remembering how small DC remains. Brooklyn alone has a similar land area to the District (71 square miles to DC’s 61 square miles) housing nearly 4x as many people:

https://twitter.com/profschleich/status/1075504407431798786?s=21

Likewise, the new estimates put DC’s population density at approximately 11,500 per sq mile, still less than Chicago, Boston, Philadelphia, and any number of other cities.

Even DC’s peak population of 900,000 would only yield a population density of 14,750/sq mile – less than San Francisco, Cambridge, Jersey City, Somerville, and others.

In other words, there’s lots of room to grow.

More charts from my obsessive Metro trip tracking

After fiddling with my spreadsheet full of tracking the individual metro railcars I’ve ridden, I’ve got a few more charts to show my year-plus worth of Metro trips.

Part of the reason I didn’t have these charts before is that dealing with time as a field in Excel/Google sheets is kinda a pain. It’s not always a clear number, but I was nevertheless able to sort it out.

So, some charts:

Trip Distribution: What does my overall trip distribution look like? Surprise, surprise! It’s peak-heavy.

The red lines indicate the break points for WMATA’s fare changes. A few obvious trends emerge:

  • Most of my rides are during the peak, right around peak commuting times.
  • Most of my off-peak riders are mid-day, using Metro to attend out of office work meetings, etc.
  • My PM commute is bi-modal, often due to two separate trips as I usually do day-care pickup.
  • Very few evening trips (again, likely thanks to that day-care pickup)

Railcar Distribution: One of the other observations was the unequal distribution of railcars across the system, particularly for the 3000 series.

I made these distribution charts for each rail car series. For example, here is the 6000 series:

You can see I’ve ridden cars across the entire 6000 series fleet. I’ve ridden in two of those cars five times each. As of the creation of this chart, I’ve ridden in 97 of the 192 cars in the 6000 series.

The distribution of my rides in the 7000 series is different:

The pattern is different, due to the continual expansion of the 7000 series fleet. The lower number cars are older and have this been around longer, and with more chances for me to ride them. And the distribution reflects that (note that this chart goes up to the eventual size of the 7000 fleet, which is not yet in full service).

But if you look at the 3000 series, the pattern is different:

As you can see, I haven’t ridden many cars above number ~3150. The reason is that I seldom ride the Red Line, and most of those cars appear to be isolated on the Red Line:

(Apologies for the automatically adjusting vertical scale) Obviously, this is not a huge sample, but the only Red Line 3000-series trips I’ve taken are on the older half of the fleet.

The Red Line is the most isolated line on the system. Also, I ride it the least (and therefore am unlikely to pick up small changes to the fleet management practices).

The next big fleet milestone will be the arrival of the full set of 7000 series railcars, along with the retirement of the 5000 series. That will probably trigger the last round of shifting yard assignments for a given fleet until the arrival of the proposed 8000 series.

One year of tracking my Metro trips

Three months wasn’t enough for me, I needed to spend an entire year compulsively tracking my Metro rides. I know I’m an outlier on this, but it’s been a fun way to ‘gamify’ my commute.

Some fun facts from a year on the rails:

  • The newest car: 7547 (August 30, 2018)
  • The oldest car: 2000 (November 9, 2017)
  • Most frequent car: five trips on 5088, interestingly all of them Orange Line rides.
  • 882 total unlinked trips
  • 592 unique railcars; of which:
    • 373 I’ve ridden in once
    • 160 I’ve ridden in twice
    • 48 I’ve ridden in three times, etc…

Or, to put it in visual terms:

Screen Shot 2018-09-24 at 2.21.41 PM

Given the ever-shifting total size of the fleet (thanks to both new car deliveries and old car retirements), my best guess is that I’ve ridden at least once in 50.8% of the WMATA fleet over the past year.

Screen Shot 2018-09-24 at 2.22.00 PM

Some fun observations:

Lots of 5000 series cars are now out of service. Some reporting suggests only 62 out of the original 192 5000 series cars remain in service. I’ve recorded trips on 75 unique 5k cars, some of which are surely retired by now.

Screen Shot 2018-09-24 at 2.22.21 PM

Railcar types are not evenly distributed across the network: Of my total rides (opposed to unique rail cars), 41% are on 7000 series train, an increase from my trip share after 3 months. Some of that is surely due to 5k car retirements and ongoing 7k deliveries, but some might also be due to my changing commute and because the railcars are not evenly distributed across the entire network. 

For example, a large portion of the 3000-series fleet appears to mostly stay on the Red Line. I’ve recorded lots of trips on cars in the 3000-3150 range (none of them on the Red Line), and far fewer on 3150+.

I ride the Red Line the least often, and thus don’t often encounter those cars, and given that the Red Line is the most isolated in the network, I’m not sure how frequently those cars ‘migrate’ to other rail yards.

Midway through this year, my toddler started at a new daycare located on the Green/Yellow lines. My old commute, both to/from work as well as daycare, was located entirely along the Orange/Blue/Silver trunk. And while the 7k cars are used all over the system, riding the Green Line more often sure seems to mean more rides on 6k trains (only 9.2% of my rides in December; compared to 16% now).

Screen Shot 2018-09-24 at 2.28.35 PM

Unlinked trips by Line

Adding in more Green/Yellow trips subtly changes the fleet mix. My rides on the Green Line are almost exclusively on 6k or 7k trains. The only exception (for two trips) occurred during the concurrent Major Improvement Projects on both the Red Line and the BL/OR/SV between August 11-26 2018, which surely scrambled all sorts of fleet practices.

Screen Shot 2018-09-24 at 2.29.06 PM

Railcar share by Line; total trips by line at the top

I ride least frequently on the Red Line, but even that small sample shows a pattern of only riding 7k and 3k cars. Likewise, some of the 3000 series cars on the Red Line tend to stay there.

Screen Shot 2018-09-24 at 2.29.47 PM

The vast majority of my trips are weekday trips. My weekend use has dramatically declined. Part of that is certainly my lifestyle, pushing a stroller around. But poor weekend service with extensive track work doesn’t help.

Still, fare policy impacts my rides. Since obtaining the SelectPass, I find myself far more likely to take incremental short trips. For example, a two stop rail trip just to beat the heat instead of walking? No problem.

Methodology:

I use a simple Google form to collect the data. I only collect two bits of information via the form: the car number and line color (I do also have an open-ended text field for any notes). Submitting data via the form adds a timestamp. This helps minimize the data input.

I considered adding additional data fields, such as origin/destination station, but opted not to do so. As a result, I don’t have any information on fares, delays, most frequent stations, etc. 

I collect data on unlinked trips, so any single journey with a transfer is recorded as two unlinked trips. I’ve also (occasionally) moved cars on a single trip due to a hot car, and those trips are recorded as unlinked trips.

Also, my riding habits are not random. Aside from my regular commuting routes, I’m often riding to or from daycare with my toddler. Traveling with a stroller puts open space at a premium, which means I’m more likely to pick the 7th and 8th cars on the train when riding with the stroller. This might not matter much with the older railcars, but might skew the data a bit with the 7000 series and the A/B cars. 

693,972 – new DC population estimates

Happy New Year to the very few people who swing by this blog…

DRgKK7YWsAEN3rD

Ah, the benefits of living in a city-state – the only American city-state. You get your city’s updated population estimates from the US Census Bureau as part of the state-level estimates. And DC’s growth has continued apace, now estimated at 693,972 residents within the District in 2017.

I moved to DC a decade ago, and since then the city’s population has increased by more than 100,000 residents.

While the pace of population growth is remarkably steady (netting about 10,000 people a year) even as the pace slows, the contributors to that growth have changed dramatically. Domestic migration substantially declined after the post-recession peak. International migration steadily increased.

Within the District, apartment growth continues at an impressive pace. The impact of the additional housing supply in the face of increased demand over the past decade has finally started to appear in the asking rents and other concessions from landlords.

On a personal note, these stats aren’t just abstractions for me. 2017 was a big year for me and my family; since February, we’re contributors to that natural increase in DC’s population. Parenting sure isn’t conducive to an increase in blogging. But, since I have a string of cheap, easy to bang out blog posts (going back to 2009) celebrating DC’s population increases, I figured why not add one more?

What I learned in 3 months of obsessively tracking my WMATA trips

A few months ago, I was curious how long it would take me to ride in the same Metro car. Turns out, the answer was about a week.

I’ve since continued to log each and every Metro trip, and after three months of doing so, I have a few observations.

First, some overall stats: in that time, I’ve taken 220 unlinked trips, riding in 195 different railcars. I’ve taken the same car for two trips 23 times. The most rides for a single car is 3 – I’ve somehow managed to ride in car 3245 three times.

Assuming the sequential numbering of cars is consistent with their actual time in service, the oldest car I’ve ridden is the oldest one still in service: 2000. The newest car I’ve ridden is 7438 (WMATA’s recently hinted that cars up to 7456 have been delivered).

All together, in three months I’ve ridden about 16% of the fleet.

B8707DD9-B30A-44CE-920D-C8DCD8E55851

24B9969F-48AD-4537-8CF0-5E98CE5D63D7

For the actual rides, about ⅓ have been on the newer 7000 series, ⅓ on the combined fleet of 2000 and 3000 series cars, and the remainder split among the 5000 and 6000 series.

4D850F88-8FC2-4D8E-B8C4-1437B00883A5
Some observations:

WMATA’s fleet isn’t evenly distributed across the system, nor are my trips.

My trips are concentrated along my regular commuting route, which mostly sticks to the DC trunk line shared by the Orange, Blue, and Silver line trains.

05B58C52-EB80-427D-84BB-467B0670615C

Among just my trips on OR/SV/BL, however, the distribution of trains isn’t exactly even (even if that’s what the schedule suggests). The death of the Blue Line has been greatly exaggerated!

DC0BBC80-1E6E-48C2-8DF9-0BEA93299EC7

Since these are my most heavily used lines, it helps to show how the trains are unevenly distributed through the system. I seldom rode in older, 2000 series trains on the Orange line, despite lots of those trains on the Blue line.

I’d hypothesize that the 2000-series are based in a particular rail yard (probably the Alexandria yard). Likewise, the relative lack of 6000 series trains on OR/SV/BL is probably because the 6000 series are based out of Greenbelt (it’s a small sample, but 4 out of 7 Green Line trips were on 6000 cars; the remainder were on 7000 series trains).

I still have a long way to go to catch Matt Johnson.

Getting my kicks on I-66

The Virginia Department of Transportation started tolling Interstate 66 this week, and boy, people are pissed.

A few thoughts after two weeks of dynamic tolling:

Lots of people were cheating the old HOV rules: The shock over paying the new tolls (which wouldn’t apply for those who were driving the road in compliance with the HOV rules) shows how many people had been cheating the system.

Some drivers feel wronged by changing policies, such as the increased HOV hours and the loss of the HOV exemption for hybrid vehicles, but those folks clearly don’t account for the huge portion of cars driving the tolls up.

Faster may not be the most efficient: VDOT was boasting that the average morning rush-hour speed was 57 mph. The speed limit is just 55 mph; and the old average was something like 37 mph.

The $40 toll isn’t the cost of reducing congestion; instead, it shows the marginal cost of keeping an urban expressway flowing at rural traffic densities, enabling free-flow conditions at the speed limit. It’s a useful reminder for all drivers of just how expensive it is to ensure the kind of speeds too many of them expect.

It’s not clear at all that this is the optimal policy for VDOT to set. They haven’t released any information about changes in traffic volume. The law requires maintaining average speeds of at least 45 mph; if VDOT were to accept lower speeds in the morning without introducing congestion, they might avoid some of the more exorbitant tolls and allow for more drivers to use the road.

There’s precedent for this: Minnesota experimented with a complete shutdown of their freeway ramp meters in 2000. The end result was a determination that they could dramatically improve the system’s user experience without sacrificing the main benefits. But, it took a legislatively mandated shutdown experiment to get MnDOT to make the changes in policy.

Clear communications matter: This isn’t Virginia’s first HOT lane project, nor is it the first time tolls have spiked (though most previous events could be traced to some kind of incident – poor weather, a bad crash, etc). So what explains the backlash?

A big part of the problem appears to be a misunderstanding about the toll rates. During the approval process, lots of folks had a $17 round-trip figure in their head. That’s obviously a lot less than $40 for one-way.

Fredrick Kunkle dug into this in the Washington Post:

“The bottom line is this is very different from what we briefed people it would be,” Del. John J. Bell (D-Loudoun), an opponent of tolling on I-66, told my colleague Luz Lazo.

Others have been blunter in saying the McAuliffe administration misled people. The Republican Party of Virginia accused McAuliffe’s administration of ensuring that the tolls would be switched on only after the gubernatorial election to choose his successor. Loudoun County Supervisor Ron Meyer (R-Broad Run), who is also a member of the Northern Virginia Transportation Commission (NVTC), urged the NVTC to pass a resolution demanding that the tolls be lowered or suspended.

The defense seems like an honest response, but it might as well be included in a reprinting of How to Lie with Statistics.

But Brian Coy, a spokesman for the governor, said the administration never misled anyone. He said that when transportation officials talked about a $17 average daily toll during peak hours, they meant what they said, an average — all short trips and long trips along that section of highway, and with peaks and valleys of demand.

So, that single number was averaging both the dynamic toll rates as well as the different potential routings. An average of an average. But most drives don’t take an average route, they take a specific one. And since the initial communications didn’t discuss a per-mile rate (or an estimated range of rates), it’s not hard to see why people might feel surprised.

Clear communications matter because policies like this have real promise. They depend on political support, and that will be harder and harder to find if people think they’ve been deceived. Tolling, particularly when perceived as a solution to congestion, can be a political winner. Whether the $17 expectation was intentionally misleading or not is beside the point; those expectations have to be managed or a program like this could lose support.

The challenges of adding housing in single family neighborhoods

Too often, news articles on housing prices fall into easy traps and cliché, whether in discussing gentrification or city vs. suburb tropes. But Conor Daugherty’s piece in the New York Times (The Great American Single Family Home Problem) hits all the right notes.

In it, he tells the tale of a modest redevelopment proposal to redevelop a single family home into three units on the same lot. The political opposition is fierce, leading to years of delay and legal proceedings. And this is for a parcel already zoned for additional density; this particular saga doesn’t even touch on the challenges of rezoning an area currently occupied by single-family homes.

A couple of things stand out to me:

The missing middle: The author frames the cost trade offs well. Lots of cities allow downtown and highrise development, but this requires expensive construction techniques, and thus requires pricey rents to pencil out. Smaller-scale development (low-rise apartments, duplexes, townhomes, etc) can pencil at much lower prices – the thorny issue is the politics of building in existing single-family neighborhoods.

The problem is that smaller and generally more affordable quarters like duplexes and small apartment buildings, where young families get their start, are being built at a slower rate. Such projects hold vast potential to provide lots of housing — and reduce sprawl — by adding density to the rings of neighborhoods that sit close to job centers but remain dominated by larger lots and single-family homes.

Neighborhoods in which single-family homes make up 90 percent of the housing stock account for a little over half the land mass in both the Bay Area and Los Angeles metropolitan areas, according to Issi Romem, BuildZoom’s chief economist. There are similar or higher percentages in virtually every American city, making these neighborhoods an obvious place to tackle the affordable-housing problem.

“Single-family neighborhoods are where the opportunity is, but building there is taboo,” Mr. Romem said. As long as single-family-homeowners are loath to add more housing on their blocks, he said, the economic logic will always be undone by local politics.

Capital-A Affordable, vs. affordable: The three units proposed for the lot wouldn’t be cheap, but (crucially) they’d be cheaper than a re-habbed SFH on the same lot – and there’d be more of them.

They are estimated to sell for around $1 million. But this is an illustration of the economist’s argument that more housing will lower prices. The cost of a rehabilitated single-family home in the area — which is what many of the neighbors preferred to see on the lot — runs to $1.4 million or more.

The “economist’s argument” might be sound, but it’s a hard sell for the neighbors.

This kind of evolutionary redevelopment would’ve been completely natural and non controversial before the advent of zoning.

It’s always worth remembering how different the Bay Area’s housing market dynamics are. Daniel Kay Hertz notes that many of the same issues are in play in weaker regional markets, though the way things play out is quite different:

Aaron Renn doesn’t think much of the Bay Area’s strategy of generating affordability through redevelopment of single-family housing:

https://twitter.com/urbanophile/status/937159416847175680

First, it’s hard to say this is a cogent strategy; the vast majority of single family homes aren’t going to be rezoned anytime soon.

Second, Renn is correct, historically – at least since the advent of zoning. This was true for the Bay Area, too – suburban development offered a then-cheap and cost-effective way to add housing to the region’s supply. But that was decades ago (the NYT article includes maps showing the expansion of the suburbs over the recent decades), and the region has run out of room for new/expanded suburbs within a reasonable commuting distance.

Renn’s implied regional strategy isn’t going to work well in the Bay Area, either. Consider the recent articles on Bay Area super commuters. Relying on Stockton to be San Franscisco’s bedroom community has severe costs, after all.

WMATA’s Parking Concession

WMATA is facing a budget gap – not just for this year, but a systemic and growing fiscal deficit.

Last year. the agency tested the waters for a long-term concession contract for parking services.  In this concept, an outside partner would manage WMATA’s parking for a 50 year period, in exchange for either an up-front lump sum payment or recurring annual payments.

Despite the cancellation of the solicitation, the concept is worth digging into. I’ll have more on the concession concept (and other parking ideas) in a future post. First, some thoughts about park-and-rides generally:

In general, transit park and rides lose money.

The Transportation Research Board estimated the cost recovery of park and ride facilities for various transit agencies. Streetsblog’s summary shows most agencies lose money operating these facilities, and none come close to recovering the costs of building the facilities in the first place. TRB estimates that WMATA is among the best transit agencies in recovering costs, but still doesn’t break even. WMATA brings in about $45 million in annual parking revenue, enough to cover only 66% of the estimated parking costs for the agency.

Because the TRB did not have access to actual operating cost data, their cost recovery estimate is based on a series of assumptions about amortized capital costs (including construction) and ongoing operating costs. The author of Let’s Go LA did a similar exercise for park and ride economics, setting up a scenario to show  a revenue-positive park and ride requires a) charging parking fees, and b) filling the parking lot/garage every day with paid parkers (and fare-paying riders):

As one might expect, free parking loses money for the agency. Since the service cost is greater than the fare, the cost of building the parking is entirely a loss. If the agency can charge a modest amount for parking, in this example $3, the surface lot turns into a little bit of a money-maker. $298k/year is not a huge amount of money, but it’s something, and this option actually performs better financially than the single-family housing or townhouse options.

Due to high capital costs, a parking garage can be either a big winner or a big loser. If the agency can charge $5 for garage parking, the result is a loss of over $8m/year, but if it can charge $10, the result is almost $4m/year in profit, by far the best option. Note, however, that this is dependent on the ability to consistently fill a nearly 1100-space parking garage at $10/day. There are some locations where this will pencil out, towards the edges of the city and some commuter rail stops. (People might pay $10 to park downtown, but then they won’t even bother to ride transit, which is sort of self-defeating from a transportation and land use policy perspective.)

It’s worth noting that transit agencies have competing and conflicting priorities here. Often, the goals aren’t explicitly stated. If the goal is to maximize ridership, free parking could make sense. For a budget-constrained transit agency like WMATA, however, maximizing overall revenue means a different approach.

Parking isn’t the only option:

The main point of Let’s Go LA’s hypothetical is to tackle the opportunity cost of parking – what about the potential for other investments, such as housing? What potential does real estate development have for improving ridership and improving the transit operator’s bottom line? Let’s Go LA concludes (emphasis added):

[P]lease note that this is a very rudimentary analysis and does not account for benefits and impacts to other policy goals. For example, a 5445-space parking garage might be a winner for the agency, but if it’s not located close to a freeway, it may cause a lot of neighborhood congestion. Building housing creates the opportunity for more people to live in the city, while building parking only creates the opportunity to live somewhere else and drive. And of course, parking lots and garages create border vacuums and dead zones in the city fabric, which is undesirable.

Bottom line: park and ride lots may make sense in suburban and exurban areas if parking fees are enough to cover the cost of lot construction and help subsidize transit operations. Otherwise, build more housing.

While the specifics may vary depending on the variables, it’s hard to argue with the conclusion. One potential catch: the market must support building more housing at that particular location. Here, specific agency policies might discourage the redevelopment of surface parking lots. Requiring that any redevelopment replace the existing parking is

TRB’s survey of transit agencies found several (including WMATA) with a policy to replace all parking spaces lost to development on a 1-1 basis. While this may still be the official policy on the books, WMATA has been flexible recently and explicitly embraced the idea of additional real estate development at stations increasing overall revenue for the system.

For a large joint development project at the New Carrollton station, WMATA will waive their 1-1 replacement requirement, thanks to a large number of persistently vacant spaces. The staff report on the project contains analysis of the parking usage at the station. Of the 5,025 parking spaces, 18% are consistently vacant, and another 16% are used by non-transit users paying a higher daily rate.

WMATA’s parking payment uses the same SmarTrip card, and thus can tell if the same card was used for a transit trip at the station. Transit riders at New Carrollton pay $5.10 per day to park; non-riders pay $8.85 per day. It’s worth noting that even this scenario doesn’t hit the assumptions in Let’s Go LA’s garage scenario of a 100% full garage with riders paying $10/day to park.

WMATA’s current park-and-ride role: 

Across the system, approximately 26% of WMATA’s trips are park-and-rides. WMATA periodically conducts an extensive survey of riders. Thanks to the parking payment system, they can ground-proof the survey results to parking payment. This data is from the 2012 survey:

WMATA 2012 Ridership Survey; mode of station access for AM peak trips.

WMATA 2012 Ridership Survey; mode of station access for AM peak trips. Click to download a PDF.

That’s a significant part of Metro’s ridership. Overall, 75% of WMATA’s parking spaces are used on a daily basis.

Stations at the end of the line tend to be busier than parking garages at the penultimate station (indeed, one of the arguments for the New Carrollton joint development project is the plentiful parking available at Landover station). The end-of-line stations also tend to serve people driving from longer distances, while mid-line parking facilities tend to serve the local neighborhoods. Most terminal stations include large parking facilities.

The TRB paper includes a literature review of park-and-ride elasticity for price. Since so few transit agencies charge for parking, it’s hard to draw too many conclusions, but it does seem that many parkers are sensitive to price changes. A system-wide change in parking pricing might further reduce parking use at inner stations, making them more attractive for eventual redevelopment.