Ethics, Risk, and Junk-Touching

TSA at Link's Opening Day – Photo by Lloyd

I’ve watched the TSA Body Scanner/Pat Down fiasco with the usual mix of outrage and dark humor.  In our collective apoplexy we have made the TSA into caricatured goons to be contrasted with grandmothers and toddlers undeserving of security scrutiny. While my own politics agrees with these sentiments, I think a couple angles on this story have been missed, one of which relates to transit and the other to ethics.

First, in this new paradigm cars win. If you care about privacy and freedom of movement, it’s a good time to drive a car. While air, rail, and ferries will likely face ever-tightening security in the coming years, private vehicles will remain largely free of restriction or inspection, despite the massive precedent of car bombings.  Traveling on any major public conveyance will submit you to inconvenience, forced data collection, and surveillance.  Not so for cars.

Second, we are partly to blame. Our expectations of government reveal a fundamental contradiction. We tend to judge as utilitarians and argue as deontologists. Put differently, we tend to argue the merits of a policy via principles (ethics, rights, justice) until something bad happens, at which point our moral calculus swiftly switches to results-based judgment.  If there is a successful terrorist attack by whatever means, we fallaciously blame the government post hoc for failing to foresee and prevent the means of attack.  This tendency also works well in reverse; when there is an extended period without an attack we reward politicians without any evidence that their actions provided any meaningful deterrence.  How many times have you heard, “Bush may have been a bad President, but his actions kept us safe”?   As a result – despite our protestations – as a society we reward governmental hyperactivity and will not tolerate inaction.   This establishes overwhelming incentives for our government to enact ridiculously cautious and intrusive security policies.

If we want to win a social argument against the “surveillance society” on the basis of rights and justice, we must relax the utilitarian standards to which we hold our governments.  As difficult as it is for our litigious society, we must accept risk, period.  We must recognize that beyond a certain point risk-avoidance diminishes the quality of that which it seeks to protect, namely a free and mobile life.  Attacks will happen, and I may die, but I would rather face that infinitesimal particular risk than assent to a general loss of liberty.  The best formulation I have seen comes from Swedish philosopher Sven Ove Hansson in his article Ethical Criteria of Risk Acceptance,

“Exposure of a person to risk is acceptable if and only if this exposure is part of an equitable social system of risk-taking that works to her advantage.”

If only President Obama had meant what he said on Inauguration Day,

“We reject as false the choice between our liberty and our ideals…those ideals still light the world, and we will not give them up for expedience sake.”

Headsigns on Snow Routes

Photo by Oran

I think most people (myself included) would applaud Metro’s handling of our most recent snow/icestorm.  Communication has been frequent and detailed, operators have been helpful and upbeat, and there has been an overall sense that Metro has been competently performing a public service under difficult operating conditions.  Our other agencies have done pretty well, too.

Yet given the extensive contingency planning for just these types of events over the past 2 years, I have been disappointed that one aspect of snow routing has gone overlooked:  headsigns. Despite ubiquitous reroutes, all headsigns that I have seen retained their usual designations.  Two examples from yesterday:  Route 2 read its usual “Madrona Park – via E Union”, despite the fact that it traveled to neither Madrona Park nor along E Union.  Rather, eastbound it was traveling to 34th/Union via Jefferson and Cherry (the same as Route 3), while westbound it traveled its regular snow route.  Secondly, every other Route 3 read its usual “First Hill” despite the fact that Metro’s website prominently warned “Route 3 is not serving First Hill.”

I don’t know how labor intensive it is to program new headsign choices, but I do think it would be helpful to have accurate displays on snow days.  A brief and incomplete(!) brainstormed list:

  • 1 – West Queen Anne via Kinnear
  • 2 – Madrona via Int’l Dist // Lower Queen Anne via Downtown
  • 3 – Madrona via Int’l Dist // Queen Anne via Int’l Dist
  • 4 – Judkins Park via Int’l Dist // E Queen Anne via Int’l Dist
  • 8 – Seattle Center via Downtown // Rainier Beach via Downtown/Cap Hill
  • 14 – Capitol Hill via Broadway
  • 16– Northgate via E Green Lake
  • 24 – W Magnolia via 15th Ave W
  • 27 – Central District via Yesler
  • 30 – Sand Point via Montlake
  • 33 – Discovery Park via 15th Ave W

What do you think?  Would this be helpful, or add unnecessary complexity?

Rail Roundup: Hi-Railing Sounder

This is an open thread.

Is Route 4 Redundant?

Ghosts of Route 4 Past (1972 Seattle Transit) – Photo by the Author

During the recent discussions of Martin’s proposal to improve mobility in the Rainier Valley, one of the central ideas to emerge was redundancy. Transit agencies shouldn’t have routes that compete with each other for the same origin-destination pairs. Whether you agreed with Martin’s proposal or not, it came from his recognition of this type of inefficiency.  Services that directly compete with each other for ridership cannibalize their own resources while diminishing productivity, frequency, and connections.

As a postscript to that conversation I’ve often wondered about the utility of Route 4, specifically whether its southern half could be eliminated and its resources distributed elsewhere.  Though in 2007 it had the highest ridership per mile of any Metro route, I have a distinct impression that it no longer serves any unique transit market and in fact diminishes the performance of Routes 3, 8, and 48, all of which serve unique destinations.  From 3rd/James to 23rd/Jefferson, the shared 3/4 provide 7-15 minute headways until 1am.  Once the 4 turns south on 23rd, it duplicates the 48.  From its turn at Dearborn it runs in a couplet on 24th and 26th, needlessly threading the needle between 23rd (Route 48) and MLK (Route 8).  Worst of all, it inexplicably terminates 1/2 a mile from Mount Baker TC, foregoing any connectivity with Link, the Rainier Valley, and other points south.  To serve Mount Baker TC, the only missing infrastructure is either a 1/2 mile of trolley wire along MLK or a left-turn trolley switch at Walker/Rainier.

What market does this route serve that could not be better met by more frequent gridded service?  As far as I can tell, perhaps only Judkins Park to Harborview.  From its southern terminus to downtown Seattle, Route 7 and Link provide some of the most frequent service in the region.  To the Central District, Routes 8 and 48 are faster and twice as frequent.  From the Central District to downtown, Routes 3/4 provide peak 7-15 minute headways, or there is half-hourly service on Routes 14 and 27.   The 2-4 buses per hour on Route 4 could be used to provide 15-minute headways on both the 14 and 27, or to provide all-day, 7-minute service on Route 3 as far as 23rd, with perhaps every other bus using Route 4’s Queen Anne terminus.  Do we need Route 4?

Winners and Losers in Amtrak’s New Ridership Report

Empire Builder at Mukilteo – Photo by the Author


Amtrak carried a record 28.7m passengers and took in $1.7B in revenue in FY 2010.  [For a North American market-share comparison, Amtrak’s passenger volume falls between Alaska Airlines (15m) and Air Canada (30m)].  Though most media outlets have reported this as significant year-on-year growth for Amtrak (up 5.7% from FY 2009), it is more accurate to view these statistics as a successful recovery to pre-recession levels.  After all, 2010 ridership is only.015% higher than 2008, or 450 additional annual passengers.

Furthermore, system-wide statistics obscure both important regional trends and homogenize an Amtrak system that offers widely varying levels of service quality.   Far more interesting are the train-by-train breakdowns listed in the report.

Mirroring some trends in the broader economy but bucking others, ridership fell in Western Michigan (Pere Marquette), Indiana (Cardinal), and California (Capitol Corridor), but grew substantially in Eastern Michigan (which introduced longer trains), the Pacific Northwest (Cascades), the Northeast Corridor, and North Carolina (Piedmont, which added a second daily train). Below is a chart that maps annual passengers, annual growth, and revenue per passenger.

Chart by the Author

The report affirms that our trains perform very well relative to the Amtrak system as a whole.  Cascades service benefited greatly from the additional Vancouver BC service, with ridership up 13% from 2009 and 10% from 2008.  The Empire Builder and the Coast Starlight, meanwhile, have the highest ridership among long-distance trains.

News Roundup: City Council Tantrums

This is an open thread.

The Lewis-McChord Conundrum

photo from

Traffic has gone from bad to worse near Joint Base Lewis-McChord.  Over the past five months 14,000 soldiers have returned from Iraq and Afghanistan, and the base is set to expand by 50% more soldiers by 2016 (from 23,000 to 36,000), bringing the total number of soldiers, civilians, and dependents at JBLM to well over 50,000 people.  This Redmond-sized contingent of people mostly reside off-base along a narrowed stretch of I-5 (SR 512 to Nisqually)with no HOV lanes and only skeletal (at best) transit service.  Travel demand has increased markedly and tempers are short, prompting nervous press releases from WSDOT announcing “immediate actions” that amount to little more than signal timing improvements.  Everyone knew this storm was coming, but the scale of the backups seems to have caught base officials as well as WSDOT off-guard.  This may now be the single worst traffic area in Washington State.

Yet soldiers and civilians alike haven’t had many other options, and what is a ‘last mile’ problem elsewhere is a ‘last 5 miles’ problem on base, ringed by mandatory security checks.  In short, transit currently has little chance to compete for mode share. (Tacoma Tomorrow blogger Evan Siroky had an excellent post series earlier this year covering many of these same issues.)  Over the years, Pierce Transit has worked diligently with base officials to design useful routes with limited resources, yet results have been continually disappointing.  Three local routes currently enter the base, #206 (Lakewood-Tillicum-Madigan Hospital), #207 (Madigan Hospital-Ft. Lewis), and #300 (Tacoma Mall-SR512 P&R-McChord Commissary).  While #206 and #300 both exceed 1,000 riders per day and meet PT service standards, PT’s planners tell me that very few of those riders actually enter the base.  The only intra-base route (#207) is PT’s least-ridden route, averaging only 36 boardings daily for an hourly service.

Longer term, a series of infrastructure investments have been planned for some time, and WSDOT recently released a Transportation Alternatives Analysis that proposes spending over $1B on widening and new ramps for I-5.  Transit receives only a cursory mention, and the planners seem to assume that nothing cost-effective can be done to get JBLM to lower its SOV/VMT levels.  Given the dismal fiscal situation government faces at all levels, it is clear that capital-intensive projects will be difficult to fund and complete, however badly they are needed.  Yet the sheer scale of this problem — cultural, structural, fiscal — prohibits thinking small.  What should be done? What should transit advocates push for?  To get the conversation going, my initial suggestions are after the jump…  Continue reading “The Lewis-McChord Conundrum”

2nd Cascades Train to Vancouver B.C. Canceled

Pacific Central Station in Vancouver – Photo by Discovery Institute

The extension of Amtrak trains 513/516 to Vancouver B.C. will end on October 31st, indefinitely truncating the trains back to Bellingham. The Canadian Border Services Agency was unwilling to relent upon their demand for roughly $550,000USD in annual border clearance fees, and neither WSDOT nor Amtrak is willing to pay the fee.  WSDOT issued a press release criticizing the decision and has urged CBSA to reconsider.  The Bellingham Herald reports that the B.C. government wished to see the trains continue without a fee but was overruled by CBSA on fiscal grounds.  Approximately 73 people per day rode 513/516 between Vancouver and points south over the life of the extension, with the Olympics and the summer months averaging roughly 100 per day.

More after the jump…

Continue reading “2nd Cascades Train to Vancouver B.C. Canceled”

Downtown Tacoma Paid Parking Begins Today

Photo by Zach

For the first time since 1976, today the City of Tacoma begins charging for parking in the downtown core.  The city will charge an introductory rate of 75¢/hour with a 2-hour maximum, a fee significantly lower than either Seattle ($2.50) or Portland ($1.60). In addition, today the city introduces a 90-minute buffer zone adjacent to the paid parking area in which parking will remain free but will for the first time be subject to time limitations.

Rates will be effective 8a-6p Monday through Saturday, with Sundays and designated holidays free, and Tacoma will use the same coin/card meters used by the City of Seattle.  Sources at the City of Tacoma admit that the 75¢ rate will not cover administrative costs, making it likely that rates will rise once drivers have adapted to the new scenario and the city analyzes ongoing parking availability.

The change is expected to to increase parking availability in the core, markedly increase parking demand at Tacoma Dome Station (which for now will remain free parking), and boost ridership on Tacoma Link.

Ferry Panel Suggests Cyclists Disembark Last

Darrel Bryan (Clipper Vacations CEO and PVA Chairman) with Governor Gregoire, David Mosely, and Paula Hammond -- WSDOT Photo

On September 9th the Passenger Vessel Association released its recommendations report for improved operations on Washington State Ferries.  Governor Gregoire had requested that the panel conduct an informal audit and make preliminary suggestions.  While many of the suggestions are sensible, such as replacing in-state bidding for capital projects with national bids, they also included a frustrating suggestion for cyclists.  From page 88 of the report:

The Panel recommends that vehicles be unloaded ahead of bikes.
Safety is of the paramount importance with efficiency second. The Panel recommends that a trial project be undertaken to change the loading/unloading sequence with bicycles being loaded last and unloaded last. This allows better separation of vehicles and bicycles and gives the Mate more control over the space allocated to bikes. Bikes are also slower than cars and can slow the disembarkation of those they are in front of. By holding back bikes, it also avoids the need for bicyclists to move through the car deck with their bikes in order to get to the front of the vessel. By off loading after the vehicles, bikes will not be sharing the road at the same time as the disembarking vehicles, allowing for a margin of safety.

Forcing cyclists to wait an extra 10 minutes would significantly disincentivize the mode when foot passengers and vehicles would remain able to disembark immediately. Would Metro buses at Fauntleroy or Vashon wait the extra time in order to accommodate cyclists? Would the extra waiting time cause Bainbridge cyclists to switch to walking or driving? Beyond these inconveniences, cyclists would have to be on the vehicle deck breathing poorly-ventilated exhaust, posing a significant health risk and significantly diminishing the rider experience. For these and many other reasons, let’s hope that this is one suggestion they ignore.

Link Ridership Down 1.5% in August

Weekday Link ridership declined slightly in August.  The weekday average of 23,771 boardings was down 1.5% from July’s record count of 24,145.  Sunday ridership declined a significant 7% (from 17,127 to 15,893) , while Saturday ridership was up by 4% (from 22,098 to 22,979). 

While perhaps disappointing, most of us expected an August lull of just this sort.  The cruise ship and general tourist markets were on the wane, and schools had not yet begun in earnest.  Looking into the autumn, the smart money is on increased ridership in September and October to be followed by another leveling-off or decline in November.

Parsing Reliability Numbers Further

I thought I’d follow up on Martin’s post on Amtrak’s long-distance reliability numbers with some Cascades data that is highly relevant to the conversation.  Commenters expressed frustration with using mean performance as a single indicator of success, and many asked not only for Amtrak’s definition of ‘on-time’, but also to see median, mode, and ‘full-distribution’ data for Amtrak trains.  The problems with ‘mean performance’ have already been kicking around the transit blogosphere lately, and I see little reason to simply reiterate something others have already said well.  But I always find it useful to add data and visuals to a conversation.

By Amtrak’s on-time performance definitions, trains can be 10-30 minutes late (depending on route distance) and be considered ‘on-time’.  With such a wide scope, the concepts of ‘on-time’ and ‘reliability’ begin to drift apart in meaning.  Reliability should mean a consistent travel experience, both qualitatively (comfort and service) and quantitatively (speed and on-time performance).

Check out the chart above.  Using data from the invaluable Amtrak Train Status Archives, I charted the performance of two morning train segments (#510 Seattle to Vancouver BC, and #513 Vancouver BC to Seattle only) for January 1-June 30 2010.

More after the jump…

Continue reading “Parsing Reliability Numbers Further”

Ridership Modeling and Fallibility


Critics of transit investment – especially rail investment – frequently cite a failure to achieve a budgeted ridership estimate as evidence of the ineptitude or corruption of the agencies planning the lines in question.  While I never wish to discourage due criticism, ridership estimates are constructed via theoretical models, and critiquing a model for being wrong is tautological, akin to critiquing a human for being mortal. Frustrated at popular confusion over the nature of modeling, I thought I’d write a post on the limitations and capabilities of models.  To my mind there are four main points:

More after the jump…

Continue reading “Ridership Modeling and Fallibility”

Rail News Roundup: Jackson Plaza

  • Work continues on King Street Station.  The remnants of the baggage area and the Jackson St. Plaza have been demolished.  (Slideshow above, picture of the ’empty hole’ here.)
  • Body found inside the Great Northern Tunnel.
  • Amtrak Cascades ridership between Seattle and Vancouver BC set a record in July, with nearly 25,000 total passengers for the four daily trains.  Ridership is also up 21% on the original trains (510/517).  CBSA has apparently reached a decision whether or not to continue free inspection services, and will be notifying WSDOT shortly.
  • To celebrate the popularity of the service (and to put timely pressure on CBSA?), Tourism Vancouver has partnered with Amtrak to offer 25% off Cascades travel between Seattle and Vancouver BC  for all of September, along with discounts on hotels and city attractions.
  • Leveling and grading work has begun on the new 3.2 mile bypass track in Vancouver, WA.
  • WSDOT and BSNF have received approval to begin work on the Stanwood siding.  On August 17 the Army Corps of Engineers issued the necessary wetland permit, the last major environmental hurdle.
  • WSDOT has applied for $80m in additional high-speed rail grants available under the $2.3b USDOT Appropriations Act. A 20% local matching commitment was required in order to apply.
  • The new Edmonds Sounder Station groundbreaking ceremony was held on August 18.

This is an open thread.

Sound Transit to Add 277 Bicycle Parking Spaces

SeaTac/Airport Station Bike Lockers and Racks
Bike Parking at SeaTac (photo by Oran)

Sound Transit announced plans on Tuesday to construct approximately 277 additional bicycle parking spaces over the next 16 months.   Funded by federal grants, the 173 lockers and 104 rack spaces (13 racks, 8 spaces per rack) will be placed at selected park & rides, Link stations, and Sounder stations.

Cycling investments such as these are some of the easiest and most cost-effective means of combating the ‘last-mile’ problem, especially at far flung suburban park & rides.  For a cost of $464,000 (or $1,675 per space) they’re not exactly cheap, yet they cost a tiny fraction of the same number of car spaces and they leave a small physical footprint on station areas.     Anything that makes non-SOV multimodality more attractive has our support.

The locations are:

  • Auburn Station, 28 lockers
  • Columbia City Station, 7 lockers and 2 racks (2010), and 46 more lockers and 1 more rack (2011)
  • DuPont Park & Ride, 6 lockers and 1 rack
  • Federal Way Transit Center, 8 lockers
  • Kent Station, 22 lockers
  • Lynnwood Transit Center, 20 lockers and 4 racks
  • Mercer Island Park & Ride, 4 lockers
  • Othello Station, 8 lockers and 2 racks
  • Rainier Beach Station, 16 lockers
  • South Everett Freeway Station, 3 racks
  • South Hill Park & Ride, 8 lockers

Rider Alert: Another Tough Day to Take the Train


Due to a fatal accident between Puyallup and Tacoma, Sounder service is delayed over an hour and trains are being turned at Puyallup. It is unknown at this time if Cascades trains will run as scheduled to Portland.

Also, no trains will run to Vancouver BC today or tomorrow as emergency repairs in White Rock continue.

Frequency Mapping the West Coast Rail Corridor

Following up on our recent discussions of frequency mapping (see Martin’s post here and Jarrett’s post here), I thought I’d share a frequency schematic for all intercity and commuter rail service available in California, Oregon, Washington, and the BC Lower Mainland. As we push to develop our rail networks it is imperative that we strike the right balance between vision and process; with too much of the former we lose ourselves in the foaming muddle of railfans (I say this as a railfan), while with too much of the latter we succumb to the sort of stale incrementalism with which we’re all very familiar.  One of the most important requirements for speaking intelligently about our rail networks is knowing exactly what we already have, for this is what we will build upon.  To that end I share this draft map I’ve been working on.

More after the jump… Continue reading “Frequency Mapping the West Coast Rail Corridor”

Riding The Much Improved Coast Starlight

Upper Klamath Lake at Sunset

Two weeks ago I traveled from Seattle to Los Angeles aboard the Coast Starlight.   Since the dark ages of 2005-2006, during which the train was late 90+% of the time, the Starlight has done much to reclaim its status as a premier passenger train.  After a mudslide knocked out service in the winter of 2008, Amtrak ‘relaunched’ the service with a renewed focus on amenities and on-time performance, and it has worked.  In June 100% of southbound trains arrived into Los Angeles on-time, while 93% of northbound trains into Seattle arrived on-time.  My own trip confirmed this improvement.  For $220 I had an on-time arrival, wireless internet, bottomless coffee and fresh produce, simple but decent meals, tablecloth service with porcelain dinnerware and real flatware, a small but comfortable room, a hot shower, leather lounge chairs, a cinema, and a panoramic view of Mount Rainier, the Tacoma Narrows, the Nisqually River Delta, the Columbia River, the Willamette Valley, the Oregon Cascades, Mt. Shasta, the Sacramento/San Joaquin Delta, and the Central Coast of California.

Somewhere Between Lompoc and Santa Barbara

Though we often discuss how well-developed passenger trains can take modal share away from car and air travel – and Cascades has a good chance of evenly splitting the air-rail market over the coming years – long-distance trains are qualitatively different.  They offer neither the on-off freedom of cars nor the speed of airplanes.  They are often the only high-capacity transit service available across a large swath of rural America and they traverse scenic pre-Interstate corridors.  Being a functionally unique service, long-distance trains compete only against themselves and their own expectations, and they will live and die on the strength of the experience they offer. Amtrak seems to have belatedly figured this out, and the Starlight is again a wonderful experience.  It’s not the ‘Star-late’ anymore.

As a footnote, I think Seattle and Portland have largely not recognized that we have arguably the highest quality Amtrak service in the country.  We’re not a hub like Chicago, nor do we have the frequency of the Northeast or California, but we have the two best long-distance trains and a unique corridor service with Talgo equipment everyone else would love to have.  While in many ways Seattle is behind the curve, we can at least be grateful for the high quality of service we enjoy.

The Little Things That Keep Us Driving


We talk a lot on this blog about why people drive.  Frequent points are made concerning perceived freedom, the motorist’s willingness to “pay time to save money”, the undercapitalization of transit infrastructure, the low marginal cost of individual driving trips once a car is owned, the modal lock-in caused by low density development, etc, etc…

But I’ve been especially frustrated lately by 3 perverse incentives that don’t get as much press:

More after the jump… Continue reading “The Little Things That Keep Us Driving”

Link Ridership Up 7.5% in June

Average Weekday Ridership

Link ridership set another record in June, with an average weekday ridership of 23,396 (Saturday 17,510; Sunday 13,919), and a daily range of 9,827 to 28,820.  This represents a 7.5% increase over May’s average of 21,774, and a 16% increase over April’s 20,129.  June represented the sixth straight month of >5% month-over-month growth.  Even with the usual caveats regarding sampling and modeling error, this is very encouraging news.

Ridership tends to follow a logistic growth model, in which ridership grows exponentially until it slows as it approaches a theoretical saturation point.  We’ll probably see ridership stabilize at its “natural level” well before University Link restarts the growth cycle, but it will be exciting to see just what that level is.  In the meantime, even with seasonal variations, we are likely to see continued year-over-year growth.  (Just for fun, a continued 7% rate of growth would yield ridership of 35,000 by the end of the year.)