Thursday, November 6, 2014

Tuesday, October 28, 2014

Wednesday, September 3, 2014

Why Republicans Should Like Buses (Part 3 of 6)

Part three of David Levinson's and my series about politics and buses is here.

Part one. Part two.

Why Democrats Should Like Buses (Part 2 of 6)

David Levinson's and my series on politics and transit continues at The Transportationist. Here is part two, why Democrats should like buses.

Here is a link to part one.

Tuesday, September 2, 2014


I contributed to a series of posts about the politics of transport with David Levinson at The Transportationist. Here is part one. (Link fixed)

Wednesday, August 20, 2014

How Much Money Do You Really Save Switching to Transit?

The American Public Transportation Association released their August transit savings report, which means my social media feeds are swamped with claims that people can save large sums of money every year if they just switched to transit from owning a car. These reports are issued regularly, and I'm always surprised that they get some people so excited each time they are released. Could it be true that the average individual will save $10,064 per year if they just get rid of their car and switch to transit? That's the claim this month.

There are three ways to look at the $10,064 per year difference. First, people are really dumb and don't realize that they are just throwing that amount of money away each year. Second, the value that people get from owning a car is worth more than $10,064 more than transit annually so they are happy to pay it. Third, the APTA numbers are wrong. I don't think the first explanation is true. If it is, then what does it mean for transit policy that $10,000 per year in cash isn't enough to get people to switch? The second is plausible but will have strong income effects. The third option is probably the best one.

APTA uses the AAA guidelines for estimating auto costs. The key aspects of the AAA estimates that distort the APTA calculations are that AAA assumes a purchase of a new car for financing and depreciation, with five year terms for loan repayment and depreciation, and that the driver averages 15,000 miles per year. These are not realistic assumptions.

If you drive 15,000 miles per year you are covering a lot of ground and almost certainly doing most of it on highways. Average annual VMT in the US is now about 9,300 and has been declining since the mid-2000s. Just as a matter of practicality, it is nearly impossible to substitute a lifestyle with 15,000 annual miles driven with one where transit is used for all trips. To do so would require relocating home, work, shopping, etc. Someone who can switch from driving to transit is someone who doesn't drive much. Let's say that if you average 6,000 miles per year (estimate based on 2/3 of US average, and the average local transit trip being under five miles) you are a good candidate for getting rid of your car. Then assume (using data from page 6 of this report) that AAA's per mile operating costs of 16.3 cents the operating costs are $978 per year. Add the cost of a brand new sedan of $4,516 per year and the cost of a new car, including financing and depreciation, is $5,494, or about $460 per month. This is still higher than transit by quite a bit, but it is a lot lower than APTA claims.

It is certainly possible to own and operate a car for less than $5,500 per year, but even if that is the baseline number. For local transit you will spend $75-115 per month on an unlimited pass. If the pass is $90, then cash savings per month are $370, or $4,440 per year. This is still substantial even if still overstated. But it is far less than $10,000. It is not unreasonable to think that people value the speed and accessibility autos provide at more than $4,440 per year. They may even want more transit options but at present are acting completely rationally by not using transit.

So bully for APTA for getting people excited about transit savings with their reports, and these are advocacy reports that should be treated as such. I just wish they would no longer be reported as matters of fact. If we are to believe APTA's numbers we should ask what is so terrible about transit that people are willing to pay so much to avoid it. Transit is, of course, mostly not terrible. We know that if transit represents good value people will take it. For most people most of the time it does not represent better value than driving, however.

Tuesday, August 19, 2014

Who Should Pay for Social Fares?

King County Metro announced a low income fare policy last week. Here is a press release. While laudable, why is it the transit agency's responsibility to pay the $8 million or so annually? This situation is something I discussed a bit in the Atlantic's CityLab, where I noted that transit agencies are  forced to be social services but without just compensation. Here is what I wrote about services, though it can be applied to fare policy:
Of course, lifeline transit services exist because somebody wanted them and people need them. Transit agencies are expected to pay for these services that serve a social purpose but that strain operating budgets.

This is from the King County press release:
"Rising housing costs are leading many families to locate in lower-cost locations that may be farther away from where they work," said Mike Heinisch, executive director of South King Council of Human Services. "Providing a low-income fare is one way we can help keep the region more affordable for working families and ensure equal access to economic opportunity." 
"As a social service agency, we work with people who are in dire need of affordable public transportation to get to training classes, meet with case managers, find and get to jobs and health care appointments, as well as other important appointments," said Mahnaz Kourourian Eshetu, executive director of Refugee Women's Alliance. "The efforts of our County Executive and County staff to make the discounted transit fare widely available to people who need it the most is admirable and will have a positive effect on the County's economy while creating stronger communities. It was an honor to serve on this task force."

"King County is one of the first regions in the nation to put a low-income fare in place, helping to make sure that our bus service really is serving the whole community," said Alison Eisinger, director of Seattle-King County Coalition on Homelessness. "We can be very proud that we are putting our values into practice in this way, by taking a big step that will help advance greater equity and access to opportunity. Thousands of people, and our community as a whole, will benefit from this progressive policy."
So Human Services, the Refugee Women's Alliance and the Coalition on Homelessness are all part of a coalition supporting low fares for certain groups. While I agree that all of these groups deserve affordable travel, I don't see why these benefits should be paid directly from the transit budget. These should be paid by someone--probably King County--to the transit agency, and I realize these are somewhat one and the same.

King County has asked voters to raise taxes to pay for transit services many times over the past decade, and they expect many service cuts coming up. Expecting the transit provider to also provide social services is not sustainable unless someone directly pays for those social services. The public has a social obligation to provide access to opportunities, but this doesn't mean the mass transit operators should pay for social programs.

There are many alternatives, but the Paris compensatory indemnity program is one place to start.

Monday, August 18, 2014

How Much Walking Should You Do in a Walkable Neighborhood?

Walkability is a current planning trend that has implications for public health, traffic, land use planning, zoning, parks and pretty much everything we like about cities. All else equal, cities should favor more walkability over less walkability. Too often planners and urbanists narrowly define walkability as the ability to walk to places where they like to spend money*, and we should think more broadly about what walkable means for all types of folks, but in most cases more walking is better than less walking.

Say you live in a walkable community where you can do most of your errands on foot. How much might you walk on any given day? How much should you walk? Paul Krugman weighed in on this question on his blog. He notes that he walks about 15,000 steps per day, and this is "fairly easy to do." Since most people take about 2,000 steps per mile, he's covering about 7.5 miles daily. This is hard to do!

I posit that if you are walking 7.5 miles daily to go about your business you are not in a walkable community. Mabe farmers and ranchers cover that kind of ground, but at that distance your daily activities are too spread out for easy walking. That distance covered daily is getting close to the hunter-gather period where humans walked 3,000-4,000 miles annually.

An realistic pace for walking is about three miles per hour. This means that 7.5 miles is about 2.5 hours of travel per day, not including subway, bus, taxi, etc. It is certainly doable, but not simply by going about your business. You have to want to do it, and be able to afford to do it. Walking is a leisure good if your income is high enough. You can buy more enjoyable and slower travel and work a bit less. Most people don't have this option, however.

But how much walking is the right amount for a walkable community? I'm not sure, but one place to start is with established travel time budgets of around 90 minutes of total travel per day. This suggests walking about 4.5 miles daily, which is less than the doctor recommended 10,000 steps.  I'm not sure this is right or wrong (it is probably too high as carrying things while walking or traveling with children limits how far and how fast you can go), but a good place to start when evaluating walkable communities. People should be able to take care of their daily business by walking less than five miles per day.  If you use transit and live and work within 1/2 mile of your stops, then your commute includes two miles of walking. As the average person takes less than four trips per day total (Table 3 in this summary of NHTS data, and I realize that national data may not accurately reflect trip totals for denser areas) it seems unlikely that a couple more trips will lead to a few more miles, unless your destination are really far apart.

In sum, to make a community walkable it must have lots of things close together. The closer together these things are, the fewer miles traveled it takes to get to them. So the most walkable places are those where people can access the most while walking the least. If you are walking 2.5 miles per day for all of your activities--which is about what I average about forty blocks north of where Paul Krugman lives (somewhere on the Upper West Side as I understand)--then walking places is easy.

Since each mile is associated with about 100 calories burned, we should also temper our expectations that the marginal increase in walking will offset the increase in calories recently seen. Five miles daily is not enough to offset fancy coffee drinks, sports drinks or even eating out a couple of times per week.

So walking is good and we should encourage more of it through design, mixed uses and street redesign. We should also keep in mind that too much walking suggests that our communities are not actually that walkable and people walk a lot because they can. To be truly walkable people should be able to reach all of their destinations without walking far. Ultimately, however, if we really want people to walk more the best thing to do is get everybody a dog.

*Central Park is one of the best places in Manhattan to walk, but for the most part you just walk about and enjoy your company, the trees, turtles, an occasional bird, etc. Central Park is not a great place to walk because you want to have lunch at the Boathouse. Fancy people take a shuttle there.

Tuesday, August 12, 2014

This Week in Dope Smoking and Transport

It is a big week for responsible transportation choices for dope smokers. First, medicinal marijuana patients sued the city of San Diego for zoning regulations that forced them to drive really far to reach the dispensaries:
Medical marijuana smokers in San Diego say the city has forced their pot shops to locate in remote areas and that means the drives to and from will increase air pollution — and ultimately, harm their lungs.
Then, in San Francisco of all places, Eaze is being scrutinized for delivering pot in under 20 minutes. Here is a story. A snippet:
As The Chronicle reported recently, there's a new San Francisco startup called Eaze that bills itself as the "Uber of pot" because it allows medical marijuana patients to use their smartphones to order pot and have it delivered by people driving their own vehicles. No word yet on whether there will be big fluffy green marijuana leaves on the cars' grilles to identify them.
I am pretty sure these are the first two legitimate dope related transport issues that do not involve impaired driving or smuggling.

Thursday, August 7, 2014

Who Blames Roger Rabbit?

Mixed traffic streetcars are all the rage right now. Just in the past week Tuscon, Arizona started operations and Washington, D.C. started training for new lines. Dozens of other cities are planning or proposing similar systems. These are almost all bad projects because the trains operate in mixed traffic. The D.C. system, while on it's first day, highlights some of the problems:
On a day when the District wanted to show how four streetcars operating together for the first time on H Street Northeast would blend with the usual traffic flow, the system encountered the kinds of problems that have raised questions about whether streetcars will be able to efficiently move people once passenger service begins this year.
Starting at the intersection of H and 3rd Streets Northeast, the first streetcar, piloted by D.C. native Saundra Harrison, lurched into traffic on its fixed track shortly after 10 a.m. 
“Today I guess there are just more people watching,” said an excited Harrison, who has been operating the streetcars on the isolated Anacostia test track.

Her excitement was short-lived. She maneuvered the hulking red and gray vehicle just a few blocks before she had to stop. A fire truck and ambulance were blocking the tracks in the heart of the H Street commercial district, tending to an injured pedestrian.
 Of course keeping a clear lane is a matter of enforcement, but I'm starting to think that Roger Rabbit is to blame for such blind support for mixed traffic streetcars, and that's a shame.

Who Framed Roger Rabbit? was a movie that helped popularize the myth that there was a grand conspiracy, masterminded by General Motors, to demolish the streetcar systems of the U.S., in particular the Red Cars (error fixed) in Los Angeles. This conspiracy myth has been popularized for decades. It is not true. It follows that if the main reason streetcars disappeared is because of nefarious action on the part of an auto company, then simply building new streetcar systems will help regain what was lost. This if/then scenario assumes away operational deficiencies that are the true reason streetcars failed. Which leads to today's mixed traffic systems.

If the dominant narrative about the decline of streetcars focused on operational problems (including not enough fare revenue to support maintenance) I doubt there would be the enthusiasm for mixed traffic streetcars we see today, and perhaps we could focus on more important public investment for transit. We should blame Roger Rabbit.

Tuesday, August 5, 2014

Are There National Transportation Priorities?

The Associated Press surveyed Americans about what they think of various taxing and spending plans for transportation. An excerpt from the piece:
Six in 10 people surveyed said the cost of good highways, railroads and airports is justified by their benefits. Among those who drive places multiple times per week, 62 percent say the benefits outweigh the costs. Among those who drive less than once a week or not at all, 55 percent say the costs are worthwhile.
Yet a majority of Americans bristle at the most commonly proposed ideas from public officials and industry. For example, 58 percent oppose raising federal gasoline taxes to fund transportation projects such as the repair, replacement or expansion of roads and bridges. Only 14 percent support an increase. And by a better than 2-to-1 margin, Americans oppose having private companies pay for the construction of new roads and bridges in exchange for the right to charge tolls. Moving to a usage tax based on how many miles a vehicle drives also draws more opposition than support — 40 percent oppose it, while 20 percent support it.
Joshua Shrank notes:
"Congress is actually reflecting what people want," said Joshua Schank, president and CEO of the Eno Center for Transportation, a transportation think tank. "People want to have a federal (transportation) program and they don't want to pay for it."
I agree that people want things and don't want to pay for them, but how do we reconcile the national attitudes with the broad local support for higher taxes for transportation projects? Here is a City Lab piece about local support for transit, and here is a Mineta Transportation Institute research project that highlights some of the factors affecting local support.

One way to think about national support for higher taxes to pay for transport is that the nation is large a diverse. If you run with particular crowds you get the notion that we, as a country, agree that transit is great, roads are the future, everybody loves bike lanes, there is no better use of money than high speed rail, we should start by fixing the stuff we have before we build anything new, etc. In truth we, as a country, don't agree on much when it comes to transportation. Transit investment is great in certain places. So are roads. We should fix the infrastructure we have first, but we should also shrink our transport networks (road and rail alike).

What we don't have in the United States is a clear national need for lots of new spending on passenger travel. Lots of transit investment is based on local economic development rather than transportation improvements, hence the new and weird "transit is supposed to be slow" defense. We don't want the federal government spending transport money on football stadiums, so I don't know why we want federal money spent on transit  just to prop up private real estate values. Lots of roads are being built simply because that's how things are done when the money flows. There isn't a national or local need for a lot of new facilities, though this obviously depends on what and where.

Transportation projects actually have a very good track record of generating local support for new taxes and spending. Partly this is because the projects reflect local preferences, to which local politicians really are responsive. We should consider that one reason, if not the main reason, national transport policy is so uncertain is that there simply aren't any truly national priorities that can build necessary coalitions of support. However, we do have lots of very important local priorities.

Thursday, July 3, 2014

From the Wayback Machine: 1992 was the year when everything changed (but not really)

I'm doing a bit of research on historical trends in transport planning, and part of this is trying to figure out the periods when we thought we figured it all out and things were going to change. Every few years we go through phases with lots of claims about how Things Are Different Now For Our Cities. Back in 1992, according to the Wall Street Journal, New Jersey was at the forefront of the nation's shift towards mass transit:
NEWARK, N.J. -- The conventional wisdom for solving the nation's transportation problems, from traffic jams, to deteriorating highways, to pollution, has always been simple: throw money at them. So why has New Jersey canceled $1.2 billion in new highway projects?

Backed by $5.65 billion in federal funds from a new six-year, $151 billion transportation bill, New Jersey officials are making an all-out effort to wean commuters from their autos and the crowded highways. Instead of spending on road projects, they have decided to gamble on mass transit, doubling the state's investments to $580 million in the fiscal year beginning July 1. They plan to tie the state together with low-pollution rail systems-and hope that tens of thousands of commuters use it. And if commuters don't, they'll find the highways even more crowded due to the lack of spending.

Despite Americans' longstanding love affair with the car, some say New Jersey is showing the way to a nation increasingly fed up with traffic and pollution. "We are watching New Jersey closely," says A. Ray Chamberlain, executive director of Colorado's Department of Transportation.
The story goes on to note that Colorado had just cancelled a large road project and did build some bike infrastructure.  There was also a description of the "life style changes" that were occurring and a new penchant for some ridesharing services:
Nonetheless, New Jersey's optimistic planners contend that the changes in its transportation policy will bring big changes in life styles. More people will share rides to work. Others will become telecommuters, linked to their offices through computers and telephone lines. And as more people use improved mass transit, suburban families may be able to shed their second cars.
The planners cite life-style changes in Portland, Ore., which froze the number of parking spaces downtown and built a light rail line. Since opening in 1986, the rail line has attracted more than $800 million of office, retail and residential development near train stations. "A lot of people are riding transit to downtown, and they are coming downtown not just for work but also for shopping and recreation," says Keith Bartholomew, staff attorney for a nonprofit land-conservation organization in Portland.
Lawrence Dahms, executive director of the Metropolitan Transportation Commission of the San Francisco Bay Area, also points to life-style changes. He says that in the mornings, commuters now line up along streets in Oakland and Berkeley to get rides to downtown San Francisco. By teaming up, the drivers and their passengers can use the high-occupancy vehicle lane approaching the Bay Bridge and save about 25 minutes compared with motorists driving alone. Mr. Dahms also says new Amtrak train service between San Jose and Sacramento has caught on fast, with ridership far above expectations.

These types of stories are compelling, but the above WSJ story could be written today almost verbatim, and there is no shortage of other similar stories from other years. For whatever meager gains we have realized toward transit, walking, cycling and shared vehicles, we haven't gotten very far for the amount of money and effort expended.  


Thursday, June 26, 2014

The Social Contract for Public Transit

I have a piece in the Atlantic's City Lab arguing that public transit is not meeting its social contract. As I conclude:
So does public transit serve its social obligations? Increasingly the answer is no. The way transit is financed in the United States distorts investment and operating priorities away from those who rely on transit service the most. Transit agencies are also asked to provide a social safety net — offeringreduced transit fares for school kids, senior discounts, or lifeline services to underserved areas that few politicians are willing to pay for. A more relevant question is why public transit agencies are solely responsible for managing disparate social goals. It need not be this way.

Eric Goldwyn's Thoughts on Transport

Columbia Urban Planning PhD candidate (and one of my students) Eric Goldwyn has been busy saying smart things about transport in the New Yorker.

His piece on the Uber challenge to professional driving is here:

And on the politics of Vision Zero is here:

Wednesday, June 18, 2014

Celebrating Transit for Others

This post is cross-posted at

Light rail is in the news this week because of the opening of the Twin Cities' Green Line. I noticed this ribbon cutting picture in the Star Tribune's coverage:

I noticed that the people taking credit for the new line (which is what a ribbon cutting photo-op is for) are excited white people. They are not representative of transit ridership by any stretch. So I looked for other recent light rail line openings. Here is the Houston Red Line from a few months ago:

From Phoenix, Arizona a while back:

Salt Lake City's Trax extension last year:

Bayone, New Jersey:

They celebrated Dallas's DART with cake:

And then (because they were full of cake?) the guys celebrating DART made the train drive through the ribbon:

These are not systematically chosen photos. They are just the first few I could find through Google that were confirmed recent light rail openings. 

Perhaps it doesn't matter than those taking credit for new transit systems are very different from the riders who rely on the systems daily. It certainly isn't a very diverse group taking credit. I suspect it does matter, however. Here is a chart from Tom Sanchez's work on equity analysis of transportation funding (also see Moving to Equity, on which he was lead author):

The takeaway is not that all transit or transport decisions should be made by key users. Rather, the decision making and credit taking people are much more likely to be white (and wealthier) than the typical user. In terms of transit, few of those cutting ribbons are even regular transit users. Perhaps if the entire planning process reflected the communities being planned than the current public process could be reconsidered. It would at least be nice to have the communities who are supposedly benefiting from these new investments share the stage and enthusiasm with the ribbon cutters. Transport planning, especially transit planning, is not something that should be done unto others. 

Friday, June 13, 2014

Transportation in Transition, Again and Always

In 1982 Milton Pikarsky and Christine Johnson published a paper titled "American Transportation in Transition."

Here are the opening paragraphs:
Today, the United States is in a transportation crisis which is of a chronic nature. It may not be sudden like the gas lines of 1974, but it is consistent and the pressures of this crisis are deepening. And because its gradual nature allows people and institutions to adjust, the crisis changes the transportation system more fundamentally than transitory gas lines or transit stop pages.

The current picture of public transportation is bleak. Indeed, each new wrinkle in the financial problems faced by the transit industry brings warnings that a breakdown of public transportation service could initiate a domino effect resulting in an urban economic collapse.
And the conclusion:
Private citizens will have to adjust to the fact that traditional transportation is likely to cost more. To reduce some of those costs, they may have to become vanpool riders or drivers, participate in neighbourhood auto mobile cooperatives, or occasionally rent automobiles or use taxis as alternatives to purchasing second cars. A variety of private transportation providers may once again become party to the transportation social contract. There is evidence that developers, too, may become party to the contract. In an attempt to make their suburban residential and commercial space more attractive, many developers are underwriting bus or shuttle services or arranging van- and carpools.  
Given the position and needs of the various principal actors, it is likely that private employers and providers will become much more involved with the direct provision of surface transportation in the future. In the best and worst of extremes, an individual could face a variety of options and a maze of prices depending on the mode, time of travel, destination, and the number of people travelling. The solution to these new transportation problems may define the future role of the public sector. Rather than owning and operating systems, the public sector may become more of a travel information broker, a facilitator, a technical adviser, and a manager of a set of service contracts.  
There is little question that the process of renegotiating the transportation social contract has begun. Each party is slowly exploring and carving out a new niche. The process will be long and progress slow. We feel certain that at the outcome, when we speak of public transportation, our concept will have grown to include a range of services and providers; rapid rail, bus, vanpools, commuter clubs, subscription services, taxis, jitneys, apartment shuttle, the private automobile, and the rental auto mobile, each serving the trip length, type, and density that is most cost-efficient. 
This piece is over 30 years old. I agree with just about all of it, and I argue many of the same things today. With all of the excitement about ridesharing and transportation network companies it is worthwhile keeping in mind such services are neither revolutionary or new ideas. Perhaps they are finally here to stay, or perhaps not. (I suspect they are, likely with different companies than exist now, but I'll save my reasoning for another time.)

Public transit's demise has been predicted for a long time, as well, and transit operates in a world of permanent financial crisis. Transit finance may not be ideal, but our transit systems have survived and many have improved. Overall, though, transportation is still in transition, and we are still expecting the next big thing to show up.

Tuesday, June 3, 2014

Two Views on Taxi Regulations

Taxi regulations are big news these days. In many ways it seems that one of the challenges facing taxi regulators and the public is that it is not clear what taxi regulations are supposed to accomplish. Here are two op-ed pieces, one from taxi industry insiders in the Twin Cities, and one from a graduate student in public policy in Vancouver.

The Twin Cities piece is about entry into the market for Uber and Lyft, while the Vancouver piece is about liberalizing entry into the market. Effectively these op-eds are about the rules that regulate entry and who should be allowed to operate. They are also about maintaining license holder value. Both sides argue they are in the right, but they both cannot be. It is also possible that both sides are wrong. Resolving these regulatory issues is much harder than most people seem to believe.

Friday, May 30, 2014

An Easy Explanation for Why Fixed Doesn't Fix as Many Parking Tickets as Expected

The tech company Fixed promises to be the easiest way to fix a parking ticket. The LA Times (reported by Government Technology, link here) did some analysis of how well the company performs as found that Fixed clients are successful 20% of the time having their tickets dismissed. This is better than zero percent but less than the 28% of tickets dismissed by regular (meaning non-Fixed client) people. The founder of Fixed takes umbrage with this discrepancy:
When presented with the numbers by The Times, company founder David Hegarty accused the San Francisco Municipal Transportation Agency of possibly "willfully discriminating against our contests."
"To rub salt on our wounds," Hegarty said, "when they deny our contests, they do not include a reason for denial."
Perhaps there is discrimination about the tech clients, but I doubt it. Here is the SF MTA response from the story:
A spokesman for the San Francisco Municipal Transportation Agency denied Hegarty's allegations, saying, "We do not have concerns if people want to use this third-party service.... There is no secret to overturning a citation. If there is a valid reason to dismiss, then that citation gets dismissed."
What is almost certainly happening is that David Hegarty has a poor grasp of economic incentives, at least as far as the Fixed pricing structure goes. Fixed clients are almost certainly not representative of the ticketed public overall. The way Fixed makes money is to take 25% of the savings for a ticket that gets dismissed, but does not charge for a failed contestation. Effectively there is no cost to whoever received the ticket for contesting the violation through Fixed, and Fixed clients will contest tickets they never would have before. Because of the high relative cost of contesting a ticket yourself people who contest in person should have a higher rate of dismissal. They likely believe in their innocence more than a Fixed client. If you take time to sit through traffic court to contest a ticket, you are either burdened by the fine or you really believe you shouldn't have been cited. Contra this, there are few reasons not to contest each and every ticket through Fixed as the only costs to the client are the time it takes to take a photo and fill out an online form. Even if you know full well you're guilty you should contest your parking tickets through Fixed. If you can get one of five dismissed you are doing well gaming the system.

Thursday, May 1, 2014

More Notes on Planning Research: Always Include Your Sample Size, and Stop It with Pie Charts

The American Planning Association just released new survey data about what people want from their communities. The link is here. This is an update to their 2012 survey, which had some interesting claims. However, the 2012 survey was poorly described and I was skeptical of the generalizability from the data. I wrote a post about it. I know the authors of the 2012 report read my previous post because they sent me emails about it. They have improved their research methods not at all in the interim.

My main complaints are that when you present survey data you must include sample sizes, and stop using pie charts. The APA loves endless sheets of pie charts without any data about sample size within each group. It is not hard to include your n. (Here is a link to the only appropriate use of a pie chart.)

The sample size matters for understanding the accuracy of the data. For the entire sample the margin of error is about 2.7 (assuming 95% CI). The data are broken into subsets, though, so we don't know how many millennials or older people are actually in the sample. Millenials are about 25% of the total population, so if they are proportionately sampled that's about 325 people, and a MoE of 5.4% for their responses. That MoE changes the interpretation of the data quite a bit by introducing much more uncertainty of the claims. Uncertainty doesn't lend itself easily to infographics, though.

This Transportation for America survey from a couple weeks ago suffers from similar sample size issues, but at least presents the data in a way that the reader can assess the veracity, and they don't have noxious pie charts. That doesn't stop reporters from gleaning far too much insight from the data. See here, here, here, or just find one of the many other examples. Again with the MoE, though, is that the data sheet reports the MoE as 3.7, yet that is for the full sample of 703, not the subgroups created. As the survey collected 70 responses from 10 different cities the MoE is actually much larger for the data as presented. For any given city the MoE is 11.7% at 95% CI, so the entirety of the data should be used with caution when analyzing subgroups.

So below is what I wrote two years ago, and it stands for these reports again. Analysis using descriptive data can be very powerful if done well, and the difference between doing it well and not doing it well isn't that big.
Reports like this bother me in part because I teach planning research courses and would be distraught if any of my students turned in a report of this quality (without additional explanation, anyway). But the larger issue is that low quality research--whether it confirms or opposes your personal preferences--reduces the signal to noise ratio. Reports like "Planning in America" are noise that cloud our ability to understand critical issues and policy (the signal in this case). At the very least the full methodology should be explained, pie charts jettisoned and sample sizes included in tables and graphs. As for planning research, reports like this are why I argue planning education should focus primarily on numerical literacy and well-crafted basic research with descriptive statistics rather than advanced regression analysis. We should train planners to communicate with data rather than claim to be psuedo-econometricians. Many of the greatest failures of planning can be directly attributed to planners' inability to understand the fundamentals of quantitative data. (See here for an explanation of the most egregious example.) Reports like "Planning in America" make the situation worse, at least as currently presented. Let's not get excited about the claims made in it.

Friday, March 21, 2014

There Isn't a Surge in Transit Ridership

Mike Smart, Mike Manville and I wrote an op-ed in today's Washington Post.  The recent APTA ridership report claims that transit ridership is surging, and we argue that this is simply not true:
But the association’s numbers are deceptive, and this interpretation is wrong. We are strong supporters of public transportation, but misguided optimism about transit’s resurgence helps neither transit users nor the larger traveling public. Transit trips did rise between 2008 and 2013. But so did the U.S. population, from 304 million to 316 million, as did the total number of trips made. Simple division suggests that, if anything, transit use fell between 2008 and 2013, from about 35 trips per person annually to 34. Many numbers look impressive without denominators, but anyone who examines transit use as a rate — whether as trips per person or share of total travel — will find that transit is a small and stagnant part of the transportation system.
We argue:
So there is no national transit boom. Why does this matter? The U.S. transportation system is deeply troubled. The country has difficulty financing improvements to its aging infrastructure, and heavy reliance on driving creates congestion, increases carbon missions, pollutes our communities, and is a leading cause of injury and death. No one should pretend these problems are spontaneously solving themselves because Americans have decided en masse to ride transit instead of driving. 
Nor should we misdiagnose problems caused by too much driving as problems caused by too little transit. Building transit systems is not the same as having people ride them, and people riding transit more is not the same as people driving less (plenty of transit riders are people who used to walk). Additionally, transit is not the only viable alternative to using a car. The environment is helped when drivers switch to buses but also when drivers switch to bikes.
Do read the whole thing for a more complete argument. I actually find the recent transit ridership statistics deeply troubling and suggestive that transit might be losing core riders at the expense of system expansions. This, of course, needs additional research, but we should not think that our current approach to transit is working well. We should be outraged at how little actual effect the billions and billions and billions of dollars of investment have produced. We can and should do better with our public transit investment.

Thursday, March 20, 2014

Can Move NY Overcome Credible Commitment Problems?

Eric Jaffe writes about Move NY* at the Atlantic Cities. In the piece he highlights some work that Mike Manville and I did on credible commitment:
Some transportation experts worry that a pricing plan won't even advance to the point of debating the economic and equity questions. Over the years some notable pricing plans — Hong Kong in the 1980s, Edinburgh in 2005 — failed to get off the ground because residents lacked faith in the funding agency to manage the new revenue. Scholars Michael Manville of Cornell and David King of Columbia call this the "credible commitment" problem of congestion pricing. A few years back, Manville and King interviewed 50-some officials in Los Angeles about road pricing. About a third explicitly said they would not support a congestion plan because they didn't trust state officials to redistribute the toll revenue as promised, the same share who feared that pricing might be unfair to the poor.

"What happens is, absent that trust, that sort of revenue promise doesn't necessarily lead to the kind of political support you might think," says Manville. "In some ways, what people were saying is it would never get far enough for pricing's regressivity to be a problem, because we would just never see this money." 
Considering the emphasis Move NY's plan places on revenue redistribution, not to mention the MTA's own mixed record of public promises, those findings give reason for pause. The way around the commitment problem, says Manville, is to stress the traffic benefits that a strong pricing plan will bring, as opposed to the revenue gains. Some believe the best way forward is to run a pilot project first, as officials did in Stockholm.

The Atlantic Cities piece nicely describes many of the problems and opportunities for the Move NY plan. Revenue distribution sounds good for assembling coalitions, but these opportunities are constrained by trust. Hopefully such concerns can be overcome.

In earlier work Mike, Donald Shoup and I looked at the politics of congestion pricing, short and ungated version available here.

*Just as a point of disclosure I have informally and infrequently consulted with Move NY on their plans.

Monday, February 17, 2014

Santiago and Transport Innovation

A couple of weeks ago I spent a few days in Santiago, Chile. This is a really underrated city that should be much higher on people’s lists to visit. The city is in the central valley of the country and is surrounded by mountains. The layout is fairly flat and easy to walk around. In some ways it reminds me of bits and pieces of California, which to my mind is a good thing. A couple of standout features include the La Vega Marquette, which is one of the “world’s best markets” according to a few travel guides. Another unique piece are the "cafes with legs" downtown that have servers dressed like they are going out to nightclubs. It’s quite a scene.

Beyond the regular cultural attractions as to why someone might go to Santiago, such as the castle in the middle of town, the city has a really interesting group of transportation policies that are of note. It even seems that Santiago has a willingness to experiment in ways that other cities haven’t.

In certain circles of which I may or may not run, Santiago is best known for their bus service experimentation where drivers were paid under one of two systems. They were either paid a fixed wage, or they were paid by the total number of passengers transported. This is innovation! It also may not be the best approach. According to this paper passenger dwell times decreased as bunching declined, but drivers drove much more aggressively and caused many more crashes. The lesson is that bus drivers should not be incentivized to pick up as many passengers as possible as this leads to inefficient and deadly competition. These compensation policies have changed.

I was extremely impressed with the downtown commercial area, which has converted (? I think converted but maybe they were always this way. In any event the paseos of Santiago are well known.) all streets to pedestrian streets. There are some cross-streets open to vehicle traffic. This is a great place to walk around. The cafes have standing tables where the street would be, there are other vendors and seemingly plenty of places to sit and linger. In my walking about (which was over 13 miles for the day, so I feel like I saw a nice slice of the city) I never was overwhelmed by curb cuts or space devoted to parking.  This may be a problem elsewhere in the city, but not where I was.

Santiago also has these awesome running man countdown clocks, which encourage RUNNING before time expires. Sort of like a video game. I watched cycle after cycle, but it does seem that you die if you don’t make it across the street in time.

I can’t say that the interior mall just off the centro was as successful. It was empty and prime space was occupied by a strip club, which was bad in the sense that it occupied lots of wall space that was just empty and had no windows—no free shows. A different outdoor mall north of the centro was a fairly typical upscale outdoor mall that you might find in the US, but had lots of bike parking in the interior:

There were lots of cyclists. Most were dressed like people but a surprisingly (to me) large share were dressed in the fancy bicycling outfits so common in the US.  I was staying in a relatively upscale area, so my observations may be skewed a bit, but overall lots of bikes, and mostly mountain bikes ridden by young adults and seemingly middle class folks. It is a very good city for biking.

Santiago takes their transit seriously now. A few years ago they reorganized the transit systems and built dedicated bus lanes through the center. See this post for details about the 2007 restructuring, which was major. The bus stops are impressive with lots of useful information and a meaningful presence on the street. The streets downtown have two dedicated lanes for buses and taxis, which is really how these things should be done.  

The freeways are tolled for free flow traffic. I was only on the freeways for my taxi rides to and from the airport, but I will attest that these trips were in free flow conditions. I don’t think there are any other cities that take this approach to the entire urban freeway network, and the fact that I didn’t know this before I went suggests that not enough people are looking to Santiago for research.

In parts of the city where the freeways run parallel to the river they have been covered with parks, which are then seamlessly integrated into the park system that runs along the river. The park along the river is great and good for running, cycling, walking or whatever. The have concerts, art installations and other good stuff there, too.

As a point of interest, here is the tallest building inLatin America. I stayed nearby.

So my advice is to go to Santiago.  It is a great city that really is at the forefront of many transport policies. I look forward to working with colleagues there about their transportation issues.

Saturday, February 15, 2014

Lisa Schweitzer's Planning Ethics Syllabus

Michael Munger highlights a session in Lisa Schweitzer's Planning Ethics course that I really like: Kids Prefer Cheese: Wind-Fall Profits? Grand Game!

I thank Lisa for this as I will happily borrow her example for my Transport Economics and Finance course next fall. Planners need to pay attention to rent seeking, among other economic concepts, now more than ever.

Friday, February 14, 2014

My Curmudgeonly Take on Sneckdowns

Sneckdowns are a thing. First coined by Streetfilms’ Clarence Eckerson all the way back in 2006, now everybody is on the lookout. Some have called it the winter of the transportation nerd. The gist is that vehicle paths through the snow in streets show that vehicles don’t use the whole street for mobility most of the time. This is a clever idea presented as a collection of suggestive anecdotes. The popularity and fervor that people seem to be embracing the concept worries me that sneckdowns will lead to bad policy.

First off, however, this post is not meant to troll or anything of the sort.  I think roads are overbuilt and we damage our neighborhoods by designing for cars and trucks over people. That said, I have two main complaints about sneckdowns.
1)     To be unnecessarily jargon-y about it, sneckdowns are observational data for partial equilibrium models. Partial equilibrium is where your observed effect holds if all else is held constant. PE isn’t stable. Neither are sneckdowns because they reflect driver activity under unusual conditions, namely snow and ice. When it snows drivers behave differently than they do when conditions are ideal (hence not everything is equal). Drivers follow the tire tracks left before, for instance, and they drive slower and more cautiously. Such behavior is also true for pedestrians and cyclists! Everybody moves differently in slipperly conditions, then once desire paths are established they get followed. Here is a picture of a Harlem sidewalk after a snow but before it was cleared. You can clearly see that people follow previous footsteps. No one would argue this is evidence that the sidewalk is too wide:

Below is a picture of the Columbia campus, and most of the time people are walking all over the place, but since paths are established people use them. This does not mean that Columbia should get rid of all the space not currently used by people walking around, unless of course any new construction was for new Urban Planning faculty offices. Local roads are bigger than they need be, but sneckdown behaviors are caused by and reinforced by the weather and residual snow conditions.

2)     So my first point was really curmudgeonly, but my second point is my larger concern in that we shouldn’t argue about allocating space for modes by how much specific parts of infrastructure are used. That is an argument that bikes and pedestrians will lose. Arguing over perceived underutilized road space is what led to awful policies such as allowances for hybrids in carpool lanes, for instance.  If you observe a bus stop you might conclude that they are underused, too, as most of the time the stop is empty.  But that’s not an argument for getting rid of bus stops. Most sidewalks are underused, but that doesn’t mean we should get rid of sidewalks or make them narrower.

Sneckdowns are a clever way for looking at street space usage, but let’s not get carried away. While I think we should make pedestrians the primary focus of street and sidewalk design, arguing such a normative view from pseudo-empirical sneckdown claims will be a losing effort and may lead to even more stupid policy. There is a fine line between stupid and clever

Thursday, January 30, 2014

Atlanta's Snow Troubles: Are a Lack of Regionalism and Transit to Blame?

The Atlanta region was hit with a winter storm that covered the city in a few inches of snow and caused massive disruptions, congestion and essentially brought the region to a halt. This is unfortunate, and as many metro areas deal with a few inches of snow on a regular basis it seems that the situation in Atlanta should be avoidable. As a technical issue, sure, the technology exists to clear snow and train drivers. As a political and economic issue, the case is less clear.

There are many articles quick to blame automobility on Atlanta’s weather related troubles. Here is one, and another. Two themes stand out in these critiques. First, the Atlanta region doesn’t practice regionalism adequately, and second, because of a lack of regionalism there is inadequate transit. From the Politico piece:

If Atlanta, the region, wants to get serious about public safety, its mayors, county officials, and state officials will need to start practicing regionalism instead of paying lip service to it. And whether threatened by a dangerous pandemic, a major catastrophe, or just two inches of snow, we need to have ways to get around—and out of—the city other than by car.

Again, there are lots of regions that are fragmented politically and are auto dependent that deal with a few inches of snow on a regular basis. It is not obvious that these are the real problems here.  

Let’s say Atlanta suddenly discovers regionalism and builds lots of transit. What can be expected? Judging by other regions, probably a situation very similar to what has happened. A few rails lines at a cost of many billions of dollars will not do much to alleviate traffic. The existing transit (MARTA) was unable to maintain operations as it was. So is there anything Atlanta can do?

Consider all those pesky auto oriented metros around the country. What they have that Atlanta doesn’t are lots of snowplows with drivers, sand and salt, and plans in place to clear the roads. This is what is missing in Atlanta. From a CNN piece:
"I've been on the road for over 16 hours now. I've not seen anybody out," he said. "They've done nothing. I have seen literally hundreds of cars parked on the side of the road. I saw a lady carrying her kid in a blanket down the side of the road. I mean, people going the wrong way on major, major interstates. It's scary stuff."

This highlights the real trade off. How much should Atlanta pay for equipment they will rarely need? Also from CNN:
"We simply have never purchased the amount of equipment necessary," he said Wednesday. "Why would you in a city that gets one snow event every three years? Why would you buy 500 snowplows and salt trucks and have them sit around for 1,000 days, waiting for the next event?"

Is investing in such equipment a good deal? We can do a cost benefit analysis of this if we want, which I don’t want to right now. I will say that buying 500 snowplows and salt trucks is probably an investment of around $50 million. That is enough to keep Atlanta running in the occasional event of snow. It is also far cheaper than building new transit as a redundant system to manage the same snow events.

None of this is to say that transit in Atlanta is a bad investment. I’m sure that it is worthwhile. The point is that transit is not a cost effective or timely solution to the problems Atlanta faced this week. Even if Atlanta starts building transit now it will be a couple of decades before it makes much of a difference at a regional scale. There are far cheaper ways--such as owning plows and salt--that would have improved the lives for most people in the region rather than the small share who would have actually been able to take advantage of a hoped for transit system.