From FiveThirtyEight: Do Americans Really Hate Flying? Or Really Love Driving?, an amateur analysis of inter-city travel statistics that includes a discussion of gravity models. The comments are worth reading as well.
Interesting discussion of HSR in The Transport Politic discussing my interview posted yesterday.
But Mr. Levinson fails to address the fact that investment is needed in both intercity and intracity corridors. Claiming that we should not fund high-speed rail because urban transit is more important is equivalent to saying that federal subsidies to air travel and non-urban highways should simply end, because metropolitan areas need more investment and travel between cities is less important.
Perhaps such subsidies should end. I am not going to be a modalist and support subsidies for mode A because mode B gets them, or even suggest any mode should get subsidies (unless a good economic rationale justifies it). Transportation, especially intercity transportation, is an economic activity foremost that should justify itself on economic grounds. There are of course social aspects to this as well, but going down that path to justify multi-billion dollar systems is indeed a slippery slope.
An argument could be made about strengthening intercity linkages to refashion the current metropolitan system into a megalopolitan system, where people more regularly interact between cities. (Switzerland writ large). If the division of labor is limited by the extent of the market, and transportation can be used to expand the market, the division of labor can therefore increase (i.e. be more specialized), which should have some positive effects for the economy (akin to agglomeration economies). The magnitude of this is uncertain (and certainly location-specific), but I think presents the best case that can be made in favor of HSR in the US.
That said, remember that real HSR (not the short term improvements to get to 90 or 110 MPH, which may or may not be a good thing, but is certainly not HSR) is a long term deployment, so it needs to be compared with cars 10 or 20 or 30 years hence, and the air transportation system over the same period. Cars are getting better from both an environmental perspective and from the perspective of automation technologies. The DARPA Urban Challenge vehicles need to be bested to justify HSR. Cars driven by computers should be able to attain relatively high speeds (though certainly not HSR speeds). Further they may move less material per passenger than HSR (trains are heavy), and so may have net less environmental impact. This really waits to be seen.
The table that Mr. Freemark assembled (link above) is interesting and is a useful way to arrange a comparison (user time, negative externalities) being two of the major considerations. Operating costs and capital costs would be good to include as well. But the table is a bit harsh on aviation. Convenience is in how the system operates, and while the security theater we face now, especially in older large airports, is a disaster, it is not necessarily what we will face 20 years from now. Airports will be closer for many people (especially in multiple airport cities like SF, LA, DC, NY) than downtown train stations, though certainly not for all.
Recently published:
Xie, Feng and David Levinson (2008) The Weakest Link: A model of the decline of surface transportation networks. Transportation Research part E 44(1) 100-113. [doi]
This study explores the economic mechanisms behind the decline of a surface transportation network, based on the assumption that the decline phase is a spontaneous process driven by decentralized decisions of individual travelers and privatized links. A simulation model is developed with a degeneration process by which the weakest link is removed iteratively from the network. Experiments reveal how the economic efficiency of a network evolves during the degeneration process and suggest an "optimal" degenerated network could be derived during the decline phase in terms of maximizing total social welfare.Keywords: Decline; Transportation network; Simulation; Welfare; Accessibility
Recently published:
Iacono, Michael, David Levinson and Ahmed El-Geneidy (2008) Models of Transportation and Land Use Change: A Guide to the Territory Journal of Planning Literature 22: 323-340.
[DOI]
Modern urban regions are highly complex entities. Despite the difficulty of modeling every relevant aspect of an urban region, researchers have produced a rich variety of models dealing with interrelated processes of urban change. The most popular types of models have been those dealing with the relationship between transportation network growth and changes in land use and the location of economic activity, embodied in the concept of accessibility. This article reviews some of the more common frameworks for modeling transportation and land use change, illustrating each with some examples of operational models that have been applied to real-world settings. It then identifies new directions for future research in urban modeling and notes the important contributions of the field to date.Key Words: transportation planning • land use • mathematical models • urban growth • gravity model • microsimulation
TCRG BrownBag Speaker Series: Wednesday, August 12, 12:00 Noon to 1:00 p.m.
Topic: Destinations Count: Access to Destinations Research
Featured Speaker: David Levinson, University of Minnesota
Where: Wilder Foundation, Room 2610
451 Lexington Pkway (at University Ave)
Saint Paul, MN 55114
For a map see the TCRG website at http://www.TwinCitiesResearch.org
Event is free -- parking in Wilder's parking lot is free -- bring your own
brown-bag lunch.
Access to Destinations is a multi-year effort by University of Minnesota
researchers to analyze and model transportation networks in the context of urban
and regional economics. "The issue is what you can reach," says David Levinson,
"not how fast you can go."
Levinson presented the keynote presentation at a May 2009 conference of 300
transportation planners, engineers and researchers. He'll join us at TCRG's
August 12 Brown-Bag to share his presentation with TCRG and answer questions.
"All of the great cities have congestion. Congestion is an indicator of economic
success; the fact that roads are crowded with people going about their daily
business shows that the city is providing what they need." Creating access
induces people to travel and participate in the economy. This activity leads to
higher land values, which encourage developers to build housing and commercial
properties, creating new destinations and higher levels of accessibility.
For more background, Levinson's research is described at
http://www.cts.umn.edu/Publications/CTSReport/2009/07/AccesstoDestinations.html
TTI's 2009 Urban Mobility Report (2007 data) is now out. Minneapolis St. Paul ranks 28th on the Travel Time Index (TTI, get it?), Should I be happy we are relatively less congested or unhappy that we have lost ranking? Clearly congestion is dropping in recent years.
William Gibson's Neuromancer defines cyberspace as a "consensual illusion" obtained when a user "jacks into" the network. Plans for cities are also consensual illusions, a community agreed upon vision of how the city will look at some future date. Planners are but illusionists, creating and shaping a fantasy for a how city imagines itself, and through this consensus, harnessing the positive feedback processes of public and private investments aiming to achieve self-fulfilling prophesies. By promising networks, development will come; by promising demand, infrastructure will come.
Riding for a conference from the Portland airport to Portland State University on Light Rail Transit (LRT) and then streetcar gave me time to reflect on the Elysian Fields of transportation engineering, the Nirvana of networks and nexi.
Portland, Oregon is one of the major battlegrounds in the mode wars (car vs. transit and the internecine rail vs. bus). It has since the 1980s been held up by planners as the exemplar American city that does almost everything right. The foremost thing they do right in the view of the planning establishment is promoting LRT and bicycling.
The fascination with rail transit in particular (especially as compared with bus) was something I have never quite grokked. As a rational observer with formal training in transportation, I have had a hard time understanding the emotional relationship people have with rail. Why do people like LRT more than bus? Is it simply how we operate them, or that it is modern capital, or is there a psychological benefit associated with deterministic tracks vs. widely diverging roads? There are lots of theories on the matter, I will identify a few below.
1. Ride quality. The quality of the ride on an LRT is smoother and less herky-jerky than a bus, and passengers have a nicer facility.
2. Navigability. It is hard to navigate current US bus systems, while the fewer number of rail lines are fairly easy to figure out. Because trains cannot steer, they cannot get lost the way a bus can.
3. Speed. Trains are faster than local buses, especially if they have their own right of way and few stations.
4. Permanence. I can make a permanent investment decision based on the location of rail lines, as the transit system is committed to this line, while a bus line may be temporary.
5. Nostalgia. People who like LRT recall (or wish they could recall) the immediately post-World War II America when streetcars were at a maximum. 1946 was a magical period in US history, a boom following the long depression, when streetcar networks if not at a maximum were really close. (Coupled with a conspiracy theory about their removal)
6. Sexuality. This is part of the theory presented by Jonathan Richmond's in his book Transport of Delight and earlier paper The Mythical Conception of Rail Transit in Los Angeles. The image of the train entering the tunnel clearly evokes a primal response.
There are logical rejoinders for the first four (though not the nostalgia or sexuality argument I suppose), the most obvious is that if you spent the kind of money you are spending on rail on buses instead, and operated them better, buses would be quite nice. Navigability could be improved with a bit of thought (and trains can divert), while permanence of the last generation of streetcars (1887-1954) clearly was temporary.
The theory I have now adopted comes from my recent trip from Minneapolis to Portland accessing the airport at both ends via LRT, and then riding the Portland streetcar almost full circle. Rail transit forms an urban superstructure. Guideway transit, esp. LRT makes the city more like a single structure, and makes everything seem closer. The LRT vehicle is continuously running, and if activities are along the path of the vehicle, everything seems quite coordinated. In a way by organizing activities linearly (or multi-linearly), it simplifies the city. Hopping on a train is much like getting on an elevator.
Preferences for civilization or frontier-crossing (or degree of each) vary across individuals. Driving of course places you in a machine, but you, not civilization, are operating the machine, so just as driving is freedom, not everyone wants that freedom to drive, they may prefer freedom from driving. The extent to which you believe in the importance of community over individuals (or vice versa) will affect your perception of the issue.
( LRT may also be more popular than traditional underground subway (Metro) systems. People of course like being able to walk out the door and step onto a train more than having to descend through the gates of hell, Metro to get to the underground subterranean system. There are many reasons, not least of which is the extra time and energy required to so descend. The advantages in principle are faster point to point travel time, but that depends on the access cost vs. the in-motion speed. )
Transit invokes further passions because of the positive feedback loop between ridership, revenue, and route frequency, especially where transit is weak as in much of the US. My riding transit creates a positive externality for you (more riders, shorter headways, and more routes), so of course if you ride transit, you want to impose your preference on me. It is only selfishly rational. Further cars use scarce roadspace. While similar feedback loops may exist on the highway side (more drivers means more closely spaced roads), congestion mitigates that and the network is largely built out, so drivers do not feel the same need to impose their modal preference on the transit riding minority. Finally, drivers may benefit in the short term if other drivers take transit. (Where transit is already congested and frequent, additional riders produce few positive externalities as diminishing returns set in).
Our Value Capture for Transportation Finance study is now out.
Detailed reports will be placed online soon.
About the StudyLarge public investments in state transportation infrastructure--such as new freeway interchanges, highways, or transit stations--can increase the value of adjacent private land, sometimes substantially. Capturing the value of this benefit through various tools is gaining interest as a finance mechanism for infrastructure investments. But many questions remain: Does "value capture" promote or hinder economic development? How high should the tax rate be? How stable is the revenue?
To answer these and other questions, the state legislature appropriated funding to CTS to study the public policy implications of value capture.
Researchers reviewed the relationship between transportation and land values, including the measurement of benefits from a transportation improvement, as well as the legal and economic frameworks for capturing the value gains. They explored the major financing techniques associated with value capture--such as joint development of infrastructure and adjacent private parcels, rezoning and reselling, impact fees, special assessment districts, and tax increment financing--and some examples of their implementation. They then evaluated several of the proposed policies and their suitability for implementation locally, based on the criteria of economic efficiency, social equity, adequacy as a revenue source, and feasibility.
On June 24th, MnDOT held a "Long-Range Funding Solutions Symposium" to examine issues associated with the long-term funding of transportation. I was asked to be a discussant. These are my comments in extended form.
Thank you for giving me the opportunity to discuss the topics raised today.
First, MnDOT has identified $50 Billion of unfunded "needs" for additional resources of which 86% are for the purpose of "mobility" over the next 20 years. I am not clear as to how these needs were identified, but several points should be kept in mind. First, this is a slow-growing region (and outside the Metro a declining state). It has 5 million people now, and at best is growing at about 1 percent per year. Second, per-capita Vehicle Miles Traveled has been flat for almost a decade, and overall VMT growth has been flat for about half a decade. There are several reasons for this, most recently recession and high gas prices, but I think the most important is market saturation. if speeds are not growing (because we have maxed out the network given current technologies and face diminishing marginal returns to new road construction), and people have finite time, they choose not to devote additional time to travel (and thus distance). Fortunately, since the I-35W Bridge Collapse, MnDOT has adopted a "fix it first" approach, so that system preservation, operations, and maintenance get the largest share of the existing budget, and comprise the first funded element of needs.
We cannot know what "needs" for mobility are if we have an unpriced (or underpriced) transportation system. People will always over-consume if they are subsidized, and people do not presently pay for the congestion externality they impose on others. Once we have something like marginal cost pricing (or a second-best version thereof), we can determine which links generate more revenue than they cost to operate and maintain, and that will signal where capacity should be added, where the benefits of added capacity outweigh the costs.
Another way of thinking about what $50 billion means is that Minnesota is a state of 5 million people, so that amounts to $10000 of new construction for each resident of Minnesota (because this is above and beyond the funded part which takes care of preservation (we hope)). Over 20 years, $10000 per capita is $500 per year, or about $0.50 per trip. But that $0.50 per trip is not to pay for existing infrastructure, that is to pay for new infrastructure those travelers may or may not use; or if we were to charge users, we would be looking at 10 to 100 times as much per trip, as the new capacity built for $50 billion will serve only 10% to 1% of trips, most trips will continue to use pre-existing infrastructure.
We could also talk about mobility vs. accessibility, and why is it important to enhance mobility, but that is another long discussion, and the reader is referred to the Access to Destinations study for details.
Attention is a scarce resource, spending time on non-starters like $50 Billion in "mobility" needs detracts from real problems with existing infrastructure.
In short, the $50 Billion suggested comprises Wants not Needs. (as Jim Erkel calls it the Rolling Stones theory of transportation finance ... You can't always get what you want, but you get what you need).
Second, we need to re-examine the institutional structure of transportation funding and administration. We should consider a public utility model where a transportation authority or utility with independence from the legislation and executive branch of government determines how much is required to maintain (and as necessary expand) the transportation system, with oversight from a Public Utility Commission or similar. This would resemble how Natural Gas and Electricity and Water and Sewer in many places are currently delivered. Like those, transportation is a utility that has costs that users should bear as directly as possible. The user fee notion would be embedded into the governance structure of such a transportation authority. The British might call this a Transportation Trust. We could consider how this is organized at different levels of government (keeping state and local separate or bringing them together?)
Third, Value Capture has not been fairly characterized in the presentation made today. If we do not have road user fees, transportation creates value for land-owners. (If we do have marginal cost user fees, a closed system, and invest the revenue in transportation, making some simplifying assumptions, we would not have additional land value associated with investment (in the absence of agglomeration economies)). Since we do not have road user fees, value is created. Several of the methods proposed by the value capture study hold promise for financing transportation systematically, not just at the project level.
Fourth, in the short-term (next decade or so), gas taxes, indexed and adjusted appropriately should be used to fund transportation, as they are administratively much more efficient than road user charges. They have several advantages: foremost they are cheaper to collect than most of the proposed VMT charges. An annual odometer reading is certainly a similar alternative, but that does not have the environmental benefits of discouraging motor fuel consumption and encouraging better mileage. Ultimately as the fleet becomes electrified, the gas tax becomes a better and better incentive to move in that direction. If today 100% of the drivers use gas and pay for 100% of roads (which I recognize is not strictly the case at the state level, but is simply illustrative), and next year only 50% of drivers used gasoline, the remaining 50% would pay for all of the roads by doubling the gas tax. That provides a somewhat stronger incentive to switch to electricity. If the following year another 25% switch to electricity, than 75% use electric and 25% use fuel and pay the motor fuel tax, which is now 4 times as high. Eventually this becomes unsustainable as the last drive of a gasoline-powered car could not possibly afford 100% of the road system's costs, but in the meantime the incentive works in the right direction for the environment, and since government is always a lagging indicator, retaining the gas tax for as long as tenable should be considered the near term solution, with continuing research into road pricing, additional demonstration, and deployment of select strategies like High Occupancy Toll lanes. See Beyond the gas tax for a further discussion.
At any rate, as I have learned today, in Minnesota transit funding depends on the Motor Vehicle Sales Tax, so I will do my part to help fund transit and buy a car.
From the European Commission (via JM)
Travel time to major cities: A global map of Accessibility
My Design and Go presentation on The Importance of Being Accessible is now posted.
From the Washington Post via Dick Pratt, an Obituary for Walter Hansen, developer of the Hansen accessibility measure.
Walter George Hansen - Transportation Consultant
"Walter George Hansen, 76, founder of a transportation planning consulting company, died of cancer May 15 at Inova Fairfax Hospital. He lived in Annandale.
"Mr. Hansen co-founded Alan M. Voorhees and Associates in 1961. By the mid-1960s, it was operating worldwide. It was acquired by Planning Research, later Ashland Technology and finally AECOM Technology. He worked for each of those firms in executive positions, including chief operating officer. He retired in 1996."
Photo here
Bob Metcalfe, Inventor of the Ethernet, famously proposed that the value of a communications network is given by n^2, where is n is the number of members on the network. This has been dubbed Metcalfe's Law.
In an article published in IEEE Spectrum titled Metcalfe's Law is Wrong, my colleague Andrew Odlyzko with Bob Briscoe and Benjamin Tilly reason from Zipf's Law (using Zipf's Law applied to word frequency, but as transportationists, we could just as easily use Zipf's Law as applied to city size distribution) why this is not the case, and that n log(n) is a better estimate. In short, not every connection is equally valuable. This is something well understood in transportation, where accessibility measures discount connections by a function of their travel impedance. However this article suggests there is something else going on, that there are, in a sense, diminishing returns to connections. The first connection is more valuable than the second.
One could organize this over time instead of just network size, and suggest that network value grows at a decreasing rate as all the best connections are made first, then the next best connections, and so on.
If this is the case, this generates the hypothesis (which I have not yet tested) that in a hedonic model of price (value) of real estate, accessibility measured as a product of the log of activities will give a better fit than one which just uses activities straight. (Results of hedonic models suggest accessibility is a significant factor in explaining house price, see Access to Destinations: Development of Accessibility Measures (esp. Chapter 5) for an example ).
Traditionally we represent Accessibility (Hansen's Accessibility Measure) at point i (Ai) as proportional to Destinations at j (say employment Ej) multiplied by f(Cij) where Cij is a travel cost, and f(Cij) is a travel impedance function (e.g. I/Cij^2) in the classic gravity model or e^(B*Cij) using a negative exponential form B<0).
Ai = ∑ Ej * f(Cij)
but the n log(n) argument suggests
Ai = log(∑Ej * f(Cij) )
might give a better fit in a behavioral or hedonic model dependent on accessibility.
(in short we discount the job for its difficulty to reach before we discount it because of diminishing returns. )
A new paper with Ahmed El-Geneidy publishes some of our results from the Access to Destinations Project ... Mapping Accessibility Over Time is now available for download from the Journal of Maps
The Journal of Maps is freely available, though registration is required (user name = email ... this is not immediately obvious).
The abstract notes:
"This study compares the changes in levels of accessibility over time in the Minneapolis - St. Paul region using two different modes (car and public transport). The importance of accessibility as a measure of land use and transportation planning performance in the region is revealed by comparing it over time. The longitudinal analysis being conducted shows increases in accessibility by car in most areas in the studied region, and a drop in accessibility by public transport over the period 1990 to 2000. The findings are compared to the levels of congestion in the region between the same time periods. This comparison shows the difference between the two measures and strengthens the importance of accessibility measures as a tool for monitoring and evaluating regional land use and transportation planning performance."
updated August 25, 2009:
For those of you who doubt I am doing work over in London, I have completed two other papers (in addition to "Too Expensive to Meter" based on my research over here):
Network growth is a complex phenomenon. Some have suggested that it occurs in an orderly or rational way, based on the size of the places that are connected. David Levinson examines the order in which stations were added to the London surface rail and Underground rail networks in the nineteenth and twentieth centuries, testing the extent to which order correlates with population density. While population density is an important factor in explaining order, he shows that other factors were at work. The network itself helps to reshape land uses, and a network that may have been well ordered at one time may drift away from order as activities relocate.
This article examines the changes that occurred in the rail network and density of population in London during the 19th and 20th centuries. It aims to disentangle the 'chicken and egg' problem of which came first, network or land development, through a set of statistical analyses clearly distinguishing events by order. Using panel data representing the 33 boroughs of London over each decade from 1871 to 2001, the research finds that there is a positive feedback effect between population density and network density. Additional rail stations (either Underground or surface) are positive factors leading to subsequent increases in population in the suburbs of London, while additional population density is a factor in subsequently deploying more rail. These effects differ in central London, where the additional accessibility produced by rail led to commercial development and concomitant depopulation. There are also differences in the effects associated with surface rail stations and Underground stations, as the Underground was able to get into central London in a way that surface rail could not. However, the two networks were weak (and statistically insignificant) substitutes for each other in the suburbs, while the density of surface rail stations was a complement to the Underground in the center, though not vice versa.
Perhaps more interesting for the non-academic, we (Ahmed El-Geneidy, Feng Xie, and myself of the Nexus group) have put together three quicktime movies
Our first report in the Access to Destinations Series: Development of Accessibility Measures
has finally been released.
The most interesting finding (which still awaits corroboration) is that despite the rising congestion of the past decade, accessibility in the Twin Cities region (measured as the number of things (jobs, workers, etc.) that you can get to in a fixed period of time) has been improving. Clearly this would be because there are more things per unit time, not because you can cover more distance per unit time. Increasing density increases accessibility, this is why cities form in the first place, it is nice to see it in the data. More in the final report. Thanks to my colleague Ahmed El-Geneidy who did most of the number crunching.
Randall Crane's Urban Planning Research discusses accessibility vs. mobility. He seems to be searching for a definition.
Accessibility is nicely defined as the ease of reaching particular destinations. That can be operationalized (and easily communicated) as how much stuff you can get to in a particular amount of time (e.g. number of jobs within 20 minutes). Our book from the conference is now out. The first of many reports on methods for measuring accessibility will be out soon.
Randall rightly notes that the importance of different things varies for different people. Accessibility measured as above is clearly a supply (or opportunities) measure, and makes no account of demand. No one measure encapsulates the entire economy.
Choices have costs. Increasing acccessibility is not free. Enabling someone to access 101 grocery stores in 30 minutes travel by auto instead of 100 will likely not be noticed unless that grocery is somehow distinct, and valuable, to an individual consumer.
-- dml
In the most recent episode of The West Wing(the penultimate episode "Institutional Memory"), White House Chief of Staff C.J. Cregg is being recruited to help run a foundation loosely based on the Bill and Melinda Gates Foundation, and her idea of what the best use of $10 billion would be to criss-cross Africa with highways, which would enable the delivery of medicine, expand trade, and do all sorts of good things.
The website Visualcomplexity.com has some really nice transportation graphics, which I became aware of after seeing Manuel Lima present at Meshforum. In particular, the travel time remapping of the London Underground is quite slick.
--dml
Manitoba or Manhattan
The Access to Destinations Conference, which I helped organize was just completed. It brought together 30 researchers from 5 continents to discuss the theory and practice of questions related to Accessibility. Accessibility is a measure of the ease of reaching destinations, and is contrasted with mobility, which simply measures the ease of use of the network. Accessibility and congestion and related phenomena, but not identical. The ability to move faster on the network generally improves both accessibility and congestion. However, accessibility accounts for land use, while mobility measures don't. Manitoba is an example of a place with no congestion, and very low accessibility. Manhattan, on the other hand, has a great deal of congestion and a slow network, but also a great deal of accessibilty, many places can be reached in a very short time.
A book with the proceedings should be out in 2005.
Article in Minnesota Daily - November 9, 2004