[sustran] Wendell Cox's Comments

[Todd Litman]

I appreciate Wendell Cox taking the trouble to reply to my comments. Below
are some responses.


At 01:55 PM 02/05/2000 -0600, Wendell Cox wrote:
>A few points in relation to Todd Littman's comments.....

> CONCERNING IDEOLOGY
> > I don't think that most critics of automobile-oriented transportation
> > planning consider cars and roads equivalent of the "Great Satan." Many of
> > us believe that automobiles have a role in a balanced transportation
> > system, but that current transport policies result in excessive driving,
> > and that given more optimal transportation and land use markets that
> > provide viable choices, fair and efficient pricing, and rational
> > investments consumers would choose to drive less and be better off as a
> > result.
>
>"Balanced transportation???" I've yet to see them support a new freeway.

Support for new freeways is not an appropriate measure of support for
balanced transportation if the current transportation system is unbalanced.
The need for new highway capacity could be avoided by reforming current
market distortions that encourage excessive automobile travel. For
disucssion of these strategies see our reports "Win-Win Transportation
Solutions" and "Road Relief", posted at our website: http://www.vtpi.org.


>Parking provided by stores or by people for themselves are not a subsidy.
>I, and most, define subsidies as monies paid by a person or business for
someone
>else, not for their own business or personal use. Transit is run, mainly on
>subsidies (in the Portland metro area, only 19 % comes from the users). A
store that
>provides parking, doesn't have to. It does so because the convenience
>allows them to make more profit.

Cox's defintion of subsidy reflects sector level analysis, which means that
costs are internal to a group. As discussed in our report "Socially Optimal
Transport Prices and Markets" this does not address efficiency issues,
since consumers do not have the opportunity to save parking costs by
reducing their vehicle trips, and it ignores the inequities that result
from cross subsidies within the group. 

Most businesses are required to provide a minumum amount of parking by
zoning laws. Partly as a result, free parking has become widespread and
expected by consumers, so individual businesses would face a competitive
disadvantage if they don't offer it. However, I think that it is wrong to
assume that this situation is necessarily optimal for consumers or
businesses, since it reflects economically inefficient market practices
(see our report "Transportation Market Distortions - A Survey" for
disucssion of the economic impacts of such distortions).


>1. FHWA study in question combines highway and transit expenditures. My
>comments relate to highways and roadways alone. For data from 1993, which
>demonstrates my points, see The Highway and Motorway Fact Book at The Public
>Purpose website:
>
>http://www.publicpurpose.com/hwy-us$93&c.htm
>
>Highway user fees are generally equal to or greater than the expenditure on
>roadways in the United States, at all levels of government. Much of these
>fees is used for other purposes, which creates a deficit that is covered by
>general taxation.

As I pointed out, this depends on perspective and assumptions. For example,
Cox's analysis assumes that all fuel taxes should be considered "user fees"
and dedicated to highway, which makes them functionally exempt from general
taxes. If you believe that fuel should be charged general taxes at the same
rate as other consumer goods then spending a portion of fuel taxes for
non-highway purposes is no more a "diversion" than to suggest that all
sales taxes on hats should be dedicated to public hatracks, and that
putting such taxes into general budget represents a "diversion".

Similary, Cox's analysis suggests that motorists user fees can legitimately
be used to fund state highway law enforcement and safety programs, but not
the same services provided by local agencies. 

Analyses that use more comprehensive analysis of user payments and public
costs consistently show that motorists pay less in user fees than what is
spent to serve them. See the 1997 Federal Highway Cost Allocation Study or
our report "Transportation Cost Analysis".


>2. As regards external costs, there are two sides to this discussion. For
>the other side, see The Urban Transport Fact Book at The Public Purpose
>Website:
>
>http://www.publicpurpose.com/ut-littman.htm
>
>Also recommended is... "The Myth of Social Cost," By Steven N. S. Cheung
>(Institute of Economic Affairs, 1978).

I welcome readers to review Cox's comments and my report "Evaluating
Criticsm of Transportation Costing". I don't believe that Cox effectively
addresses my comments about his article, nor do I belive that Cheung's
analysis "proves" that social costs and externalities are inappropriate for
consideration in public policy. Just as a consumer wants comprehensive
information on the full costs they would incur before purchasing a vehicle,
communities need comprehensive information on the full costs incurred by
public policy decisions.


>3. Would be interested in the transit investments that are less expensive
>than urban roadways.

Critics of public transit expenditures often point out that public transit
service frequently costs several dollars per addional trip. However, when
car trips are properly accounted, they often incur even greater external
costs. Let me give an example from our report "Evaluating Public Transit
Benefits and Costs".

A few years ago our provincial highway agency considered two options for
improving mobility between the city of Victoria and the nearby Western
Communities. Adding two lanes of highway capacity was estimated to cost
about $250 million, while a light-rail transit system would have cost an
estimated $300 million. Thus, they concluded that the highway was the
cheaper way to accommdodate 2,000 to 4,000 additional peak period trips.

However, this overlooks many additional transportation costs. A car trip
requires more than just a highway. It also requires a vehicle, surface
streets, and parking spaces. If the highway is widened to accommodate
2,000-4,000 additional peak period trips, the additional downstream traffic
will increase congestion on Victoria's streets, imposing costs on other
motorists and requiring additional roadway improvements. Parking problems
will also increase, requiring somebody (businesses or governments) to build
more parking capacity at $5,000-30,000 per space. These additional costs
alone would have shifted the analysis from favoring the highway to favoring
the rail option.

Conventional highway project investment analysis usually only considers
users' out of pocket expenses, assuming that everybody has an automobile
available that would otherwise sit unused (i.e., there is no opportunity
cost to vehicle use). High quality transit provides additional consumer
benefits: some households can reduce their vehicle ownership or defer
vehicle replacement, or a vehicle is available to another household member. 

Although the costs of providing public transit service in congested urban
areas is significant, so is the cost of providing highways in such
circumstances. One study by FHWA planners estimates that accommdodating an
additional peak-period car trips costs about 62 cents per vehicle mile
(Patrick Decorla-Souza and Ronald Jensen-Fisher, "Comparing Multimodal
Alternatives in Major Travel Corridors," Transportation Research Record
1429, 1997, pp. 15-23). This high marginal cost is often ignored by using
average values, which represent a cross subsidy from motorists who don't
drive on congested highways to those that do.

Public transit costs are also distorted by other factors, including low
demand due to under pricing (lack of congestion pricing, free parking,
fixed insurance and registration fees, external accident costs and
environmental impacts) and zoning and development practices that mandate
automobile dependent land uses. If these market distortions were corrected
demand for public transit should increase significantly. Since transit
service enjoys economies of scale this would substantially reduce per-trip
costs and subsidies. This conclusion is supported by international research
indicating that less automobile dependent regions have much better transit
service cost recovery than those that are more automobile dependent
(summary data available in the report "Automobile Dependency and Economic
Development" posted at our website).

Cox's website includes the factoid that it would be cheaper to give each
user of Seattle's proposed rail system a new Jaguar car. This is a false
comparison. Rail includes right-of-way and avoids parking, so a more
appropriate comparision would be the cost of the car, the road and a
parking space, which would tend to be quite high in the high density urban
corridor that the rail line will serve.


>4. Public service costs tend to be higher, not lower, in higher density
>inner cities --- this is the cost of providing the services and
>infrastructure within the boundaries of the central cities. This has to do
>with an inherently more costly cost structure, greater special interest
>control, a comparative lack of innovation and other factors. See the
>following paper at "Demogaphic Briefs and Urban Policy, at The Public
>Purpose website:
>
>http://www.publicpurpose.com/tor-demo.htm
>
>Also see Helen F. Ladd, "Population Growth, Density and the Costs of
>Providing Public" Services, Urban Studies, Vol 2, 1992, pp 273-295

Cox confuses city size with density. Higher costs associated with large
cities that Cox cites, if they exist, would not be affected by changes in
development density or other Smart Growth objectives. A number of
independent studies show that per capita public service costs increase with
sprawl-type development patterns and decline with moderate densities and
urban infill (see citations in our report "Land Use Impact Costs of
Transportation").


>Further, public service costs are typically fully paid by people who live in
>suburban municipalities. Sometimes they get some of their services from the
>central city, but pay the central city. The "failure to charge residents for
>the higher public service costs associated with lower-density locations" is
>thus puzzling --- lower densities do not mean the costs are higher, and
>whether or not they are higher they are being paid by the users who live in
>the lower density communities.

Because lower-density, urban periphery development imposes higher public
service costs there are often cross-subsidies from higher- to lower-density
neighborhoods within political jurisdications. One study (Subhrait
Guhathakurta, "Who Pays for Growth in the City of Phoenix? An Equity-Based
Perspective on Suburbanization," Urban Affairs Review, Vol. 33, No. 5
(www.urbanfutures.org/j102898.html), July 1998, pp. 813-838) found that
suburban residents in Phoenix receive 77% more government infrastructure
investment per household than residents of older urban neighborhood, but
pay less in taxes.


>5. Disagree with the view that suburbanization is not generally a market
>based response. Suburbanization occurs largely because of affluence and the
>availability of more modern forms of urban transport. It began long before
>the auto. It is absolutely overwhelming, and occurs at different rates
>probably based upon differing market forces variations in the extent of
>government intervention in the market.
>
>In the United States and Canada, only three central cities that have not
>annexed since 1950/1 have gained population --- Vancouver, which is up 169k,
>compared to an increase of 1.1 m in the suburbs, Miami, up 115k, compared to
>nearly 3m in the suburbs, and Los Angeles, which was more than 1/3 rural
>land in 1950. Any review of inner city census data (core areas) generally
>shows population losses of 20-50 percent, even in places like Portland and
>Indianapolis, where annexation could mislead one to the view that inner city
>growth has occured.
>
>It is no less stark in Europe, where cities that have not annexed territory
>have lost population, with all population growth in the suburbs --- examples
>are virtually all of the UK cities, Paris (down as great a percentage as
>Chicago), Copenhagen, Stockholm, Amsterdam, Rotterdam, etc.
>
>In Japan the losses started later, largely because it took Japan much longer
>to recover economically from WW2 --- but Tokyo is down more than 1m from
>1965 and Osaka is below its 1940 population. All growth in metro Tokyo and
>Osaka has been suburban since that time. Generally, in other Japanese urban
>areas, the bulk of growth has been in the suburbs.
>
>With respect to Europe and Japan, one might expect government intervention
>in the market with respect to excessively high taxation of motor fuel and
>auto taxes, combined with the high land costs, and the lower income
>(purchasing parity) to militate against suburbanization. But these
>government interventions and market forces have only slightly restrained the
>trend.
>
>Historical US city population data can be found at Demographic Briefs on the
>Demographia website...
>
>http://www.demographia.com/db-uscity1850.htm
>
>Historical US metropolitan area populations (1998 area definitions) can be
>found at Demographic Briefs on the Demographia website at:
>
>http://www.demographia.com/db-usmetfr1900.pdf
>
>Historical city and metropolitan area population information for the UK and
>Europe can be found at Demographic Briefs on the Demographia website...
>
>http://www.demographia.com/dbx-europe.htm
>
>Additional European data will be found at Demographic Briefs and Urban
>Policy on The Public Purpose website...
>
>http://www.publicpurpose.com/dmx-eur.htm
>
>Historical city and metropolitan area population information for Japan can
>be found at Demographic Briefs on the Demographia website...
>
>http://www.demographia.com/dbx-japan.htm
>
>Historical city and metropolitan area population information for Canada can
>be found at Demographic Briefs and Urban Policy on The Public Purpose
>website...
>
>http://www.publicpurpose.com/dmx-canada.htm\

My point is not that suburbanization is "bad" or is the factor of
consideration. My concern is with the manner of development. I believe that
current development patterns result in part from market distortions that
encourage many households to choose lower-density, more
automobile-dependent locations than what they actually prefer. 


>6.. Finally, the "growing popularity of loft apartments, neotraditional
>development and gentrified urban neighborhoods" justifies a few comments....
>
>a. Loft apartments and the movement back to the city is an important and
>encouraging development. It is largely people with no children... because
>the schools are lousy ... and often such developments have extra security,
>if not special additional patrol efforts by local police.  Even so, the
>numbers are miniscule compared to the population growth in the suburbs of
>the same cities.
>
>b. Neotraditional developments are an interesting case. They are, as often
>as not, agents of suburbanization themselves.
>
>For example, Celebration, in Florida, is being built on a greenfield
>suburban site (as was the neotraditional icon, Seaside) ... it is
>contributing to sprawl. There is no public transit. Much of the community,
>when built out, will be more than walking distance from the commercial core.
>And, the "mix of housing" ranges from well above average price to way
>above --- hardly the type of development that one can imagine replacing the
>1950s to 1980s middle income housing stock occupied by most Americans. But
>most amazingly of all, it is being built at a density less than half that of
>the average suburban density built since 1950. At this rate, an additional
>100+ acres would need to have been developed in US urbanized areas of more
>than one million from 1950 to 1990 (nearly 45 hectares).
>
>McKenzie Towne in Calgary is a bit more successful, with higher densities
>and paltry (rather than no) transit service. But it too is on a greenfield
>site and contributes to suburbanization. Transit work trip market share is
>less than the city of Calgary in general.
>
>For a review of McKenzie Towne, see Demographic Briefs on the Demographia
>website...
>
>http://www.demographia.com/db-mckenzietowne.htm
>
>Reviews of three other developments... Celebration, Habor Town (Memphis) and
>Har-Ber Meadows (Springdale, AR) will soon be posted, with references at the
>Demographic Briefs Urban Development Index on the Demographia website...
>
>http://www.demographia.com/dbx-nu.htm
>
>Other reviews will be posted in the future. Neotraditional development needs
>to be considered for what it is --- an issue of architectural taste, not a
>fundamental shift in urban development. Before long you will see early 1900s
>treatments applied to traditional suburban developments on 1/2 and one acre
>lots, and perhaps even on mobile homes. The style is attractive, and can be
>placed on any size lot... even the large lots that Americans prefer.
>
>c. Gentrified urban neighborhoods are already beginning to price the poor
>out of their homes in places like Austin and Portland, as it did before in
>its previous life in other cities in the 1970s and 1980s. I don't believe
>that this is a positive development, and neither do the people to whom it is
>happening.
>
>Finally, I for one, applaud the limited success being achieved in central
>city revitalization, and hope that it continues and expands. It would be
>nice to have the cities compete again for middle income people, but they
>will first have to get their houses in order, which they are not (poor
>schools, high crime rates, high taxes, poor public services, special
>interest control, etc). There is no doubt that if the government failures of
>the past 50 years had not interfered in the market, US central cities would
>be stronger today.

This is exactly my point. It is both futile and foolish to invest in more
highway and parking capacity to "solve" transportation problems is the root
of thsee problems is poor services and environmental conditions in existing
urban areas. It is far better in terms of economic productivity, consumer
benefits, and the environment if resources that are currently dedicated to
highway capacity expansion were instead used to improve public service and
provide amenities that make higher density urban areas more attractive to
consumers. Much of this I believe can be achieved by removing market
distoritions that currently favor automobile travel and automobile-oriented
land use. This is the idea behind "Smart Growth", which Cox rails against
while demanding that its objectives (better urban environments,
middle-class returns to the cities, more neutral transportation and land
use markets) be achieved.

In terms of housing affordability, many of the forces of automobile
dependency increase housing costs, including the use of residential
property taxes to fund roadways and traffic services, minimum residential
parking requirements, lower-density mandates, and the higher transportation
costs associated with suburban living that tends to be ignored in mortgage
assessments (thus further skewing household location decisions toward
automobile dependent homes). Cox suggests that affordable housing should be
provided by keeping some neighborhoods so unpleasant that most households
would not consider living there despite depressed real estate prices. Some
of us think that there are better ways to produce more affordable housing
through good design and management practices. For discussion see Wenya Jia
and Martin Wachs, "Parking and Affordable Housing," Access, Vol. 13, Fall
1998 (http://socrates.berkeley.edu/~uctc/access.html), pp. 22-25; "Parking
Requirement Imapcts on Housing Affordability," VTPI (http://www.vtpi.org),
1999; Kim Hoeveler, "Accessibility vs. Mobility: The Location Efficient
Mortgage," Public Investment, American Planning Asso. (Chicago;
www.planning.org) and Center for Neighborhood Technology (www.cnt.org/lem),
1997; David B. Goldstein, Making Housing More Affordable Correcting
Misplaced Incentives in the Lending System, NRDC (www.nrdc.org), 1996.


=====================


Let me make a few points in response to Mr. Coxes comments:

CONCERNING IDEOLOGY
I don't think that most critics of automobile-oriented transportation
planning consider cars and roads equivalent of the "Great Satan." Many of
us believe that automobiles have a role in a balanced transportation
system, but that current transport policies result in excessive driving,
and that given more optimal transportation and land use markets that
provide viable choices, fair and efficient pricing, and rational
investments consumers would choose to drive less and be better off as a
result. The point is not that cars are "bad", but rather that they impose
significant costs on society.


CONCERNING ROADWAY COST RECOVERY
According to the "1997 Federal Highway Cost Allocation Study," USDOT
(www.fhwa.dot.gov) current vehicle user charges (fuel taxes and
registration fees) only cover about 70% of total roadway costs, indicating
that these charges would need to increase by about 43% to achieve full cost
recovery. That is, the price of fuel would need to increase to about
60-cents per gallon just to cover road expenditures.

It is true that the highway cost study did include some transit
expenditures as part of road system costs (which can be justified on
least-cost principles, since transit investments are often cheaper than
building new urban highways, and because automobile dependency reduces the
efficiency of public transit). On the other hand, it omits a number of
additional facility costs and tax exemptions currently provided to
motorists. For example, the costs of local traffic services and municipal
parking subsidies are not included, and many states that apply general
sales taxes to most consumer goods exempt fuel. 

There are a number of other external costs imposed by driving, including
parking subsidies, uncompensated accident costs, the opportunity cost of
roadway land, traffic congestion (traffic congestion is usually defined
only in terms of delay to other motor vehicles, but pedestrians and
cyclists are also delayed by traffic), and a number of negative
environmental impacts. These are discussed in our paper "Transportation
Cost Analysis" a summary of which is posted at our website,
http://www.vtpi.org. It summarizes the results of a number of studies on
the full costs of various modes of transportation.

The 1982 Federal Highway Cost Allocation Study examined the issue of
whether local roads should be funded by users or through general taxes and
concluded that once basic access is provided to an area there is no good
reason to continue general tax funding. This implies that there is
virtually no justification for general tax funding of current roadway costs.

A few automobile advocates have criticized these studies, arguing that
motorists pay their fair share of costs, but virtually none of the critics'
reports are published by peer-reviewed journals, while many of the
comprehensive cost studies are. For a discussion see our paper "Evaluating
Criticism of Transportation Costing".

INEVIDABILITY OF SUBURBANIZATION
Although most developed cities have experienced suburbanization, there is
some question as to how inevidable this is and how much it reflects free
market choice. Just as automobile use is underpriced, resulting in
excessive driving, lower-density development is underpriced, resulting in
excessive suburbanization. Examples of land use market distortions that
favor lower-density, urban fringe development include dedicated highway
funding that encourages automobile oriented solutions to transportation
problems, zoning laws that require generous parking and lower density
development, mortgage interest payment income tax deductions, and failure
to charge residents for the higher public service costs associated with
lower-density locations. 

As a result, residents of higher-density older neighborhoods tend to
overpay their public service costs, while residents of lower-density, newer
residences tend to underpay (see Subhrait Guhathakurta, "Who Pays for
Growth in the City of Phoenix? An Equity-Based Perspective on
Suburbanization," Urban Affairs Review, Vol. 33, No. 5,
www.urbanfutures.org/j102898.html, July 1998, pp. 813-838; Robert Burchell,
et al., The Costs of Sprawl - Revisited, TCRP Report 39, Transportation
Research Board, www.nas.edu/trb, 1998). To put this another way, the
current land market fails to provide with effective price incentives,
because it does not return to individual consumers the savings that result
when they choose lower-cost locations. This encourages suburban development
over urban infill.

There are also some self-fulfilling prophesies at work. As more
middle-class households move to suburbs, urban neighborhoods become less
desirable places to live. But note that this has little to do with physical
conditions, rather it represents economic and social trends that favor
suburbs over cities.

This is not to say that in a less distorted land market everybody would be
living in highrise apartments, but it does suggest that some, perhaps much,
of suburbanization reflects a LACK of consumer choice and FAILURE of the
market, not true consumer preferences as Cox claims. The growing popularity
of loft apartments, neotraditional development and gentrified urban
neighborhoods indicates that given even moderatly attractive conditions,
some households would prefer moderate- to high-density residences.

ECONOMIC IMPACTS OF AUTOMOBILE DEPENDENCY
There is both theoretical and emperical evidence that excessive automobile
dependency has negative economic development impacts, since it imposes many
costs on governments, consumers and society overall, and it tends to
require a large amount of imported goods. A more balanced transportation
system could provide more economic stability to developing countries such
as Hungary. For discussion see "Automobile Dependency and Economic
Development" posted at our website.


Sincerely,

Todd Litman, Director
Victoria Transport Policy Institute
"Efficiency - Equity - Clarity"
1250 Rudlin Street
Victoria, BC, V8V 3R7, Canada
Phone & Fax: 250-360-1560
E-mail:  litman at vtpi.org
Website: http://www.vtpi.org



>---------- Forwarded message ----------
>Date: Sun, 30 Jan 2000 20:11:48 -0600
>From: Wendell Cox <wcox at publicpurpose.com>
>Reply-To: sustran-discuss at jca.ax.apc.org
>To: sustran-discuss at jca.ax.apc.org
>Subject: [sustran] Fw: Sprawl in Hungary
>
>> Sorry for the discordant view... but...
>>
>> I understand that a lot of people think of autos and roads as being the
>> equivalent of the "Great Satan," but that does not change the facts.
>> Contrary to the claims in the "sprawl in Hungary" article...
>>
>> The highways in Calif and Texas, not to mention across the US were paid
>for
>> by users --- through taxes assessed on fuel alone for the purpose of
>> building the roads. These taxes are specific to fuel, and not charged on
>> other commodities. A small percentage of user fees is tolls. Data for the
>> past five years is available at....
>>
>> http://www.publicpurpose.com/hwy-us$93&c.htm
>>
>> There is some general taxation support of roads, but it is approximately
>> canceled by the diversion of user revenues to other sources, such as mass
>> transit. Moreover, virtually all general taxation support of roads is for
>> LOCAL roadways, not for the motorways that are the backbone of the
>national
>> system.
>>
>> It is an inventive argument to connect sprawl to the S&L crisis. While
>some
>> weak connection might be made, the fact is that suburban expansion was at
>> its weakest in the 1980s, and much of the development would have occured
>> without the S&L crisis. The problem was that the national insurance
>program
>> was poorly administered... it was one of our most unfortunate government
>> failures, and we have had a few. A couple of larger ones have been the
>> abysmal failure of central city education and explosion of central city
>> crime rates from 1960 to 1990, which in and of themselves were of
>sufficient
>> concern to drive millions of people out of the central cities into the
>> suburbs. One would hope that the same will not occur in Hungary, and that
>as
>> a result, the inevitable movement to suburbs that is attendant to
>increased
>> affluence will simply reflect preferences in the market, rather than the
>> "bleeding" that has resulted from government failure in US central cities
>> (FYI, the city of St. Louis will show a population of 325k in 2000, down
>> from 857k in 1950 --- virtually all US inner cities have declined in
>> population, though the trend has been masked by annexation in some).
>>
>> Best regards,
>> Wendell Cox


Sincerely,

Todd Litman, Director
Victoria Transport Policy Institute
"Efficiency - Equity - Clarity"
1250 Rudlin Street
Victoria, BC, V8V 3R7, Canada
Phone & Fax: 250-360-1560
E-mail:  litman at vtpi.org
Website: http://www.vtpi.org