Moore and Thorsnes present an interesting take on the transportation/land use
situation. Their analysis is quite solid and internally consistent, but it
does betray their background in civil engineering. As they explicitly note,
their interest lies solely in efficiency, and they make no effort
to analyse the fairness of their proposals. Their policy recommendations
are quite sound, but need to be taken with a grain of salt due to the limited
scope of their analysis.
Personally, I contrast this with Jane Jacobs' work , which I'm
reading simultaneously. She approaches the issue from a more sociologic
point of view, and reaches different conclusions. The interesting approach
will be a combination of the two ways of thinking: an efficient and fair
The history is also telling. The social and environmental movements were
much quicker to realise the damage of the automobile than the engineers were;
Jacobs' book is from the early 60s, while Moore and Thorsnes (and much of the
engineering community) only began tackling the issue in the mid-80s. While
they're late to the table, it's good to see a real consensus forming:
auto dependency is a serious problem to both our society and our economy.
It has to be dealt with.
I would add a little analysis to their comments. Much of the history they
describe is a shift from high property values, dense, centralised development
around a rail import/export node, to low property values, sprawling,
polycentric cities. They tend to describe this as a more efficient system,
eliminating manufacturers' requirement to compete for scarce central land
and allowing them to spread across the landscape to cheaper land. However,
from an environmental perspective, the cheapness of this land is illusory.
Land itself is underpriced, since the ecosystem services provided by that
land are not reflected in its price. Furthermore, access to undeveloped land
also has a value, as I've noticed since I moved to Vancouver and began
enjoying the nearby mountains and hiking; this value is also completely
ignored by Moore and Thorsnes' analysis.
Overall, their analysis of transportation is quite solid, but I feel they
could still learn more about land use, and about social issues.
The rest of this document is just notes, drawn chapter-by-chapter from the
text of their report. This entire document is really just my private notes
on the report, but I've put it up on the web for search engines to slurp up.
- Good references: [3,13,20]
- Good references: [2,16,7]
- The authors start by stating acceptance of the optimality of the market
economy. The final sentence of this paragraph reveals their bias: while they
see problems in the transportation system, they still come from the typical
Civil Engineering background, and believe that our current system is
already quite effective "despite rhetoric to the contrary"
Understanding tends to reside in the temperate zone rather than at the poles:
the market may reveal itself as a friend of public goals, not a foe. A
principal contribution of modern economics to social thought is the
demonstration that, under the right conditions, the decisions made by millions
of self-interested individuals in the market will result in an allocation of
resources that is optimal for society in the aggregate. Those conditions
include (1) the initial distribution of wealth and income is deemed fair or
acceptable by society, and (2) the price each individual pays for each good or
service equals the cost to society of producing it. We readily admit that
these conditions often are not met precisely in urban markets. In particular,
prices often do diverge from cost (transportation and land-use policy is
almost always limited to mitigating the misallocation of resources that results
when prices diverge from costs). But prices reflect costs sufficiently well to
make modern cities extremely productive places. Despite rhetoric to the
contrary, those who set transportation and land-use policy are in the
unenviable position of trying to improve an already effective system.
- They approach the problem starting from simple model and building
to the complex, using reductionism as the rationale. (This is one of Jane
Jacob's classic complaints, where models for towns are applied to cities; I
don't think she'd argue with the approach here, though.)
- "The economic advantage of clustering is one of the two primary forces
that drive urbanization." They also call these economies of
agglomeration or external economies of scale.
- "The economic advantage that comes from putting people together with
machines under one roof is the second of the two primary forces for
urbanization." They call these economies of scale in production.
- The industrial revolution gives example of conditions in the absence of
effective transportation system: massive overcrowding around factories, etc.
- Central Place Theory  focuses on differences in demand and
supply between types of retailers. Demand depends on
- efficient scale of business
- number of people within travelling distance of business
- amount of goods and services demanded by each person
- The industrial revolution and changes in transportation technology
changed firms' locational preferences
- In the Monocentric City Model (in [16,14] or any urban
Offices need face-to-face contact, and can use land more intensively (towers),
while manufacturers need more land, and cannot outbid offices. Retail is also
in the centre to maximise access to customers under radial transport system
(seems like an artefact of the strange transportation system to me...)
- idealised topography, with a single import/export node and a flat,
- all travel is radial about the node, and no congestion exists
- rings are formed, with offices at the centre, manufacturing
surrounding the offices, and residential in the outer ring.
- Monocentric model can be extended to have a "waterfront" import/export
area, in which cases offices still cluster but don't take the entire
waterfront, allowing manufacturers to occupy much of the water. Arterials can
also be modelled to show clustering of retail.
Accessibility plays the key role in determining land use in the
monocentric city. It is not unreasonable to claim that costly transportation
causes and shapes cities. Even in the smallest urban areas, workers cluster
near their work place to reduce their expenditure on travel; they trade-off
land for accessibility. In larger urban areas, firms compete for central
locations to reduce the cost of travel. Economies of agglomeration exist
because transportation is expensive. The price of accessibility for both
workers and firms is higher land prices. Higher land prices encourage firms
and households to economize on land, which increases the intensity of
development in central areas. Cities look the way they do because people
compete for access. [p. 15]
- isochron map: lines showing equal times instead of equal distances
The fundamental factors that influence the characteristics of an efficient
transportation system are (1) the value travelers place on the service the
roadway provides, (2) the value of the land which the roadway occupies, and
(3) the technologies that exist to transport people and freight. The price of
the land used for transportation is determined by the competing demand of
other valued uses—factories, office buildings, stores, and houses. In the
monocentric city, the price of land increases as one approaches the center of
the city. Because roads occupy land, the cost of building a road with given
characteristics increases as the price of land increases (just as the cost of
building a house with given characteristics increases as the price of land
increases). [p. 15]
- Cities tended to have a single employment centre from about 1850 to
1930. American city populations grew much faster than rural populations during
this period, as horse-drawn and electric streetcars on rails lowered the
price (in time and money) of commuting.
- Streetcars encouraged suburbanisation, and increased downtown congestion
by increasing the number of workers with access to the CBD.
- In the monocentric model, it becomes clear that distance is really a
proxy for travel time.
- The development of the truck significantly decreased the cost of
freight transportation to/from the import/export node. Highways made it
feasible for manufacturers to locate far from rail: same freight cost, but
easier access to labour.
- High-tech manufacturers later had high-value lightweight products,
and hence relatively low freight costs, lessening dependence on freight
- Polycentric cities are the modern incarnation of monocentric cities.
In suburbs, clusters form due to agglomeration effects, and arterials then
form between clusters.
- Giuliano and Small  classify subcentres in Los Angeles
region by type
- specialised manufacturing (clustered for face-to-face contact)
- mixed industrial (larger mixtures near a transport node)
- mixed service (downtowns of the older towns absorbed by Los Angeles)
Moreover, the body of empirical work supports our argument that major changes
in urban form have resulted primarily from changes in the characteristics
of transportation. The auto, truck, and highway have given both firms and
households greater locational flexibility. That flexibility has changed
dramatically the character of urban development. The change has not come
without cost: some parts of most urban areas have experienced increases in
traffic congestion, declining environmental quality, and lack of development
and redevelopment in older areas. Policy makers hope that changes in policy
can address these problems. To provide insight into the potential of policy
to evoke desirable change, we discuss in Chapter 3 the effects past policy
has had on the changing pattern of urban development. [p. 25]
- Carefully ignoring all equity and social issues, they argue:
We have no doubt, however, that urban economies benefit from decentralized
development. The truck and the highway open up large quantities of land for
manufacturing, which reduces the price manufacturers have to pay for land.
In a competitive market, those savings get passed on to consumers. Moreover,
with improvements in electronic communications and decreases in the cost of
moving people, providers of services can operate more efficiently by splitting
their operations between subcenters. Finally, subcenters give consumers
better access to retailers and providers of culture and recreation.
Our analysis in Chapter 2 leads us to an answer to a question now topical among
planners: what leads, land use or transportation? The typical answer is both;
they interact to determine development patterns simultaneously. But if one
were pressed to select one or the other as the prime mover, which would it
be? Our answer: transportation. The most significant changes in land-use
patterns over the past 150 years have resulted from major changes in the
characteristics of transportation. Land cannot develop without access. The
nature of that access determines the characteristics of development, given the
technology of producing goods and services. [p. 27]
- Public policy plays a much larger role in transportation than in land
- In this chapter, the authors discuss the three failures of market
economies: natural monopoly, public goods, and externalities. Government
policy in three periods has seen each failure in turn. Private railroads prior
to the interstate system were natural monopolies; interstate highways were
examples of public goods, and suffered from congestion; and externalities
still abound in the transportation system, such as air and noise pollution.
They explicitly avoid addressing fairness issues, focussing solely on
The hope is that changes in technology and demographics, combined with
increases in roadway capacity, will reduce problems significantly. This hope is
unlikely to be realized. Public policy has contributed greatly to current
problems and must change to solve them. [pp. 28-29]
This is the case in the rail industry, and when it was combined high start-up
costs and limited regulation it led to natural monopoly in the rail industry.
Natural monopoly leads to price diverging from costs, as the monopolist
raises prices above costs, and in the absence of competition this can
An unregulated market allows the formation of a natural monopoly when economies
of scale allow a single large firm to produce enough to supply the entire market
at lower cost than could two or more competing firms. [p. 29]
- The public response to natural monopoly is either regulation or public
provision of the service. Governments in the early 20th century chose the
latter for railroads, water and sewer systems.
- The Depression marked the start of big public works projects. These
projects, while initially aimed at reducing unemployment, also had real
public value, and were unlikely to have been provided by private enterprise
due to the difficulty of recouping costs (trails in national parks, rural
roadways, construction of dams). This class of goods are known as public
Roadways are a congestable (i.e. non-pure) public good, and are particularly
difficult because it is impossible to exclude people who refuse to pay
from them (unlike, for example, a rock concert).
A good is public if it can be used or consumed by more than one person
simultaneously. A pure public good is not affected by the number of
people simultaneously consuming it. ... Most public goods, however, at some
point suffer from congestion. ... Private goods (e.g., consumer items bought
in stores) are extremely congestable: once purchased, they are not
available to other customers. [p. 31]
Government intervention in the market solves the private supplier's problem of
collecting payment for use of the public good (the supplier gets paid by the
government), but it shifts to government the problem of determining how much
of the public good to provide. Private businesses decide how much to produce
based on the price of the product in the market. As price increases, so does
production. People reveal their willingness to pay for the product by
purchasing it at the price established in the market. Since people cannot be
excluded from using or consuming some public goods, the government has
difficulty using price as a signal to guide production decisions.
- Government has to develop policies for allocating (1) available funds to
roadway projects, and (2) space on roadways to roadway users. Space is
currently allocated on a first-come, first-served basis. Fuel taxes fund most
infrastructure improvements, and these taxes bear no relation to the price of
space on a roadway, which may vary by location (rural vs. downtown) or by
time (rush hour vs. late night). Funds are allocated more carefully, but
the collection process still provides little insight as to where funding
would be most valuable—the places where the most fuel is bought are not
always well-correlated with the places that most need transportation
investment. Planners use a prioritisation process to decide (Also, 44% of
funds still come from property tax and general revenues.
Ref: Federal Highway Administration, Highway Statistics, 1991, p. 42)
- Existing computer models of traffic provide good short run estimates
of traffic flows resulting from transportation changes. In the long run,
however, changes in the network will shift land uses. If traffic volumes are
light, people can relocate to areas that require longer trips and traffic
volumes are consequently affected. Existing models do not account for changes
in land use in any way.
Consumers benefit from [the modern variety of goods and services]. Though
work-related travel has increased, much of the increase in VMT [Vehicle Miles
Travelled] has resulted from increases in the amount of travel dedicated to
shopping and recreation. Many planners today see this travel as a sign of the
inefficiency of urban sprawl, but it is also, and more importantly, a measure
of how much mobility most people enjoy. (Much of the public debate about
congestion emphasizes the need to decrease auto trips in particular and, by
implication, all trips. The people making those points are sometimes making
the judgment that trips are bad. This position appears at odds with one that
some people take simultaneously: that mobility is good. Our normative position
derives from economic principles: if a person is willing to pay for a trip
because he believes its value to be greater than its cost, then that's a good
trip for him. The problem that we address later is that, if the price he pays
does not include all the costs imposed, it may be a bad trip for the rest of
us. Our main point: there is nothing inherently bad about making many trips.)
- Modern transportation issues are often framed in terms of externalities:
congestion on the roadways, auto traffic increasing the comfort, convenience
and amenity of other modes of transport (walking and bicycling), parking
space, urban noise and air pollution, and automobile accidents.
External costs or benefits exist when people impose costs on or provide
benefits to other people who are uncompensated for the costs imposed or
uncharged for the benefits received. Industrial pollution is the textbook
example of an external cost (negative externality). Externalities are also
called "spillovers," and spillovers increase the closer people and their
activities get to one another. [p. 37]
- The authors argue (citing ) that governments are to
blame for environmental problems, as economists were "ready and waiting" for
the environmental issue, and advocated that government should implement
policies that create conditions that would exist in an efficiently operating
market by pricing external environmental effects. Governments ignore this
Pricing is more efficient [than directly controlling the quantity of
emissions] because it automatically allocates the right to pollute to those
who benefit the most from polluting. [p. 39]
Policy makers have not acted as if congestion is fundamentally an externalities
problem. Their objective has been to eliminate congestion (an objective
certainly inefficient and probably impossible to achieve short of draconian
measures) using tools unsuited to the task.
To combat one externality (congestion), transportation agencies have
sacrificed progress against others. New capacity on existing routes increases
traffic volumes, which increases the level of noise and air pollutants in the
area they traverse. Bypasses, despite their logical objective of separating
through from local traffic, encourage businesses and households to move
from the established developments to locations near the bypass, often leaving
economic blight behind. Those changes in land use then contribute to
congestion on the bypass, leaving local governments to deal both with
congestion and with economically depressed areas. Finally, increasing road
capacity often increases the costs of using other modes of travel. Bigger,
busier roads mean longer, less enjoyable walks to the transit station,
and longer, more dangerous trips for bicyclists. Decreasing the cost of auto
trips by increasing roadway capacity leads to land uses and building designs
oriented to the auto, not to walking, transit or bicycle. [p. 40]
- Subsidizing transit is not the solution to congestion; the problem is
the mispricing of auto travel, and while further mispricing may slightly
improve the situation, it will not solve it.
- Even with emissions testing, few drivers pay anything approaching the
cost of the damage they do to the environment.
- The authors conclude that transportation policy has had profound effects
on urban land development.
- Zoning regulations have attempted to curb externalities, by separating
residential uses from noisy and polluting industrial uses; by forcing large
lot sizes for noise buffering; and by limiting auto traffic with cul-de-sac
and winding street design. The effect, however, has been to increase auto
dependency, causing a simultaneous increase in externalities.
- The authors conclude that the conventional land-use policies
have tended to reinforce the trends in land development that resulted from
changes in the characteristics of transportation.
Analysts in the 1950s failed, not surprisingly, to foresee all the ways in
which people would respond to lower-priced travel and rising incomes.
Households bought more cars, changed driving habits, had babies who would grow
to drive, put women into the workforce, split into two, moved from central
areas to suburbs, and moved from rural areas and small towns to large cities.
Businesses made analogous changes: they moved from small towns to large cities,
split their operations between central city and suburb, and moved factories to
the urban fringe. Land-use policy reinforced the pattern of more and longer
trips by segregating origins and destinations and limiting densities. All
these changes in location and travel behavior, largely a response to unpriced
access to the impressive system or urban roads and highways, cause the
problems policy makers face today. [p. 43]
In this chapter, the authors present an analysis showing the current
subsidies given to automobile and truck transportation. They follow with a
range of ideal policies for addressing efficiency issues in roadways and
land use. Many are not realistic, implementable policies.
- pricing roadways
- pricing parking
- pricing auto noise and emissions
- pricing industrial emissions
- pricing commercial externalities
- pricing neighbourhood externalities
- pricing extensions of urban services
- improving the efficiency of the property tax
- One of the most interesting aspects of this chapter is the following
Cost of Roadway Travel in 1989 (Table 4-1). Sources: [21,12,22]
|Component of Cost
||Cost per Mile (US cents)
||Total Cost ($US billions)
||% of Total Social Cost
||% Unrecovered from Drivers
|Direct Costs of Highway Capital and Operation
|Roadway construction and repair
|Related services (police, etc.)
|Parking (lots and structures)
|Costs that Accrue to Drivers
|Other Private and Social Costs
|Air and noise pollution
- The average cost of travel by auto is nearly 78 cents per mile, about
three times the rate of 25 to 30 cents used by most governments and companies
to calculate auto expenses
- The total cost on roadway travel was US$1.6 trillion in 1989, compared
to a GDP of $5 trillion.
- The roadway costs most people think of (roadway construction, repair and
maintenance) make up only 3% of the total here.
- While the total unrecovered cost (12.6%) seems low, it would require a
fuel tax hike of US$1.80 per gallon (taxes are currently about US$0.38 per
- HOWEVER, these numbers are very, very rough. Some important inaccuracies:
- The "related services" category is local government expenditures
on police, fire, courts, attorneys and other social services that result
from auto traffic. The figure itself comes from combining federal costs
(0.6 cents per mile, from ) with the average of two studies:
 reported 3.2 cents per mile in Pasadena, California; and
 reported 2.8 cents per mile in San Francisco. This gives
3.6 cents per mile, and is extrapolated to the entire United States.
All told, it's a very large figure, but my (totally unsubstantiated) gut
feeling is that it's probably reasonable. I'm not sure where the 88.9%
unrecovered figure comes from, but I don't have much argument with that;
user fees definitely don't cover many of these kinds of services, with
the exception of small claims court.
- The parking figure is even more tenuous. They start with a guesstimate
of US$1000 per parking spot ( gave this figure, and
the authors cite figures of $500-$2500 for downtown structured
parking, surface parking costing 40% of that, and suburban
surface parking costing 20-25% of that; this figure yields something in
the ballpark of $1000 per spot). From this, they use the fact that there
are 95 million commuters to reach a cost of $95 billion. They then
state that one-third of all household trips in 1990 were work-related,
and make a conservative estimate that non-work trips are slightly more than
half of all trips, raising the $95 billion estimate to a very
rough $200 billion, and state that they wouldn't be at all surprised
if they were off by a factor of two. Finally, the 25% unrecovered
figure comes from the fact that free parking places offered by employers
and retail outlets are untaxed. They assume that "free" parking is
actually paid by the drivers: when commuting, it is paid through lower
wages; when shopping, it is paid through higher retail prices. They
do not account for the fact that non-drivers subsidize drivers' costs.
All told, I suspect their $200 billion number is very conservative, even
though it seems shockingly high. My gut feeling is that their price per
parking place is solid, but they underestimate the number of shopping
trips (two-thirds of all trips, although the household/commuter
correlation confuses things), and they underestimate the percent
unrecovered by ignoring non-drivers' subsidy to drivers in free parking
- The vehicle operating and capital costs are uncontroversial, and
in line with typical business prices per mile. Capital costs are derived
by depreciating the costs of automobiles over the life of the car.
- The time costs seem suspicious, but are based on fairly reasonable
assumptions.  reviewed a range of studies to conclude that
drivers value their time at about half their gross wage. The average gross
wage in the U.S. was $9.60, which gives $4.80 per hour. With an
average speed of 40 mph, this gives 12 cents per mile. The hourly
price is probably a little optimistic, since many VMT are not commuting,
and probably valued at less than half. This is balanced out, however, by
the optimistic average speed, which is probably lower due to congestion.
- The "schedule delay" figure is due to people taking fewer trips
and changing the timing of the trips they do take; it comes directly from
. My gut feeling is that this is a little high, and may
be somewhat double-counted with travel time costs.
- The pollution costs are very conservative—the authors cite an
extreme estimate of $200 billion. I'm curious to know if anyone has
attempted to estimate the value of noise pollution by comparing arterial
(noisy) and off-arterial (quiet) property values. The healthcare costs
of air pollution (and noise pollution, for that matter) must be quite high
- The accident costs are based on figures from [12,21].
The bulk of this figure ($229 billion in , $186 billion
in ) is due to pain, suffering and loss of life. Lost wages
amount to $58 billion and damage to property about $38 billion. Although
they do not state as much, I assume that vehicle insurance is not included
in operating expenses above, and pays for most of these costs. The figure
of 15% unrecovered comes from . In a country like
Canada, where vehicle insurers do not pay healthcare costs, I imagine
that much more of this expense would be unrecovered and paid by society
as a whole.
- They have also ignored some costs: the costs to land developers for
meeting auto-friendly design standards, or the value of land devoted to
streets, which can be quite substantial in downtowns, especially when
street parking is included. They also ignore the risk of global warming,
military expenditures to secure an oil supply, structural damage from
- Although the previous table made substantial arguments of
subsidisation of drivers by non-drivers, there are also major arguments
to be made for subsidisation within the driving group. Their argument
hinges on one idea: that people value trips differently. If the price of
an automobile trip was higher, some drivers would choose not to make the
trip (by car), because they do not value the trip highly. Those drivers
might instead travel by foot, bicycle or transit; or they might simply not
make the trip at all. Other drivers value the trip highly, and are willing
to pay the higher price.
- The authors use microeconomics to argue for congestion pricing of
the roadway. Given a demand curve for a trip, and a "private trip cost"
curve (essentially a supply curve) indicating the price of a trip as
perceived by the driver, we can establish the equilibrium traffic volume.
However, the actual cost of trips is given by the higher
"social trip cost" curve, which includes (some) externalities, such as the
delay a given trip imposes on other users of the roadway. The optimal
traffic volume is given by the intersection of the STC curve and demand
curve, and reflects a lower volume. The authors argue that by imposing
a roadway toll, the demand curve can be inferred, and hence a toll can
(eventually) be selected that will raise the PTC curve until the
equilibrium volume is the same as the optimal volume. The graph shown here
only applies at a single time of day; demand curves would need to be
inferred at different times, and varying tolls could then be established.
This will amount to congestion tolling, with tolls typically only paid
during peak periods.
- The authors also make the valid point that the cost is being paid
already: at present it is paid by drivers, in terms of lost time due to
congestion. The difference with congestion tolling is that the cost is
collected as government revenue, instead of being wasted as lost time.
This is the true efficiency gain in congestion tolling.
- Tolling also gives a direct idea of the value of added capacity:
roadways should only be expanded if the toll from the new capacity will
pay for the cost of expansion. Roadways should be self-supporting, in
other words. The authors hence propose shifting taxation from fuel taxes
to roadway taxes.
No rational concert promoter would decide how big to build a stadium based
on the number of people who would come to see the Grateful Dead if the
tickets were free. But that is often how transportation planners decide highway
capacity: they estimate how many trips would be made on an unpriced facility
big enough to accommodate that number of trips. [p. 56]
- The authors note that implementation is a major barrier. Toll booth
systems have been tried in pilot projects in Hong Kong  using
a centralised toll both system; the Autostrada in Italy 
uses a more sophisticated electronic credit card system (similar to Highway
407 in Toronto, or the London congestion charging).
- The authors note that research in transportation demand management
(TDM) [11,23] has shown that drivers respond strongly to
increases in the price of parking. As explained earlier, drivers currently
fail to perceive the cost of parking, even though it is typically provided
privately. The main reason for this is the tax code: the total cost of
free parking is currently lower than the total cost of paid parking, due to
the tax incentive for free parking. By eliminating the federal tax
break, or counteracting it with a municipal or regional tax hike, the incentive
to provide free parking will cease to exist.
- However, collecting parking fees is expensive. Redesigning lots is
probably the only way to correct this.
- Furthermore, if one retailer switches to pay parking while a neighbour
has free parking, the first retailer will suffer. A regional system is
required to encourage all retailers to switch to pay parking.
Parking charges will have effects on development patterns in the long run
similar to the effects of increasing roadway capacity. [p. 60]
- The authors suggest a noise and emissions charge similar to the
congestion charge: rates would depend upon the location of driving, with
downtown high-intensity areas being more expensive that suburban or rural
areas. The rationale for this differential charging is based on population
density: driving downtown harms more people (and hence imposes more cost to
society) than driving in a suburban area. This proposal is a double-edged
sword: it could encourage low-density development, since auto emissions are
more acceptable in such an environment. On the other hand, it will discourage
driving downtown, and will make downtown environments more appealing due to
their reduced noise and improved air quality.
- The authors also propose adjusting noise and emissions charges based on
the characteristics of the vehicle, to provide an incentive for automobile
manufacturers to improve emissions and noise.
- As an alternative to zoning, the authors suggest pricing industrial
emissions. They go beyond standard emissions trading permits, and suggest
a spatially sensitive system. This system would be similar to the auto noise
and emissions policies described earlier, where the cost of an emissions permit
would vary with the location of the emitter. This gives firms a choice:
by locating near residential areas, they gain access to workers and businesses,
but pay more in emissions fees. By locating further away, they must pay more
to attract workers, but pay less in emissions fees.
- The authors argue that this system is more efficient than zoning,
requiring less classification of industries into rigid niches and allowing
greater flexibility in industrial location.
- Implementation of industrial emissions pricing is difficult. The
government must know the impact of a wide range of pollutants on society, and
must be able to measure firms' emissions. Furthermore, selling such a scheme
to the public would be politically difficult, unless the taxes collected
could mitigate the damage done to a neighbourhood by nearby industrial
- The authors also briefly suggest pricing of commercial externalities,
in a similar manner to industrial emissions. The externalities in this case
include generated automobile traffic, rowdy customers (e.g., bars), and noisy
truck deliveries. The main benefit of such a scheme would be to allow
non-intrusive commercial uses in zones that are currently purely residential.
Examples include convenience stores, small grocers, and coffee shops. The
authors also suggest that such a system might encourage better
externality-limiting building design.
- In a similar manner, residential externalities could be priced. The
authors suggest taxes on poorly maintained property combined with a subsidy
for maintenance costs. The main goal here is to deal with the problems of
mixed incomes in a neighbourhood without resorting to the typical
exclusionary zoning tactics found in many parts of the U.S. The authors
readily acknowledge the difficulties here.
- Echoing many others, the authors suggest pricing urban service
extensions. Until recently, most new developments were charged the same rate
for sewer, electricity, etc. as older developments. New regions cost more
to supply due to their distance from the centre, and due to typical
low density development patterns. Cities are already beginning to address this
- Property tax could be shifted from a tax on improvements to a tax on
land. In other words, rather than basing property tax on the value of
the building and land, base it purely on the value of the land. This would
encourage denser development, since no additional tax would be paid for a
skyscraper vs. a bungalow on a similar sized parcel.
In general, decreasing the tax rate on improvements increases the intensity of
development, which would decrease VMT. Decreasing the tax rate decreases the
price of improvements relative to other goods, including land. As the price
of one input decreases relative to another, builders substitute that input
for the other. If the price of building a square foot of apartment buildings
decreases, developers will build larger apartment buildings. Developers of
single-family residential land will economize on relatively high-priced land
by building bigger, better houses on smaller lots. Smaller lots (higher-density
development) would decrease travel distances and increase the demand for modes
of travel other than the auto. [p. 65]
This chapter describes policies that influence prices, but do not
necessarily bring about a system of efficient prices.
- New lanes for congested highways have been a common response to
congestion issues. In the short-run, this policy has problems: since
highways are not efficiently priced, capacity will be increased beyond
the optimal volume.
Though drivers may appear better off because of the added capacity (the
number of trips increases and the average price of all trips decreases),
the costs of the land, labor, and materials needed to build the capacity
exceeds the value of the reduced congestion. [p. 68]
In the long run, the problem is worse: it encourages changes in driver
habits, auto ownership, and development patterns in ways that increase the
volume of traffic.
- Bypass highways are sometimes built when new lanes cannot be added to
a congested highway. The idea is to separate through traffic on the highway
from local traffic by placing fewer exits on the bypass, but the effects are
often highly negative. The new bypass highway tends to have better access
characteristics than the old highway, and commerce moves from the old
highway to the offramps of the bypass. This leads to two simultaneous
difficulties: dealing with growth along the bypass, and decay along the
older highway. One solution is to build express lanes within the old
highway, which may seem more expensive, but may be cheaper in the long run.
Another solution is for governments to restrict access points to the bypass,
or control land use at access points; the best option there is for
government to buy the land around access points, to prevent landowners
from lobbying for changes to land use (which amounts to a windfall for
- Subsidies to transit are often proposed as a means of making transit
a viable alternative to the automobile. However, the level of subsidy
required to match inefficient automobile trip prices may be exorbitant.
The authors argue that the transit subsidy increases transit ridership and
accessibility, but probably does little to decrease VMT in the congested
areas. Finally, subsidies to transit exacerbate the problem by increasing
the total subsidisation of transport.
Eliminating fares, for example, would probably not suffice. The subsidy
would have to be large enough to provide frequent, comfortable service
between a wide variety of origins and destinations, most developed at
relatively low densities. The level of transit service would have to be
similar to that provided by school bus systems ... but with higher-quality
vehicles and serving destinations located throughout the metropolitan
area. [p. 71]
The authors also suggest that rail transportation cannot be cost-effective in
most metropolitan areas. They do not that Toronto made decisions in the
1950s to invest heavily in transit and rail, and with accompanying land use
decisions managed to make rail cost competitive with the auto.
[U]nderpricing transportation results in too much travel in the sense that
too many resources get taken from the production of other goods.
- Transportation system management (TSM) aims to increase the capacity
of a roadway without adding new lanes. Methods may include lanes for high
occupancy vehicles (HOV lanes), signaled onramps for freeways and
controlling signalling of arterials. They conclude that these methods are
good in the short run, but still lower the time cost of travel, encouraging
land development patterns that bring back congestion in the long run.
Furthermore, methods such as metering onramps may favour suburban
long-distance commuters (who enter an empty roadway) over urban commuters
(who enter at a more congested point).
- Fuel taxes are currently the primary means of funding road
infrastructure. While the tax is correlated with use of the facility (a
per-km charge), it does not take into account variations in cost associated
with location (e.g., downtown vs. rural) or time (e.g., peak vs. night).
Fuel taxes are consequently a better proxy for emissions taxes than for
congestion taxes. Finally, raising fuel taxes now may be useful for a later
substitution with congestion charging.
- Parking should be priced at cost, according to the authors.
While some have argued that parking surcharges (i.e. at peak times) could act
as a proxy for congestion pricing, the authors feel that this merely
further distorts a distorted pricing system.
- Transportation Demand Management (TDM) usually refers to a
specific package of policies, usually employed by firms or residential
developers. See [4,17,11,23]. A successful program must
(1) provide access to a good alternative to the auto, and (2) increase the
relative cost of an SOV trip through increased parking fees or cash payments to
users of HOVs. In general, unless TDM is implemented regionwide and
significantly increases the cost of SOV trips, its effectiveness will be
Demand management policies and their effect on the demand for auto
trips. Sources: [17,4,11,23].
||Determinant of Demand Affected
||Expected Strength of Effect on Auto Trips
|Promote HOV Use
|Decrease fares or HOV costs
|Improve access to transit
|Improve transit service
||Waiting and riding time
|Provide HOV parking
|Guarantee a ride home
|Increase parking fees
||Auto trip cost
- Cordon congestion pricing (they call it restricting access to
congested areas) can be useful in congested (downtown) areas. Such a scheme
was implemented in Singapore in 1975 , and reduced traffic in
the central area by 70% while increasing average travel speed by 20%.
Queuing and lotteries are alternatives to pricing access, but these are not
- Do Nothing as a policy ("planned congestion") is also an option.
The authors were
initially hostile to the idea, but came around in favour eventually.
The authors argue that planned congestion combined with shifting funds to
improvements in other modes would change travel patterns in the right
direction. They cite 's observations in Zurich where transit
priority policies were implemented, increasing transit trips by 33% over
While technically inferior to congestion pricing, this policy is probably
easier to implement and may help ease the transition from the current
policy environment to a more efficient system of prices in the future.
- Zoning regulations, according to the authors, are not the cause
of current development patterns; many developers exceed minimum lot sizes,
citing market forces. These market forces arise from underpriced
transportation: transportation leads land use, in the authors' opinion. As
an interesting aside, the authors note that funding via property tax often
encourages segregation by income, since low-value properties don't share
the urban service burden as evenly as others. Another aside mentions urban
platting, where low-density developments must include plans for transition
to higher densities.
- Urban growth boundaries have potentially strong effects on
land-use and travel patterns. They limit the spatial extent of urban
development; in Oregon, they surround every large city. (Vancouver has one: the
Agricultural Land Reserve.) A UGB limits land supply while demand rises. It
prevents sprawl at the fringe, and simultaneously promotes more intense
development in the centre. However, they have problems, typical to
command-and-control policies (as opposed to market policies): the political
pressure to relax the boundary will only grow with time.
- The jobs and housing balance could be directly regulated,
mandating mixed use development. The authors argue that this will not have
much impact; with ongoing cheap transportation, many people will still
choose to live far from their jobs.
- Design standards can have significant effects on pedestrian
travel in an urban environment, even at a relatively low cost. Large-scale
commitments to development oriented to pedestrians and transit are more
difficult to achieve in the presence of underpriced auto travel, however,
according to the authors. Their concern is mainly that transit-oriented
development won't sell; too much automotive convenience must be given up by
- Public purchase of land or development rights could alleviate
some of the problems of UGBs. The motivation is purely political: by buying
the urban boundary land, the municipality avoids any pressure from
landowners at the fringe. Naturally, it is much cheaper to simply regulate
- Transferable development rights extend this idea by allowing
development at the fringe, if and only if the developer pays for
compensation to landowners in other parts of the fringe. This helps to make
the purchase more affordable, by pushing the cost onto the benificiaries of
- Performance/flexible zoning could potentially reduce other
urban externalities: emissions, noise, etc. They are similar to pricing
strategies to land-use, but do not price at cost; instead, a city planner
negotiates with developers for amenities to compensate for emissions and
- Urban service boundaries limit the region where urban services
are provided, or charge more in lower-density, harder-to-service regions.
The problems are similar to those of urban growth boundaries: the location
of the boundary becomes a political issue.
- Development impact fees will have a small impact on land
development, but are still inferior to pricing-based solutions.
||Direct to commuters
||To all society
||Ease of administration
|Rapidly removing accidents
|Improving highway maintenance
|Building added HOV lanes
|Building new roads without HOV lanes
|Upgrading city streets
|Building new off-road transit systems, expanding existing ones
|Increasing public transit usage by improving service, amenities
|Coordinating signals, TV monitoring, ramp signals, electronic signs,
converting streets to one way
|Instituting peak-hour tolls on main roads
|Parking tax on peak-hour arrivals
|Eliminating income tax deductibility of providing free employee parking
|Providing income tax deductibility for commuting allowance for all workers
|Increasing gasoline taxes
|Keeping densities in new growth areas above minimal levels
|Encouraging formation of TMAs, promoting ride sharing
|Encouraging people to work at home
|Changing federal work laws that discourage working at home
|Staggering working hours
|Clustering high-density housing near transit station stops
|Concentrating jobs in big clusters in areas of new growth
|Increasing automobile license fees
|Improving the jobs-housing balance
|Adopting local growth limits
- Subregional planning in the absence of regional planning is not
likely to be successful; neither is regional planning in the absence of an
empowered regional government [...]
- Regional government may be a necessary condition for successful
regional planning, but it is not a sufficient one [...]
- The stated goals for regional growth do not vary across metropolitan
areas [...] We suggest four goal categories:
- Environment and Natural Resources (or, the Natural Environment)
- Development and Built Resources (or, the Built Environment)
- Employment and Cultural Resources (or Amenities) (or, the
- Government and Institutional Resources (or, the Institutional
- The possibilities for future urban form are few [...]
- Expansion. [...] "Let the current trends continue" [...]
- Concentration. Emphasize growth in the existing urbanized area
- Concentration of the expansion. This concept takes a little from
both of the previous ones. [...]
- Deflection. [...] This concept would emphasize growth outside the
region, either proactively or reactively.
- Public involvement is necessary, expensive, and typically
ineffective; regional planning efforts will be simultaneously criticized as
too little and too costly [...]
- Measurement in a multi-objective world is always faulty; regional
planning led by technicians for technicians will be interesting to
technicians only. [...]
- Focus on direction, not destination; the only way for most people to
evaluate a long-run vision is to focus on the short-run policies that are
the first steps towards it [...]
- Work with the market to change behavior: change prices. [...]
- Evaluations of regional policies focus on efficiency; interest groups
and the public care about equity—what will this mean for me?
- Bundle complementary policies [...]
- Preserve long-run opportunities [...]
- Integrated regional planning needs champions [...]
- If you really want to affect the long run, take a long-run attitude
toward change [...]
- Squeeze more out of existing highway capacity without squeezing
- Help transit
- Slowly increase emissions standards
- Continue to experiment with TDM
- Organize regional institutions for regional problems
- Incrementally reform land-use and design policies
- Use more pricing policies for infrastructure
- Some short-run improvements in travel times on some routes resulting
from increases in capacity
- Transit ridership will increase, but not enough to reduce congestion
or operating subsidies
- TDM, combined with congestion and HOV lanes, will provide a small
increase in the average number of riders per vehicle
- New highway capacity will continue to provide economic incentives for
low-density development in suburbs
- Increase the price of auto trips toward cost
- Parking should also be priced properly
- Control the environmental externalities of trip making with pricing
- Apply more rigorous benefit-cost criteria to highway and transit
- Organize regional institutions for regional problems
- Make a greater effort to control externalities through pricing and
- Increase the use of pricing policies for infrastructure
- Reduction of congestion on limited-access highways to efficient
- Large shift from the auto to other modes and higher occupancy of autos
- Reduction in the rate of growth in per-capita VMT
- More stability in CBDs and suburban subcentres
- Slower development at urban fringe
- Greater density and more mixed-use in residential areas
- Higher-intensity development, especially near transit stations
- Reduction in the rate of growth in noise, air emissions, and other
- Improvements in the character of public spaces
W.J. Baumol and W.E. Oates.
The Theory of Environmental Policy.
Cambridge University Press, Cambridge, UK, 2nd edition, 1988.
Central Places in Southern Germany.
Stuck in Traffic: Coping with Peak-Hour
Brookings Institution Press, Washington, D.C., USA, 1992.
Transportation demand management: Planning, development and
Journal of the American Planning Association, 56(4):442-456,
G.J. Fielding and D.B. Klein.
How to franchise highways.
Journal of Transport Economics and Policy, 27(2):113-130,
F. Fitzroy and I. Smith.
Priority over pricing: Lessons from Zurich on the redundacy of road
Journal of Transport Economics and Policy, 27(2):207-214,
Genevieve Giuliano and Kenneth A. Small.
Subcenters in the Los Angeles region.
Regional Science and Urban Economics, 21(2):163-182, 1991.
Electronic road pricing: Developments in Hong Kong.
Journal of Transport Economics and Policy, 24(2):203-214,
The Death and Life of Great American Cities.
Vintage, New York City, NY, USA, 1961.
N. Krumholtz and J. Forester.
Making Equity Planning Work: Leadership in the Public Sector.
Temple University Press, 1990.
Evaluation of Travel Demand Management Measures to Relieve
U.S. Federal Highway Administration, 1990.
J.J. MacKenzie, R.C. Dower, and D.T. Chen.
The Going Rate: What it Really Costs to Drive.
World Resources Institute, Washington, D.C., USA, 1992.
John R. Meyer and Jose A. Gomez-Ibañez.
Autos, Transit and Cities.
Harvard University Press, 1981.
Edwin S. Mills and Bruce W. Hamilton.
Harper Collins, New York City, NY, USA, 4th edition, 1989.
Terry Moore and Paul Thorsnes.
The transportation/land use connection.
Technical Report 448/449, American Planning Association, Chicago, IL,
USA, January 1994.
Cities and Housing: The Spatial Pattern of Urban Residential
University of Chicago Press, 1969.
Can management of transportation demand help solve our growing
traffic congestion and air pollution problems?
Transportation Quarterly, 44(4):483-498, 1990.
Irwin, 2nd edition, 1993.
C.V. Patton and D.S. Sawicki.
Basic Methods of Policy Analysis and Planning.
Prentice Hall, 2nd edition, 1993.
Boris S. Pushkarev and Jeffrey M. Zupan.
Public Transportation and Land Use Policy.
Indiana University Press, Bloomington, IN, USA, 1977.
Kenneth A. Small.
Urban Transportation Economics.
Hardwood Academic Publishers, 1992.
U.S. Federal Highway Administration.
Highway Statistics 1989.
U.S. Department of Transportation, 1989.
Policy implications of recent behavioral research in transportation
Journal of Planning Literature, 5(4):333-341, 1991.