This paper examines three related but distinct questions:
·
Is urban transit investment economically worthwhile? What kind of
transit investments yield the greatest economic pay-off?
·
Is transit subsidy economically justified? and
·
Is there a compelling case for expanding the federal financial role
in urban transit?
Question one represents an inquiry into the fundamental value proposition for transit; whether the economic benefits justify the economic costs. The payment of subsidy, question two, hinges on more than a sound value proposition. If transportation markets were reasonably well behaved, good investments would be self-sustaining. The economic appropriateness and degree of subsidy must thus be examined in relation to market failure, including factors such as monopoly, externalities (congestion and environment), economies of scale, imperfect information, public goods and equity.
Question three is part of a much wider discussion about how public policy functions should be allocated among the different levels government (“federalism”). While a federal role in transit finance would demand both a strong value proposition and a reasoned economic case for subsidy, more is at issue. Policy here must turn on, (i) the strategic significance of transit in the national context; (ii) the fiscal capacity of different levels of government; and (iii) the extent to which federal contributions would substitute for rather than add to provincial and local outlays.
Plan of the Report
The report is organized around the three main questions outlined above. Section 2 examines the investment worth of transit. The level of justifiable subsidy is investigated in Section 3. Section 4 addresses implications of the findings for the federal role in transit support.
Unlike highways, for which a reasonably rigorous micro-economic analysis framework has been in place for more than 30 years, the appraisal of transit investment has been given to more subjective evaluation schemes. Although highway decision makers are not habituated to economic counsel, highway investment options are often examined in terms of their economic benefits, economic costs, net present values and rates of return: something of their economic worth is known accordingly. In contrast, transit projects are traditionally characterized for decision makers in terms of “planning balance sheets,” “multi-criteria scorecards,” “cost-per-trip” indices and other schemes that reveal little about transit’s economic value or the benefits of transit relative to its costs.
Recent shifts in common practice have however endeavoured to align decision support work in the transit sphere with the principles of economic analysis. A reasoned cost-benefit analysis framework has emerged. Peer reviewed and applied in a number of field situations, the framework is beginning to reveal a coherent account of transit’s net economic value.[6]
Costs and Benefits of Transit
Costs
The economic costs of transit include capital expenditures on vehicles, facilities and equipment; outlays for maintenance and repairs; spending on wages, fuel and other operating costs; and the opportunity cost of capital employed. Some economists believe that distortions in transit labour markets create deadweight losses which must be treated as an additional cost. Probably more relevant in the United States than in Canada, the argument is that labour conditions “grand-fathered” when transit systems shifted from private to public ownership have driven transit wage rates above the value of the marginal product of transit workers. If so, wages would incorporate an element of rent, the value of which would properly be treated as a cost.
Although transit proponents often seek to include construction employment on the benefits side of the ledger, there is no evidence that either building transit facilities or operating them reduces structural unemployment or underemployment in urban labour markets. Because of the scale of some transit projects (subways in particular), any macroeconomic effects are more likely to be negative than positive. This is because large-scale construction projects fuel inflation (by bidding up wage rates) during times of economic expansion. In short, construction labour is a cost, not a benefit of transit.
Benefits
The economic benefits of transit fall into three categories:
· Congestion management. Increased use of transit in lieu of automobiles leads to improved highway travel times and travel time reliability. Benefits accrue to both the passenger and freight sectors. Whereas the benefits of highway capacity expansion in congested corridors diminish as new demand is induced to use the facility, rail systems show evidence of stabilizing congestion growth over the longer term.
· Affordable mobility. Increased use of transit by low income individuals in lieu of taxis and other higher-price modes creates a gain in consumers’ surplus. Household budgetary resources are released for more highly valued uses, including shelter, nutrition and childcare.
·
Land-use. Well designed, at-grade
rail transit stations create increased property values in high density
neighbourhood environments. Although a
portion of the increased value is attributable to time savings, there is
evidence that transit facilities give rise to non-use benefits in the form of
“amenity value” (value associated with urbanized living arrangements.
The
Economic Benefits of Congestion Management Transit-induced highway delay savings
are commonly valued at between $5.00 and $10.00 per hour for passenger traffic
and $150.00 per hour for freight.
However, new scientific evidence indicates that actual values are some
three-times higher during periods of unreliable and unpredictable traffic
conditions. The research finds that
individuals are willing to pay substantially more for improvements in the
predictability of journey times than for improvements in average speed.[7] While the finding applies equally to highway
travel time improvements induced by roadway expansion and transit, the implications
for bus versus rail investment are important.
As discussed next, this is because rail systems in congested corridors
can yield a more time-stable (sustainable) improvement in travel times and thus
greater travel time predictability.
For the
past several years, researchers of traffic systems have observed that in
congested urban corridors served by a high-capacity transit mode, door-to-door
journey times tend to be equal among modes.
New research postulates an economic theory and empirical evidence that
supports these observations.
In
general, the amount of time it takes to make a trip during peak hours, and the
number of users who decide to use roads versus transit, depend on a number of
factors: the highway capacity, the
costs of using a car versus taking public transit, and individual tastes. In spite of all of these variables, a travel
pattern emerges in congested urban corridors:
the time it takes to complete a journey, door-to-door, tends to be the
same across different modes of transportation.
Furthermore, it is the journey time by the transit mode that seems to
determine the journey time for other modes.
This pattern of converging travel times is predicted by utility theory.[8]
Current
planning practice usually does not allow for the convergence of travel times
and, in fact, proceeds quite differently.
The standard planning practice consists first of predicting the number
of trips that will be made between two locations, based on the number of
inhabitants in both places, the location of jobs, etc. Then, these trips are apportioned among the
different modes based on the traveller's income, personal tastes, and so
on. It is at this point that standard
practice departs from the theoretical and empirical results set forth by
convergence theory. The standard approach does not account for travellers who
move back and forth between modes, much as motorists move between lanes on a
highway in their search for a faster-moving lane. It is the presence of these "explorers" that allows for
the travel times to converge across modes, toward those for rail transit.
Triple Convergence or Travel Time Convergence?
Downs[9] discusses as
a principle of traffic analysis the notion of "triple convergence"
whereby peak hour traffic speeds converge spatially (across the road network),
in time and across modes. Under the
triple convergence principle, an improvement in peak-hour travel conditions on
high-capacity roadways "... will immediately elicit a triple convergence
response, which will soon restore congestion during peak periods, although
those periods may now be shorter".
The prospects for improving transportation performance through transit
investment are just as promising. Downs
states that a new fixed-rail public transit system should initially reduce
peak-period traffic congestion, "...[b]ut as soon as drivers realize that
expressways now permit faster travel, many will converge...onto those
expressways during peak periods".
However,
in congested urban corridors the observed convergence of peak-hour, door-to-door
journey times--by the highway and transit modes suggests that a different
dynamic is at work. If the convergence
dynamic were in effect, a fixed-rail investment would yield an improvement in
journey times by highway. In general,
the convergence of all journey times to the journey time by the rail mode
implies that a change in the performance of transit will result in a stable
change in the performance of highways.
Modal Explorers What
explains the phenomenon of travel time convergence? One claim is that a dynamic relationship exists which parallels
that of a multi-lane highway. Speeds
across lanes tend to be equal because some drivers are "explorers"
who seek out the faster-moving lane thus driving the system to an equilibrium
speed shared by all lanes. By the same
token, in congested urban corridors some travellers and commuters are explorers
who value travel time improvements highly.
They are not committed through circumstance or strong preference to
either mode and they behave as occasional mode switchers.
If the
transit mode has a high-speed, non-stop segment, then the door-to-door journey
time by this mode will be relatively stable and small shifts in ridership will
not significantly effect the journey time by the transit mode. On the other hand, under congested
conditions even a one-half percent increase in highway traffic volume in the
peak period can have a major impact on journey times.
Because
the journey time by transit is stable and determined by the speed of the
high-capacity mode, transit "paces" the performance of the urban
transportation system in the congested corridor. The modal explorers, like exploring drivers on the multi-lane
highway, serve to bring about an equilibrium speed across modes as they seek
travel time advantages across modes.
Travel Time Equilibrium and Modal Choice
While travel time represents a dominant component in the cost of trips,
the generally accepted models of modal choice and the assignment of trips to
networks would not predict travel times to be equal. Rather, the theory behind current practice is that individuals
choose a mode based on income, car ownership, price differentials and modal
preferences which account for non-money factors like convenience, uninterrupted
travel, etc. The persistence of equal,
or near equal, travel times across modes in congested corridors suggests that
current theory fails to correctly capture modal interrelationships in a
multi-modal system.
Empirical Evidence
Economic theory indicates that if congestion is sufficiently high, then
journey times will tend to equal the journey time by the transit mode under the
assumption of growing marginal disutility.
This assumption can be tested empirically by estimating the
relationships between travel time differentials, congestion and additional
factors.
Source of Data In an
ongoing study for the Federal Transit Administration, door-to-door travel time
tests were conducted on 17 urban corridors.
The testing was conducted between February and October of 1995. The corridors were selected based on
criteria which included: congestion,
population density, the existence of mature dedicated-guide-way transit
systems, and public transportation headways.
A list of the seventeen corridors where data was collected is given in
Table 1. The corridors span a range of
moderate to high congestion. In each
corridor random routes of origins and destinations were selected. Survey crews conducted peak hour trips on
the different modes under comparable conditions. Over 1000 trips were recorded and some of the average results are
reported in Table 2. Of the trips
taken, 570 pairs of comparable auto/transit trips were observed. Congestion data for the metropolitan areas
in which each of the corridors was taken from recent Transportation Research
Board studies on urban congestion. The
Metropolitan Planning Organizations in each corridor provided information on
transit headways.
Analysis of Data A
regression analysis of time differentials was conducted in which the absolute
value of the travel time difference between auto vs. transit was regressed
against the metropolitan area congestion index and the transit mode headway
(minutes). As explanatory factors,
congestion and headway do little to explain the variation between each of the
570 trip pairs. This is not surprising
since these variables have no variation within the corridor and transit
mode. However, we observe that the
coefficient for congestion is negative while that of headway is positive and
both coefficients are significant at the 99 percent level. This means that as congestion increases and
as transit headways decrease, the travel times between the automobile mode and
the transit mode become more equal.
Table 1: Strategic
Corridors Measured
|
Corridor |
Modes Measured |
|
Atlanta—I-20 |
Auto, Heavy Rail, HOV |
|
Atlanta—I-85 |
Auto, Heavy Rail |
|
Boston--Mass Pike |
Auto, Commuter Rail |
|
Boston--Southeast Expressway |
Auto, Heavy Rail |
|
Chicago--Midway |
Auto, Heavy Rail |
|
Chicago--O'Hare |
Auto, Heavy & Commuter Rail |
|
Cleveland--Brook Park |
Auto, Heavy Rail |
|
Philadelphia Schuylkill--Bryn Mawr |
Auto, Commuter Rail |
|
Phila. Schuylkill--Upper Merion |
Auto, Commuter Rail |
|
Philadelphia--Wilmington |
Auto, Commuter Rail |
|
Pittsburgh--Parkway East |
Auto, Express Bus |
|
Princeton--New York |
Auto, Commuter Rail |
|
San Francisco--Bay Bridge |
Auto, Commuter Rail |
|
San Francisco--Geary |
Auto, Express Bus |
|
Washington--I-66 |
Auto, Heavy Rail, HOV |
|
Washington--I-270 |
Auto, Heavy Rail |
|
Washington--I-95 Woodbridge |
Auto, Commuter Rail, HOV |
Source:
Federal Transit Administration
Undoubtedly
there are additional factors contributing to the explanation of travel time
differentials, some of which are location-specific while others are associated
with price and other variables.
However, the evidence supports the theory that in congested urban
corridors the growing marginal disutility from time spent travelling causes
door-to-door journey times to converge to the journey time by the
high-capacity, transit mode. Furthermore,
data indicate that reducing transit headways contributes to shorter trip times
and also contributes to a reduction in the time differentials between modes.
Table 2 Door-To-Door Travel Times for Peak Journeys
|
Corridor |
Auto Mode (Minutes) |
Transit Mode (Minutes) |
|
New York,
Jamaica, Queens-Midtown Manhattan |
63.9 |
64.4 |
|
San
Francisco Bay Bridge |
72.3 |
73.1 |
|
Phila.
Schuylkill Expressway--Bryn Mawr |
48.4 |
52.5 |
|
Chicago—Midway |
54.2 |
60.6 |
|
Chicago--O'Hare |
53.9 |
59.3 |
|
Pittsburgh
Parkway East |
38.1 |
42.5 |
|
Princeton--New
York City |
113.4 |
104.9 |
|
Washington--I-270 |
71.9 |
67.4 |
Source:
Federal Transit Administration
Table 3: Strategic Corridor Regression Results
|
Dependent Variable: Absolute
Value of Trip Time Difference (Auto--Transit) |
|
|
Variable |
Coefficient (t-values) |
|
Constant |
15.30 (5.54) |
|
Congestion Index |
-3.48 (-2.45) |
|
Headway |
0.506 (7.80) |
|
All coefficients are significant at the one
percent level |
|
|
Summary
Statistics |
|
|
Number of Observations |
570 |
|
R? |
0.098 |
|
Mean Dependent Variable |
15.68 |
|
F-Statistic |
30.97 |
Source: Federal Transit Administration
The Economic Benefit of Affordable Mobility. Although the use of transit by people from low income households is typically viewed from an equity rather than an efficiency perspective, real resource gains do arise from transit’s availability in this sector. Assessments of the demand curve for transit for low income travelers indicates that the cost of the “next cheapest” modal alternative creates much of their willingness to pay for bus or rail service. The alternative is often taxi, which tends to consume three to four times more of the household budget among poorer households as compared with higher income families.
As
illustrated in Figure 1, transit thus creates a “consumers’ surplus” for the
poor, the cash value of which represents resources available for other
necessities, including housing and food.
The U.S. Federal Transit Administration estimates that the value of this
surplus in the United States lies in the vicinity of $34 billion annually. Compared with total annual outlays (capital
and operating) on transit in the U.S. of $20 billion, this finding suggests
that existing transit systems are economically justified on the strength of
affordable mobility benefits alone. As
shown later, however, this does not appear to be true for new transit investment
Figure 1:
Consumer Surplus Associated with Affordable Mobility
Are Affordable Mobility Benefits Additive to Congestion Management Benefits? The risk of double-counting benefits here is low to moderate. The question turns on the extent to which measured time savings under the congestion management framework accrue to low-income, transit-dependent people, in which case time savings and consumer surplus are simply different ways of measuring the same economic value. To the extent that low income people transfer to transit from taxis or from non-travel activities (induced demand), the mobility benefit they accrue will not be time, but rather cash dollars.
The Economic Benefit of Transit’s Effect on Urban Land-Use. Hedonic studies of how transit stations affect property values reflect both the capitalization of transportation benefits (i.e., the manifestation of delay savings) and any non-use benefits of transit due to improved neighbourhood form and general liveability. Studies performed by the Federal Transit Administration indicate that at-grade rail stations yield in the region of $16.00 greater residential equity value for each foot closer a property is to the station. Findings in San Francisco, for example, indicate that the average home carries $15,000.00 more value for each 1000 feet closer it is in relation to a BART station (other things being equal).
Are Light Rail Community Development Benefits Additive to Congestion Management Benefits? Although improved property values can reflect enhanced access and thus time savings, the hedonic studies reported above indicate that the measured enhancement in land value is often greater than the capitalized value of time savings. This implies that non-use benefits arise from transit, an inference consistent with the hypothesis of “amenity value.” Such “non-use amenity value” refers to property value created by the demand for higher density, walking oriented and less auto-dependent neighbourhoods, neighbourhood attributes promoted by transit. While additional research is warranted on the validity of this inference, urban scholars do observe that some people do indeed choose to live transit-oriented residential locations without any intention of actually using the transit service.
Are Light Rail Community Development Benefits Additive to Affordable Mobility (Consumer Surplus) Benefits? The risk of double-counting benefits here is low to moderate. The question hinges on the propensity of transit-induced increases in property value to reflect the demand for housing by low-income, transit-dependent people in transit-oriented communities. The greater this propensity, the more likely it is that increased property values are, in part, the capitalized value of consumer surplus benefits and thus not additive to the affordable mobility estimates given earlier. To the extent however that the residential re-location decisions of low income households living beyond the immediate vicinity of fixed guide-way stations are rigid, empirically estimated transit-induced increases in property value will be mainly a reflection of other value-enhancing factors (see above).
Return on Transit Investment
The handful of recent studies that apply the Benefit-Cost Framework outlined above provide some evidence of transit’s investment worth. Studies are also indicative of the relative economic worth of bus and rail system investments. The evidence for Cincinnati summarized in Table 4 is typical of other results. Whereas bus improvements yield higher rates of return than rail investments, the absolute magnitude of economic benefits is greater with rail. The difference is due principally to the congestion benefits of rail. As shown in Table 5, delay savings account for more than 60 percent of total economic benefits associated with new rail systems. Bus improvements, on the other hand, yield relatively greater affordable mobility benefits.
Another notable finding is that rates of return in relation to rail systems, while capable of reaching upwards of 10 percent (against an opportunity cost of capital of four percent), are more vulnerable to economic failure than bus investments. A prospective new line examined by HLB in Austin, Texas indicated that the present value of total investment and operating costs would likely exceed the present value of total benefits by $176 million. It is also noteworthy that affordable mobility and community land-use benefits taken together are often insufficient to justify rail system investment. This finding indicates that the economic justification for rail stems principally from a market distortion in the delivery of highway services, namely the absence of road pricing. This finding is significant in the consideration of subsidy, as discussed in the next section.
Are the U.S. findings reported above indicative of transit’s economic value in Canadian cities? The answer rests, among other things, on how similar the two nations are in relation to urban congestion, urban poverty and transit dependence, personal tastes relating to urban form, and the general propensity among car-owners to use transit. On the one hand, the likelihood that congestion and poverty rates are somewhat higher in the U.S. than they are in Canada would suggest that transit delivers relatively more value in the United States. This might be offset, on the other hand, by the higher propensity among the Canadian general public to use transit. While detailed Cost-Benefit transit evaluations in Canada would be needed to draw firm conclusions, the impression of off-setting factors appears to suggest that the U.S. findings are broadly indicative of the Canadian situation.
Table
4: Comparison of Economic Rates of Return for Bus, Light Rail and Highway
Capacity Investments
|
|
Bus Improvement, Region-wide (Cincinnati) |
Light Rail Region-wide (Cincinnati) |
New Highway Capacity (Cincinnati) |
|
Present Value of Total
Cost (Million dollars) |
$522 |
$6,218 |
$1,209.1 |
|
Present Value of Total
Benefits (Million dollars) |
$1,141 |
$10,784 |
$1,365.2 |
|
Net Present Benefits
(Million dollars) |
$619 |
$4,566 |
$1,56.1 |
|
Internal Rate of Return |
27.14% |
8.37% |
4.91% |
Source: HLB Decision Economics, The Economic and Community Benefits of Transportation Options for
Greater Cincinnati, February, 2001
Table
5: Summary of the Economic Benefits of Light Rail Projects in the U.S.
|
Category of Benefits |
Light Rail, Green
Line (Austin, Texas) |
Light Rail, Orange
Line (Austin Taxes) |
Light Rail,
corridor 1-71 (Cincinnati) |
|
Total Benefits
(Million U.S. dollars)
Congestion Management
Affordable Mobility
Community Economic Development Total Cost (Million
U.S. dollars) Net Present Value
(Million U.S. dollars) |
$852.5 $224.0 $293.5 $1,035.4 $334.5 |
$106.5 $32.5 $94.6 $410.0 -$176.4 |
$1,153 $3,23.5 $353.9 $1,043.7 $786.6 |
Source, HLB
Decision Economics, Economic and
Community Benefits of Transportation Options for Greater Cincinnati, February,
2001, prepared for Ohio-Kentucky-Indiana Metropolitan Planning Organization
(and) Light Rail Transit in the Austin
Urbanized Area: A Cost-Benefit Analysis, prepared for Austin Transit
Authority, March, 2000
In 1978, Glaister and Lewis[10] reported a method for estimating the “second-best” subsidy level in relation to mass transit. The premise is, that in absence of “first-best” pricing of highway services (ie, road pricing), lower fares and higher service quality than free markets would supply encourage transfer from private vehicles, alleviating the congestion externality. The quantitative significance of the argument depends on the elasticities of demand for transit, the cross-elasticity of demand between road and transit, and the magnitude of the social marginal costs of congestion associated with passenger cars and with buses. The model has been applied in both the U.K. and the U.S. In those cases existing levels of subsidy were found to be in line with optimal levels. This paper presents the first attempt to apply the model in Canada.
The results, shown in Table 6, indicate that optimal subsidy for urban transit in Canada lies in the range of $5 to $6 billion. This compares with current subsidy levels of about $2 billion annually. Based on the environmental costs of road traffic alone, current subsidy levels appear about right. The social marginal costs of congestion, however, drive economically desirable subsidy level well above current outlays.
Table 6: Second-Best Subsidy Results for Canada (1999)
|
Elements Included Estimated Marginal Social Costs |
Operating Optimal Subsidy (Millions) |
|
Environment, Safety and
Operating costs |
$2,642 |
|
Congestion, Safety and
Operating Costs |
$5,490 |
|
Environment, Safety
Congestion and Operating Costs |
$5,661 |
Source: HLB
Decision Economics Inc (see Appendix A)
Tables 7 and 8 repeat the analysis specifically for Toronto and Montreal. In Toronto, where total subsidy support from all levels of government in 1999 was just under $730 million, optimal subsidy for bus and rail together exceeds $1.7 billion (bus and rail combined). The $1.2 billion optimal subsidy level estimated for Montreal (bus and rail combined) is about four-times the amount of current support reported by Canadian Urban Transit Association[11]. However, since reported subsidy levels for Montreal exclude certain categories of spending, it is most likely that the ratio of optimal subsidy to current subsidy is closer to the number we report for the Toronto Transit Commission.
Table
7: Second-Best Subsidy Results for Toronto Transit Commission (1999)
|
|
Environment, Safety & Operating Costs |
Congestion, Safety & Operating Costs |
Environment, Safety, Congestion & Operating Costs |
|
Optimal Subsidy (Bus) (Millions) |
$317 |
$659 |
$679 |
|
Optimal Subsidy (Rail) (Millions) |
$476 |
$988 |
$1,019 |
Source: HLB Decision Economics (from) Appendix A and
Appendix Table 3.
Table 8: Second-Best Subsidy Results for Société de transport de la Communauté urbaine de Montréal (1999)
|
|
Environment, Safety & Operating Costs |
Congestion, Safety & Operating Costs |
Environment, Safety, Congestion & Operating Costs |
|
Operating Optimal Subsidy
(Bus) (Millions) |
$228 |
$473 |
$488 |
|
Operating Optimal Subsidy
(Rail) (Millions) |
$342 |
$710 |
$732 |
Source: HLB Decision Economics and Appendix Table 3.
Risk Analysis
The results given above are especially sensitive to two factors, (i) estimates of the cross-elasticity between auto use and transit fare and service levels; and (ii) estimates of social marginal costs of automobiles and buses during peak period. Unfortunately, these estimates are also especially uncertain (see Text Box 1).
Box 1: Marginal Social Costs of Automobiles
and Buses Measures of the
marginal social costs of auto and transit use are scarce and uncertain. The principal uncertainties arise in
relation to the congestion and environmental costs of automobile and bus
traffic. Congestion Costs. Mohring (HLB) and Anderson (University of
California) provide one of the few estimates of marginal operating and
congestion costs in the U.S. They
report costs of $0.37 per passenger kilometre for automobiles and $0.29 per
passenger kilometre for buses (each in heavy traffic). There are no comparable estimates for
Canadian cities. Adding in the social
costs of highway accidents in Canada to the marginal operating and congestion
values given by Mohring and Anderson provides a rough guide to overall
marginal social costs of congestion, safety and operating costs under
congestion levels and values of time similar to those of Canadian urban
areas. The values are $0.29 per
passenger kilometre for automobiles and $0.46 per passenger kilometre for bus
(shown in the first row of the table below). Environmental Costs. There is even less evidence regarding the marginal
environmental costs of automobiles and buses. Canada’s Royal Commission on Passenger Transportation reported
only the average environmental cost of congestion. The Commission did so moreover only in relation to intercity
auto and bus movements. Environment Canada reports that urban auto and
transit traffic generate (respectively) 2.4 and 1.7 times more green house
gases than intercity traffic. If we
assume average and marginal environmental costs to be equal, factor these
numbers up by 2.4 and 1.7 for the urban/intercity differential in emissions,
and add in the marginal operating and safety costs of auto and transit (which
we assume to be equal to average costs in Canada) we obtain a rough estimate
of the marginal social cost of environmental, safety and operating cost. The values are $0.23 per passenger
kilometre for automobiles and $0.22 for transit (shown in the second row of
the table below). Overall Marginal Social
Costs. The third row of the table combines the marginal
social costs of safety and environment estimated above with the marginal
operating and congestion cost estimates of Mohring and Anderson. This gives our “best estimate” of overall
marginal social costs of $0.30 per passenger kilometre for transit and $0.47
for automobiles. As shown in the
table, the implication is that the marginal environmental costs of automobile
and bus traffic are very low in relation to congestion and other cost
factors. Risk Analysis. The estimates are clearly uncertain. The findings reported in the main text
reflect the third row in the Table below.
The risk analysis reflects subjective range of plus or minus 20
percent (see text Tables 4 and
5). The risk analysis assumes that
the estimates are equally likely to be too low and too high (the probability
distributions are “symmetrical”). Marginal Social Costs ($
per km) Elements Included in Estimated Cost
Transit Auto Congestion; Safety; Operating Costs $0.29 $0.46 Environment; Safety; Operating Costs $0.23 $0.22 Congestion; Environment; Safety; Operating Costs $0.30 $0.47 Sources: Canadian Urban Transit Association, Canadian Transit Fact Book: 1999 Operating Data 1999: Ministry of Supply and Services, Final Report of The Royal Commission on National Passenger Transportation, 1992: Lewis, David, and Fred L. Williams, Policy and Planning as Public Choice, Ashgate Publishing Limited, England, 2000: Environment Canada, “Foundation Paper on Climate Change – Transportation Tables”, December , 1998: HLB Decision Economics, “Cost Benefit Study on NHS in Canada” Inter-provincial Council of Ministers of Transport, 1998: Mohring, H. and D. Anderson, “Congestion Costs and Congestion Pricing,” University of California, Irvine (Working Paper), March, 1996 |
We conducted a risk analysis based on subjective probability ranges for the cross-elasticities and marginal social costs (See Figures 3, 4 and 5 and Table 9). Based on a Monte Carlo simulation, the risk analysis indicates an 80 percent probability that the optimal transit subsidy level for urban Canada lies above $4.5 billion a year. Shown in Figure 2, the risk analysis indicates an even higher probability (greater than 90 percent) that the existing level of subsidy (of $2 billion annually) lies beneath that required for an efficient allocation of transportation resources. Thus, even though a great deal of uncertainty exists in relation to key assumptions, the likelihood is quite high that, from an economic perspective, spending on urban transit in Canada today is too low.
Figure
2: Simulation and Risk Analysis Results
Figure
3: Cross-Elasticity of Demand for Peak Auto with respect to Peak Transit Fares
|
|
Median |
Lower 10% |
Upper 10% |
|
Elasticity |
0.02500 |
0.02000 |
0.02625 |
Source: Oum, Tae Hoon, Transport Economics, Korea Research
Foundation, 1995; Lewis, D. and William F.W., Policy and Planning as Public Choice: Mass Transit in the United States,
Ashgate, 1999
Figure
4: Marginal Social Cost of Peak Private
Vehicle Travel
|
|
Median |
Lower 10% |
Upper 10% |
|
$ Per km |
$0.47 |
$0.37 |
$0.56 |
Sources: Canadian Urban Transit Association, “Canadian
Transit Fact Book: 1999 Operating
Data”, 1999: Minister of
Supply and Services, “The Final Report of
The Royal Commission on National Passenger Transportation”, 1992: Lewis,
David, and Fred L. Williams, “Policy and
Planning as Public Choice”, Ashgate Publishing Limited, England, 2000:
Environment Canada, “Foundation Paper on
Climate Change – Transportation Tables”, December Ottawa, 1998: HLB
Decision Economics, “Cost Benefit Study
on NHS in Canada”, 1998: Mohring, H. and D. Anderson, “Congestion Costs and
Congestion Pricing,” University of California, Irvine (Working Paper), March,
1996.
Figure
5: Marginal Social Cost of Peak Transit
Travel
|
|
Median |
Lower 10% |
Upper 10% |
|
$ Per km |
$0.30 |
$0.23 |
$0.35 |
Sources: Canadian Urban Transit Association, “Canadian
Transit Fact Book: 1999 Operating
Data”, 1999: Minister of
Supply and Services, “The Final Report of
The Royal Commission on National Passenger Transportation”, 1992: Lewis,
David, and Fred L. Williams, “Policy and
Planning as Public Choice”, Ashgate Publishing Limited, England, 2000:
Environment Canada, “Foundation Paper on
Climate Change – Transportation Tables”, December Ottawa, 1998: HLB
Decision Economics, “Cost Benefit Study
on NHS in Canada”, 1998: Mohring, H. and D. Anderson, “Congestion Costs and
Congestion Pricing,” University of California, Irvine (Working Paper), March,
1996.
Table 9 : Other Assumptions in the Risk Analysis
|
Variable Description |
Median |
|
Elasticity of Demand for Peak Transit Travel with respect to Peak
Fares |
-0.35 |
|
Elasticity of Demand for Off-Peak Transit Travel with respect to
Off-Peak Fares |
-0.87 |
|
Cross-Elasticity of Demand for Peak Auto Travel with respect to
Off-Peak Fares |
0.001600 |
|
Cross-Elasticity of Demand for Peak Transit Travel with respect to
Off-Peak Fares |
0.029000 |
|
Cross-Elasticity of Demand for Off-Peak Transit Travel with respect
to Peak Fares |
0.040000 |
|
Operating Costs of Off-Peak Transit Travel, $/km |
$0.220437 |
|
Demand for Peak Auto Travel, Million km |
149,500 |
|
Demand for Peak Transit Travel, Million km |
8,125 |
|
Demand for Off-Peak Transit Travel, Million km |
4,375 |
Does the likelihood that Canada under-spends on transit signify the need for federal financial participation in the subsidization of municipal transit systems. The matter turns (inter alia) on transit’s strategic importance from a national perspective. The main argument in favour of a federal role is that congestion is a nation-wide phenomenon and a major barrier to national productivity growth. In light of the strategic role of productivity growth in promoting national economic performance, under-spending on transit can be viewed a matter of federal concern.
A common counter-argument is that provincial and local governments often take the lead role in nationally important policy functions, especially in cases where, like traffic congestion and urban mobility, problems vary in size and type from one region to another. Non-federal governments might answer that, notwithstanding this logic, they do not have the fiscal capacity to align transit spending with economically optimal levels. In Toronto alone, the results of this study suggest that subsidy to the Toronto Transit Commission would need to increase from about $700 million a year to $1.7 billion a year. In Montreal, subsidy would need to increase three to four-fold. The “affordability” argument is ultimately a matter of priorities and can only be resolved in that context.
Another concern with federal intervention is the possibility that federal contributions would be used to substitute for rather than complement provincial and local spending. Nothing of national economic significance would be gained by federal financial intervention if non-federal governments were to back-out their contributions proportionately. In the United States, where federal grants and contributions are tied to local matching formulas, empirical studies indicate some “leakage” of state and local dollars in response to federal funding. Such leakages can be minimized, though potentially not eliminated, with innovative federal control mechanisms.
