Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 51
Guidelines for Evaluation and Performance Measurement of Congestion Pricing Projects 51
however, requires establishing a good working relationship with that agency and coordinating
data collection efforts.
Economics
Economics is not generally assessed for a variably priced toll facility project. Effects on local
businesses and regional competitiveness are of extreme interest in a region implementing an area
or cordon pricing project, but this is not normally the case with single toll facilities. Nonetheless,
improved travel-time reliability along highly traveled corridors, such as the 407 ETR in Toronto,
would be expected to have a positive economic effect. However, it is extremely challenging to
measure the precise effect of an individual transportation improvement on regional economic
trends. This type of analysis would be more likely to rely on the results of economic models,
which would allow a comparison to be made between model outputs and data collected on
regional economic activity and real estate prices.
Land Use
Performance measures to evaluate a variably priced toll facility's effects on land use are not
commonly used in practice and are not generally recommended by these guidelines. Nonethe-
less, a facility sponsor may want to consider these measures (such as residential or commercial
land use trends) if land use is a particular issue of concern.
3.4 Performance Measurement for Cordon
and Area Pricing Projects
Given their extremely high visibility and sensitive nature, performance measurement for cor-
don and area pricing programs is especially important and integral to their ongoing success. (The
distinction between cordon and area pricing is provided in Sections 1.2.3 and 3.4.2.) Although
the use of congestion pricing on individual facilities or lanes affects travel pattern in given cor-
ridors, cordon and area pricing programs have profound effects on travel patterns across entire
regions. In addition, cordon and area pricing programs are also likely to have important effects--
both real and perceived--on other important issues, such as regional emissions and air quality,
business impacts, and economic competitiveness--issues that are not likely to be high-priority
concerns with other forms of congestion pricing and are likely to require creative approaches in
order to be monitored in a meaningful way.
Given the regional nature of their influence on travel patterns and congestion, performance
monitoring programs for cordon and area pricing projects should involve the collection of com-
parable sets of data in different locations around the region, both within the pricing zone and
outside it. Although the benefits in terms of reductions in traffic volumes and congestion and
increased travel speeds will be greater in the pricing zone than outside it, the benefits may actu-
ally be the greatest at pinch points leading into the pricing zone, including bridges, tunnels, and
major arterial streets or highways. Monitoring programs for cordon and area pricing projects
should focus on these types of locations and generate data showing how the benefits of pricing
accrue to surrounding areas as well as the pricing zone itself. Project sponsors should also con-
sider collecting baseline data in any neighborhoods or communities that may oppose the imple-
mentation of cordon or area pricing programs and then monitor appropriate metrics such as
traffic and environmental conditions in those locations once the system is implemented. In many
cases, it is likely that conditions could improve, allowing project sponsors to use the performance
monitoring data to garner support for the pricing program.
Another area of concern with the use of cordon or area pricing is the ability for existing tran-
sit infrastructure to accommodate the increased passenger loads that would be expected as
OCR for page 52
52 Evaluation and Performance Measurement of Congestion Pricing Projects
motorists shift their trips to transit.4 Given the heavy utilization of rail transit and the longer lead
times needed to expand rail capacity, preparations for the area pricing programs in both Lon-
don and Stockholm included the purchase of new bus fleets to augment transit capacity. This
was a particularly strategic move because increased travel speeds, particularly within the pricing
zones themselves, made bus travel far more attractive than in the past, once the pricing systems
were activated. Performance monitoring programs for cordon or area pricing schemes should
track utilization, crowding, and travel times on all relevant rail and bus lines likely to be affected
by the new programs.
Equity is also an inevitable concern with cordon and area pricing programs. Together, the high
cost of parking and excellent transit availability in locations where cordon or area pricing may
be introduced limit the number of people in lower income groups who make regular trips to
these areas by automobile. However, residents in areas with poorer transit access or other low-
to moderate income earners who happen to have free parking at their places of employment may
be affected by the introduction of cordon or area pricing. If these types of concerns arise, it may
be helpful to develop specific performance metrics to track how these communities are affected
by the pricing program. Social equity is discussed in greater depth in Section 3.1.4.
Lastly, by their very definition, cordon and area pricing programs will likely require the instal-
lation of new toll collection systems and technologies, including character recognition systems
capable of reading license plates, as well as back office accounting systems to process payments
and manage accounts and customer service centers. The performance monitoring programs for
cordon and area pricing systems should include appropriate parameters to track the perfor-
mance of these systems, the accuracy of the data collected, and the extent to which desired per-
formance levels are achieved.
The findings and recommendations that follow on performance monitoring for cordon or
area pricing projects are informed by the case studies prepared for the congestion charging pro-
grams operating in Singapore, London, and Stockholm, which represent the three largest appli-
cations of cordon or area pricing in the world. The findings and recommendations are also sup-
plemented by the experiences of the guidelines' authors in supporting the exploration of the
introduction of congestion pricing in Manhattan, together with industry standards and best
practices. In all cases, the use of cordon or area pricing should be expected to be a highly sensi-
tive issue and of interest to elected officials and community and stakeholder groups of all types.
As such, performance monitoring efforts for these projects should involve extensive exploration
of public opinions and concerns. The information gathered through this outreach process
should be used to identify a tailored set of performance measures that track parameters of par-
ticular interest to different stakeholder groups, as well as to identify those analysis areas likely to
be affected by the use of cordon or area pricing.
3.4.1 Distinguishing Characteristics of Cordon
and Area Pricing Programs
As with the other forms of pricing, some variables distinguish cordon or area pricing schemes
from one another and are likely to influence performance monitoring programs.
Toll Collection Technologies
Two primary types of technologies can be used to collect the entry fees associated with cor-
don or area pricing: transponder-based ETC systems and camera-based character recognition
4The presence of comprehensive rail and bus transit networks is essential for any metropolitan region considering the use of
cordon or area pricing.
OCR for page 53
Guidelines for Evaluation and Performance Measurement of Congestion Pricing Projects 53
systems. In certain cases, one or the other technology may be used exclusively, and in others, the
system may use both. Operating costs for character recognition systems are likely to be higher
than those of transponder-based systems, but it is possible that pricing schemes could charge dif-
ferent rates for entering the pricing zone, depending on which technology is used. Camera-based
technologies are also more likely to raise privacy concerns. Performance monitoring programs
for cordon or area pricing schemes may need to be developed to track and compare the perfor-
mance of these different toll collection technologies in terms of accuracy, reliability, cost, and
public perception.
Cordon Versus Area Pricing
There are two approaches for collecting entry fees with cordon and area pricing programs. The
first is the cordon approach where motorists are charged a fee each time they enter the pricing
zone, regardless of the number of trips made. This model is used in Singapore. The other approach
is to charge motorists a single fee to enter the pricing zone on multiple occasions during a des-
ignated period, such as 7:00 a.m. to 6:00 p.m., Monday through Friday, as is the case in London.
The back office accounting programs used to operate cordon and area charging schemes need to
be capable of making these distinctions, and it is likely that performance monitoring programs
will also need to be capable of tracking the net number of trips made by individual vehicles of
different types (e.g., private versus commercial) on a daily basis in order to gain a full under-
standing of the ways in which the charging schemes affect different types of motorists.
Fixed Versus Variable Price Rates
As these guidelines are being written, two of the three major cordon or area pricing
schemes--Stockholm and Singapore--vary their rates by time of day on a fixed schedule. It is
possible that, in the future, new cordon or area pricing schemes could vary entry fees in real time
based on actual travel conditions, with higher rates charged during periods of higher congestion.
The rationale for using variably priced fees is to use higher toll rates as a further incentive to
encourage motorists to make trips by alternative modes or during non-peak periods. Perfor-
mance monitoring programs for cordon or area pricing programs using variable rates need to
be capable of demonstrating the effects of changing toll rates on travel behavior. Regardless of
the structure of the entry fee, performance monitoring programs for all cordon and area pricing
programs should also be capable of tracking travel conditions by time of day--including those
days and periods where no charge is levied--in order to provide a full understanding of how
pricing influences travel patterns.
Intra-Area Charges
In some cases, cordon and area pricing programs may involve levying a fee for vehicular trips
made entirely within the pricing zone in addition to those that originate outside the zone. In sit-
uations where cordon or area pricing programs levy different fees for different types of trips (or
trips made by residents who live in or next to the priced zone), performance monitoring pro-
grams need to be able to track the number of trips for each of the different fee structures and
enable analysts to assess how these different fee policies influence overall travel behavior.
Geographic-Specific Concerns
In certain cases, the implementation of cordon or area pricing can result in comparatively
severe effects on residents in certain geographic areas. This is the case in Stockholm, for instance,
for people living on the island of Lindingö, for which the only road access involves traveling
through the pricing zone in the City of Stockholm. Given that there are no alternative routes for
local residents and their visitors to use, trips to and from Lindingö are free, provided vehicles
enter or exit the pricing zone within 30 minutes of arriving or departing from the island. In cases
where there are certain locations where local residents are provided with equity-based discounts,
OCR for page 54
54 Evaluation and Performance Measurement of Congestion Pricing Projects
performance monitoring programs should be able to provide separate data tracking the effects
of the charging scheme on populations who qualify for these types of discounts in order to
understand how they affect local travel patterns and the extent to which local residents find the
discounted fees fair and acceptable.
Level of Public Interest
Perhaps to a greater degree than other forms of congestion pricing, the level of public interest
in cordon or area pricing schemes can be expected to be extremely high. Performance monitor-
ing programs for these schemes should (1) provide comprehensive information on all the ben-
efits of congestion pricing and (2) be tailored to address specific areas of public concern. As a
result of the regional nature of their effects, as well as the potential for meaningful improvements
in congestion levels and regional emissions and the heightened level of interest and concern, per-
formance monitoring programs for cordon and area pricing schemes should be robust and com-
prehensive in order to demonstrate their multitude of potential effects on the region and to gain
support for them.
3.4.2 Selection of Performance Measures for Cordon
and Area Pricing Programs
This section provides specific factors for consideration, summaries of experience, and recom-
mendations on the selection of performance measures for cordon or area pricing schemes. The
section's organization follows the order of the eight areas of evaluation identified earlier in the
study. Evaluation areas are tied directly to the goals of a project. Specific project goals can be for-
mulated and measured within the context of the evaluation areas. These evaluation areas and the
full set of identified performance measures were introduced in Table 3-1 in Section 3.1.5.
The number of operational cordon or area pricing schemes is limited in practice. All are cur-
rently located outside the United States, and the three most extensive ones (as measured by geo-
graphic extent and population served) were selected for close examination as part of the research
behind these guidelines. The number of distinct performance measures captured within each
evaluation area is shown in Table 3-11, along with those measures used by the schemes studied.
Because of this limited subset, it is more difficult to conclusively extract performance measures
most commonly used in practice to a set of general guidelines than in the case of variably priced
managed lanes. These Guidelines' recommendations take this into account and also draw from
the authors' experience with the attempted implementation of cordon pricing in New York City
and other industry knowledge.
The full spectrum of performance measures used in practice by at least two of the three area
or cordon pricing schemes examined is shown in Table 3-12. As with the performance measures
identified for variably priced managed lanes and variably priced toll facilities, the table also
identifies whether the measures are generally applied in an operations or validation capacity,
Table 3-11. Total performance measures by evaluation area.
Total Measures Measures Used by Measures Used by
Identified 2+ out of 3 Schemes 1 out of 3 Schemes
Traffic Performance 20 7 9
Public Perception 15 1 3
Facility Users 14 3 1
System Operations 15 3 8
Environment 3 2 1
Transit 7 3 3
Economics 9 3 6
Land Use 2 0 2
OCR for page 55
Guidelines for Evaluation and Performance Measurement of Congestion Pricing Projects 55
Table 3-12. Performance measures in practice--cordon and area pricing (2+ out of 3 schemes examined).
Total No. of Purpose Importance
Evaluation Area Performance Measures
Measures Used (No. of Measures) (No. of Measures)
Operations Validation Key Secondary
Traffic Performance Speed & Travel Time Travel times 3 3 1 1
Traffic Performance Speed & Travel Time Speeds/ average speed 2 1 1 1
Traffic Performance Volume Vehicle volume (hourly/daily/weekly/monthly) 2 2 2
Traffic Performance VMT/VKT VMT/ VKT 2 2 1 1
Traffic Performance Congestion Delay/ wait times 2 2 2
Traffic Performance Mode Share Mode share (SOV, HOV, transit) 2 2 1
Traffic Performance Bike/Ped Bike/ped traffic counts 2 2 1
Public Perception Social Impacts Specific activities/populations 2 2 1
Facility Users Trip Characteristics O-D/ travelshed determination 2 1 2 2
Facility Users User Characteristics Vehicle classification 2 2 1
Facility Users Trip Characteristics Trip purpose 2 2 1
System Operations Finance Revenue (toll/ charge) 2 2 1
System Operations Finance O&M Cost 2 2 1
System Operations Safety Collisions/ accidents 2 2 1
Environment Air Quality NAAQS criteria pollutants/ VOCs 2 2 1
Environment Air Quality GHG/ CO2 2 2 1
Transit Performance Travel time/on-time/excess wait 2 2 1
Transit Performance Average speed 2 2 1
Transit Occupancy Ridership/ boardings 2 2 1
Economics General Benefit-cost analysis 2 2 1
Economics Business Impacts General performance/openings/closings 2 2 1
Economics Business Impacts Retail traffic & sales 2 2 1
and whether they play a key (primary) or secondary role in a typical performance evaluation
program.
In the case of area and cordon pricing programs, however, the distinction between operations
and validation is less significant than for the other two forms of congestion pricing. Scheme
sponsors may choose to use any particular performance measure in either an operations or val-
idation capacity, given the complexity of these programs' implementation and the high level of
public scrutiny they assuredly will undergo. That is, any one measure may inform an operational
change on an ad hoc or systematic basis, and similarly, any one measure may help to communi-
cate to users and observers a scheme's successful (or unsuccessful) achievement of goals and pre-
dicted benefits. Nonetheless, several performance measures (specific to Singapore's Electronic
Road Pricing program) are indicated to be used in an operations capacity because they feed
directly into an established periodic review of scheme function; if certain thresholds or charac-
teristics are observed, adjustments to toll rates or an expansion of the priced zone can be trig-
gered. It can be assumed that the potential application of dynamically variable area or cordon
charges would require the selection of operations-based performance measures as real-time
inputs to a pricing algorithm.
In addition to the measures in Table 3-12, many other performance measures are used by
one of the three facilities, 33 measures in total, as indicated in Table 3-11. These measures are
listed in Table 3-13. Again, given the wide variety of special considerations necessary for suc-
cessful implementation and sustained operation of an area or cordon pricing program, these
measures are also presented (without Purpose or Importance distinctions) because any one per-
formance measure may be critical for project sponsors to use. Generally though, those mea-
sures captured in at least two of the three examined schemes are detailed further in the sections
that follow.
Nonetheless, as with the other two forms of pricing, because of the limited sample size of
existing programs, as well as the specialized nature of these schemes, other performance measures
OCR for page 56
56 Evaluation and Performance Measurement of Congestion Pricing Projects
Table 3-13. Performance measures in practice--cordon and area pricing
(1 out of 3 schemes examined).
Evaluation Area Performance Measures
Speed & Travel Time LOS
Speed & Travel Time Travel time savings
Speed & Travel Time Cost of delay/ VOT
Volume Person volume (hourly/daily/weekly/monthly)
Traffic Performance Congestion Congestion coefficient
Occupancy Avg. vehicle occupancy (auto)
Parking Park-n-ride activity (lot counts)
Parking Off-street parking activity (counts/occupancy)
Parking On-street parking activity (counts/occupancy)
Awareness Of the facility/general/how much?
Public Perception Acceptance General/fairness/equity
Effectiveness Congestion reduction
Facility Users Trip Characteristics Trip length
Finance Total transactions
Finance Average toll/ highest toll
Finance Revenue (fee)
Enforcement Violations/citations/fines
System Operations
Customer Service Inquiry activity (call, email)
Customer Service Performance (quantitative measures)
System Function Facility availability
System Function Equipment availability
Environment Noise Noise levels
Occupancy Average vehicle occupancy
Transit Finance Farebox revenue
Service Quality/satisfaction/reliability
General Gross regional product/ economic indices
Business Impacts Specific sectors/services/populations
Business Impacts Business costs and prices
Economics
Business Impacts Tourists/ visitors
Property Residential sales/rentals/values
Property Commercial sales/rentals/values
Residential Housing decisions
Land Use
Commercial Business locations
not listed in Table 3-12 or Table 3-13 could be significant or necessary to collect, based on the
goals set for a particular area or cordon pricing scheme. These measures may not have been cap-
tured by these guidelines' research; however, the issues discussed for each evaluation area can
be applicable to those performance measures not identified.
Traffic Performance
Traffic performance describes the fundamental purpose of a roadway network: its ability to
provide mobility to people and goods. An important distinction among cordon and area pric-
ing programs compared to variably priced managed lanes or toll facilities is the greater empha-
sis placed on including transit, bicyclists, and pedestrians among the users of the roadway net-
work and measuring "traffic" performance for these modes. In general, traffic performance is
measured by various traffic engineering measures answering the how much/many?, how fast?,
and by what mode? questions pertaining to the roadway network.
OCR for page 57
Guidelines for Evaluation and Performance Measurement of Congestion Pricing Projects 57
Representative Traffic Performance Goals. Primary goals of area or cordon pricing pro-
grams involve traffic performance. Achieving congestion reduction in a city center or central busi-
ness district (as well as its surrounding areas and routes into it) is one prominent example. This
goal, in turn though, may be further characterized on a more "measurable" basis or within a con-
text that better resonates with users or those interested in improved performance. To that end,
the goal of reducing congestion could more specifically be stated as reducing the volume or
extent of peak-period congestion, improving vehicular access (specific to commuters or goods and
service providers), or improving travel time reliability into the priced zone. Accomplishing this
goal may require a shift in travel time to less congested periods, a shift to an alternate mode (e.g.,
transit, bicycling, or walking), or not making the trip at all. As with variably priced managed lanes
and toll facilities, goals related to reliability can be subjective and dependent on location-specific
contexts. Further specification by project sponsors may be required.
What Are the Traffic Performance Measures? Measures of traffic, as indicated in Table 3-1,
include vehicle and person volumes, speeds and travel times, mode share and vehicle occupan-
cies, vehicle miles traveled, and indicators of congestion, such as delay, queue lengths, and spe-
cially developed coefficients comparing specific metrics during congested and uncongested con-
ditions. Other measures that incorporate traffic include bicycle and pedestrian measures and
parking, potentially significant considerations for area or cordon pricing schemes. Transit
performance, closely tracked with these programs, is captured in its own evaluation area.
Nearly all the metrics captured in the research for these guidelines have been applied to mea-
sure traffic performance, indicating a broad range that may offer project sponsor utility. Research
has shown that key performance measures of traffic for area and cordon pricing depend signif-
icantly on scheme context. The extent of the scheme's physical coverage, existing roadway con-
figuration, policy and method for charging a fee, and many other issues can all affect the impor-
tance attached to particular traffic performance measures. In one example, the priced zone may
contain a mixture of low-speed city streets, arterials, bridges, tunnels, and highways, unlike vari-
ably priced managed lanes or toll facilities, which are uniform road type.
How Are Traffic Performance Measures Applied?
Volumes. As with variably priced managed lanes and toll facilities, traffic volumes are criti-
cal to understanding system usage (the system in this case being both the priced zone itself and
the surrounding region that may be directly or indirectly affected). Typically these volumes
would be measured at the cordons (boundary) of the priced zone, effectively measuring the total
volume of "system" users inside the zone. Volumes are also likely to be measured at any num-
ber of other critical locations both inside and outside the priced zone to assess the scheme's
effect on particular roads or corridors of interest, especially those known to be highly congested
and targeted for relief. These are likely to be the most heavily traveled routes that lead to the
priced zone. Shifting traffic volumes and patterns are also likely to occur because of users seeking
alternate routes to avoid the charge or parking near the boundary to take an available alternate
mode, such as transit, into the zone. Volumes can be measured using system equipment installed
at the zone's cordons, existing loop detector or camera infrastructure, or through manual counts
where these options may not be available, such as streets near the zone boundary that may see a
spike in traffic from those avoiding passing into the zone.
Speeds/Travel Times/Delay. Traffic speeds measured along specified corridors or averaged
within specified zones help inform common traffic performance goals, as do travel times along
defined routes or from identified origins and destinations. Speeds and travel times often are used
to indicate reliability for journeys into the priced zone. A similar metric that represents speed
and travel time in a reciprocal fashion is delay. The difference between actual speeds and travel
times compared with a baseline accepted speed or travel time represents delay. Reduced delay
OCR for page 58
58 Evaluation and Performance Measurement of Congestion Pricing Projects
can indicate improved traffic performance. Speeds along a sin-
Example: Using Speeds to Adjust Pricing Policy
gle corridor or crossing the zone's cordons can be calculated
using system equipment (ETC transponder equipment and/or
The pricing policy for Singapore's Electronic Road cameras); otherwise a probe vehicle or other proxy equipped
Pricing (ERP) is reviewed on a 3-month cycle, tak- with GPS (such as taxis) would be required to capture average
ing into account a wealth of collected data and speeds across more complicated networks or within a defined
computed traffic engineering metrics based on zone.
speeds. Speed-flow analyses are performed for
all travel routes (expressways, major arterials, Vehicle Miles Traveled. Given a need to aggregate traffic
and minor arterials) to examine congestion levels performance across an often extensive roadway network,
relative to target LOS. This review duration is rather than just single lanes or corridors, vehicle miles traveled
considered optimal to allow enough time for provides another means to measure traffic performance. VMT
traffic patterns to readjust--passing through a requires calculated estimates from other traffic data, such as
transient period and accounting for altered volumes or extrapolations from volunteer vehicles outfitted
driver behavior. A formal process is followed to with GPS.
make an adjustment to the ERP charge schedule.
Approvals are required from the Minister of Other Modes. Encouraging the use of alternative modes to
Transport, and the new rates are formalized access the priced zone is a primary goal measured through traf-
through appropriate legal documents or law. fic performance. In this respect, transit usage (as noted in the
Transit evaluation area) as well as pedestrian and bicyclist
counts measured at the zone's cordons can be applied.
Parking. Although captured for only one priced zone, on-
and off-street public parking counts can provide a good secondary indication of improved traffic
performance. Research has shown that a significant percentage of traffic volume in city centers is
caused by people searching for parking. This means that not only does parking volume provide
an indication of reduced traffic volume as a whole, but it can help provide an indication of fur-
ther congestion reduction resulting from increased ease in obtaining parking. Outside the priced
zone, parking counts are of interest to measure the extent to which users are avoiding entry into
the zone by simply parking outside it. This information can help identify areas for cordon
adjustment or the need for policies to avoid oversubscribing parking (and potentially roadway
capacity) near the zone's boundary.
Public Perception
Area and cordon pricing have been deployed in only a few select cities--none of which is in
the United States as these guidelines are being developed. Because of their untested application
in the United States and the dramatic effects these schemes can be expected to have on how
mobility and accessibility are both perceived and managed, perhaps with more significance than
any other evaluation area, obtaining public buy-in to implement an area or cordon pricing
scheme will require positive public perception. The public's knowledge of a program's purpose
and acceptance of it as a new paradigm for managing access to the selected priced zone are crit-
ical to address prior to implementation. The roles of area and cordon pricing facilities before and
after implementation, along with satisfaction with the service these schemes provide, are char-
acterized qualitatively through public perception.
Representative Public Perception Goals. Gauging public perception is at the heart of goals
that seek to validate an area or cordon pricing project--both before and after implementation.
Representative goals may include achieving acceptance or sustaining a prescribed level of satis-
faction with the facility's operation. Specific targets of the perception of the scheme's effect on
congestion reduction, equity, or social impacts within the priced zone can be established and
tracked. In addition to its relation to public perception, the special case of equity is discussed in
depth in Section 3.1.4.
OCR for page 59
Guidelines for Evaluation and Performance Measurement of Congestion Pricing Projects 59
In general, measuring public perception is an attitudinal exercise that requires an appropri-
ate instrument such as surveys, focus groups, or interviews. Clearly, public outreach becomes a
prime factor in establishing these goals and measuring their achievement. A detailed discussion
of integrating performance evaluation and public outreach, including means of collecting atti-
tudinal information, is provided in Chapter 4 of these guidelines. Provided here are details of the
most relevant performance measures for capturing and quantifying public perception.
What Are the Public Perception Measures? Public perception measures (as itemized in
Table 3-1) focus on awareness, acceptance, and satisfaction. Among all three of these measures,
specificity can range from the very broad to the more explicit. For example, awareness of a
scheme's features (e.g., hours of operation, extent, and exceptions to the charge), planned charge
adjustments, or future cordon expansion can be queried. Similarly, acceptance and satisfaction
measures can be general or specific.
One additional public perception measure found uniquely among area and cordon pricing pro-
grams relates to gauging a scheme's effects on specific activities or populations. Activities could
be industry, commercial, or tourist, for example, while particular populations could include the
elderly, schoolchildren, or specific types of workers. It is somewhat surprising that this public per-
ception measure was the only one found to be common among at least two of the three schemes
examined for these guidelines' research. However, this finding may be more of an indication of
how public perception measures must be specifically tailored to each program's application
rather than a lack of applicability, leading to unique sets of measures for any one particular
scheme. Additionally, sponsors' performance monitoring programs often focus on the results of
post-scheme implementation and report less on their proposed implementation, resulting in a
smaller number of public perception measures employed than expected.
What is most difficult about gauging public perception, however, is that there are no "loop
detectors" for measuring it. That is, to make measurements that are inherently qualitative or sub-
jective, a different set of tools is required, those that capture attitudes, as detailed in Chapter 4.
In addition, many measures are stakeholder group-specific and must be tailored to a specific
issue of significance.
How Are Public Perception Measures Applied? All public
perception measures can be characterized as serving a valida-
tion capacity, but could very well lead to operational decisions
as well, including significant modifications to a scheme's extent
(see the associated Example). A sponsor contemplating the Example: Public Perception Leads to
implementation of an area or cordon pricing scheme may view Significant Operational Changes
certain public perception measures as key to the facility's per- A major proposed change to London's Conges-
formance evaluation program if, for example, a particular issue, tion Charge in 2010 was the elimination of the
such as user equity, is expected to be highly visible. Addition- Western Extension, which had effectively dou-
ally, results of public perception measures may dictate necessary bled the original Central London charging zone
changes to customer service functions or public communica- when added in 2007. The proposed retraction
tion policies. was initiated by Mayor Boris Johnson, elected in
2008, and a public vote/survey was used to
Survey instruments, focus groups, or interviews are generally
inform the decision. Public perception of effects
used to collect data for public perception measures. Generally
to the local economy and the zone's residents
speaking, these measures are more demanding and costly to
were the impetus for the operational change--
collect and synthesize because of the user-specific, manual col-
despite measurable reductions in traffic,
lection process required of attitudinal information. Because of
increased use of alternative transportation
this, their collection is often done on either a "before-and-after"
modes, and improvements to the environment.
or periodic basis. Surveyed public perceptions can be collected
prior to the start of an area or cordon pricing program, either
OCR for page 60
60 Evaluation and Performance Measurement of Congestion Pricing Projects
once or in several waves, and compared with similar results after implementation. Once opera-
tional, it may be desirable to continue to collect these types of measures on a periodic basis, such
as annually or biannually, or as resources allow. Before-and-after surveys may focus on more mar-
ket research, acceptance, and awareness issues, while periodic, post-opening-day performance
measurement will likely focus on user satisfaction.
What is important to keep in mind when formulating measures of public perception is that
they should address issues of public concern identified through a public outreach process. Given
that no two cities' geographies, populations, transportation infrastructure, politics, and a host
of other issues are the same, the key issues worth tracking and responding to before, during, and
after project implementation are certainly more unique than alike. Public perception measures
should be tailored appropriately to each project application.
Facility Users
The term facility users refers to other characteristics of those who make trips into, within, and
out of the area or cordon pricing zone and the characteristics of the trips themselves. (For con-
sistency with the other two forms of pricing, the term facility is retained, but in actuality a priced
zone is not a facility per se.)
Representative Facility User Goals. Understanding who are the users of a facility--users of
the priced zone--is critical to gaining acceptance of an area or cordon pricing program and
ensuring its fair and successful deployment. One primary goal may be to identify and mitigate
negative equity change for those who may be disadvantaged by the introduction of the priced
zone--for example, lower income commuters who drive into the priced zone for work and have
few travel alternatives available (the special case of social equity is further discussed in Section
3.1.4). Goals may also be established for trip users' trip purposes such as a reduction in discre-
tionary trips to ease others given higher priority such as transit or goods movement. Character-
istics of a facility's users can be used as inputs to developing and measuring goals formulated
under other evaluation subjects. For example, users' departure times, trip times-of-day, or ori-
gins/destinations can inform decisions on setting charging policies, which can be tied to goals of
congestion reduction or revenue generation.
What Are the Facility User Measures? Measures of facility users primarily focus on char-
acteristics of the users themselves or the trips they take. Specific data on their accounts or charge
transaction type is also found among those measures used in practice. The full list derived
from current operating schemes is shown in Table 3-1. User characteristics include demo-
graphic and socioeconomic data, vehicle data, and home zip code or other residence-identify-
ing measures. Trip characteristics include, among others, frequency, departure times, trav-
elshed determinations, overall trip length, and trip purpose.
How Are Facility User Measures Applied? Measures of a facility's (priced zone's) users are
made in a combined validation and operations capacity early in the implementation and initial
evaluation period of area or cordon pricing programs. As these schemes become more common,
facility user data is likely to become less significant and may only be necessary to measure on either
an infrequent basis or when a significant change in operation has occurred. As a cordon or area
pricing program is considered and initially becomes operational, capturing the characteristics of
its users (or non-users if the priced zone is avoided) such as socioeconomic and demographic char-
acteristics is important in understanding if detrimental or inequitable effects are occurring to cer-
tain groups. Operational or policy changes may be warranted to correct such findings. For exam-
ple, physical adjustment to the priced zone's boundary or special accommodation (e.g., rebates,
discounts, or exemptions) to disadvantaged user groups may need to be introduced.
OCR for page 61
Guidelines for Evaluation and Performance Measurement of Congestion Pricing Projects 61
Performance Data in Stockholm Underpins a Successful Referendum
Stockholm became the second major urban area in Europe to implement congestion pricing with the permanent
implementation of the Stockholm Congestion Tax on August 1, 2007. The decision to implement the system per-
manently was based on the outcome of local consultative referenda held in Stockholm and several surrounding
municipalities on September 17, 2006. City residents approved the congestion tax by a margin of 51.3 percent.
Local transportation planners in Stockholm credit this positive outcome to the extensive performance monitor-
ing effort associated with a 7-month trial of the congestion tax from January 3 to July 31, 2006.
The prospect of a new and controversial tax, coupled with complicated legal and privacy issues, was cause for
sharp political debate in the Swedish capital. Local polls showed that support for the tax was lowest right
before the start of the trial period. However, support increased rapidly once the positive effects of the charg-
ing scheme became visible.
In preparation for the trial, the Swedish government established a Congestion Charge Secretariat to plan, coordi-
nate, and evaluate the outcome and communicate with the public. As part of its work the Secretariat established
performance goals for the program together with a comprehensive evaluation program to assess the extent to
which the goals would be achieved. The Secretariat's key findings from the trial included the following:
ˇ Decrease in traffic volumes of 22% at the cordon during charging hours (half from commuters who shifted from
driving to public transport, and half from consolidation, reduction, or new destinations for discretionary trips)
ˇ Reduction in peak-period delays of 33% on arterials leading into the city
ˇ Public transport ridership increase of 6%
ˇ Reduction of vehicle emissions in the inner city of 8 to 14%
ˇ Marginal effect on trade and commerce
Overall, the Secretariat concluded that the goals for the trial were met, with an even greater-than-expected
reduction in congestion, improved levels of CO2 and particulates, and an improved city environment.
The Secretariat's comprehensive monitoring program was critical to validating the success of the trial and con-
veying the benefits of congestion pricing to voters in Stockholm. The decision to hold the trial and institute rig-
orous performance monitoring turned out to be a tactical success, without which transportation officials in
Sweden do not believe it would have been possible to gain the needed approvals to make the congestion tax
permanent. On a related note, officials involved with the failed campaigns to implement congestion pricing
programs in Manchester and Edinburgh agree that the unsuccessful outcomes of referenda in those cities--
82% voted against congestion pricing in Manchester--might have been different if similar trial and monitoring
programs had been implemented prior to the vote.
Of special significance to area or cordon pricing scheme sponsors is an understanding of the
characteristics of users' trips. This information can validate whether the scheme is having the
desired effect on managing trips into the priced zone--where the trips originate and conclude,
how long they are, and for what purpose they are taken, as with user characteristics, can lead to
operational or charging policy adjustments.
Collection methods and frequencies vary for user measures. Some measures, such as basic
demographic data or vehicle classification (e.g., automobile, taxi, small truck, large truck, and
public service vehicle), can be tracked through a customer registration/management process, if
used by the scheme. The level of data available will depend on the technology used for the
scheme. Many user measures can be obtained only through survey work, such as socioeconomic
data and trip characteristics (e.g., trip length and purpose). Collection of such data is naturally
done on an infrequent, as-needed basis. Comprehensive travelshed determinations may even
require travel demand forecasting or modeling efforts.
OCR for page 62
62 Evaluation and Performance Measurement of Congestion Pricing Projects
System Operations
For these Guidelines, system operations refer to operational aspects of a priced zone that are
not directly related to measures of traffic, as discussed in the Traffic Performance section. They
are categorized in five ways:
ˇ Finance
ˇ Enforcement
ˇ Safety
ˇ Customer service
ˇ System function
Representative System Operations Goals. A wide variety of goals can be set by and evaluated
against system operations. A significant system operations goal is to collect a certain level of revenue,
most likely to recoup the initial investment in establishing the scheme and to cover operating costs,
but also potentially to improve or subsidize other travel options, such as transit. Safety is also an
important goal for all transportation infrastructure. Finally, priced zone sponsors may want to
achieve established levels of customer service or targets of system equipment availability/accuracy.
What Are the System Operations Measures? Because of system operations' broad scope, a
wide variety of measures are used to track this evaluation area as detailed in Table 3-1. Finance
measures include revenue (e.g., charges and fees) and expenditures (O&M). Enforcement mea-
sures track data that includes violation data, fines, and penalties. Measures of safety often look at
accident rates. A long and very detailed number of performance metrics can measure customer
service--from volumes of inquiry and comments received (positive or negative) to customer
service center response time and average inquiry resolution time. Application of these measures
is highly dependent on facility sponsor preference, as discussed below. Finally, measures of sys-
tem function focus on system and specific equipment availability and accuracy, numbers of
equipment incidents, and repair rates.
Research for these guidelines has shown that finance and safety are the two most prominent
types of system operations measures used for area or cordon pricing schemes. Customer service and
system function are also significant, although tracked by only one of the three schemes examined.
How Are System Operations Measures Applied?
Finance. Among the five categories of system operations
Example: Revenue Usage in Stockholm performance measures, financial performance data feature the
most prominently. In analyzing revenue collection targets and
In 2008, revenue from Stockholm's congestion tax trends, total revenue and O&M costs were collected by two of
was approximately 850 million kroner, inclusive the three schemes examined (and are certainly collected for the
of the tax, administrative and late payment fees, third, but not publicly available). The ability for priced zone
and enforcement revenues. Operational costs programs to (1) cover their operating costs and (2) repay their
amounted to about 393 million kroner, although initial capital costs is a significant consideration for project
this included several one-time charges. Estimated sponsors because of the high level of resistance that can be
operational costs in 2010 and beyond were expected when implementing these schemes. The use of signif-
approximately 250 million kroner. Net revenues icant public subsidies will only detract from their acceptance.
from the permanent charge (estimated to be However, existing experience indicates that the level of charge
600 million kroner per year starting in 2010) necessary to have the desired (significant) effect on traffic
have been reinvested in the Stockholm region's reduction should yield revenue that will cover ongoing operat-
road network, unlike during the congestion tax's ing costs and result in a surplus. (London' and Stockholm's
trial period when net revenues were invested in pricing schemes yield net revenues that exceed operating costs
improving public transportation. by a factor of two to three.) Excess revenue can be reinvested to
improve alternate modes of transportation and/or the existing
OCR for page 63
Guidelines for Evaluation and Performance Measurement of Congestion Pricing Projects 63
roadway network within and around the priced zone. Such improvements are likely necessary
both to absorb and attract users who switch modes (especially to transit) and to further bolster
public acceptance for the scheme by transparently reinvesting the money collected, rather than
having it appear to be "just another tax." For pricing schemes with a variable charge rate struc-
ture, the average charge paid, highest charge paid, and total number of transactions are of inter-
est to sponsors who look to manage the revenue collected.
Collection of toll revenue data is managed through ETC equipment and does not represent a
significant cost once a facility is operational. The data is captured on an ongoing, real-time basis
and can be considered a must-have among performance evaluation measures.
Enforcement. Enforcement of charge payment requirements is an important measure to pre-
sent to a public that expects a high level of integrity for a service that requires payment for use.
Measures of enforcement such as violation rates and volume and revenue from penalties assessed
are relevant in this case and help to (1) validate the expectation of fair application of the facil-
ity's rules and requirements and (2) inform the sponsor how effective their enforcement prac-
tices are. Enforcement will likely take the form of a camera-based system to photograph license
plates of those without a valid transponder, or, if a license plate reader system is used (as in Lon-
don and Stockholm) to identify vehicles for which the charge is assessed, a bill is generated post-
trip, with the option to charge a higher rate if not paid in advance.
Safety. Measuring safety is an important means to validate the benefits of area or cordon pric-
ing. Reductions in vehicle collisions as well as reductions in accidents involving pedestrians or
bicyclists can be tracked before and after scheme implementation. A reduction in traffic volume
inside the priced zone should affect safety conditions positively.
Customer Service. Confirmation of delivering high-quality customer service can be evaluated
by many measures--such as levels of customer inquiry (by phone or email) and quantitative cus-
tomer service measures (e.g., inquiry answer time and resolution time). Scheme sponsors will
want to consider tailoring a selection of these measures, based on the role the agency plays in
providing customer service functions, public outreach outcomes, and other needs. If a private
entity collects the charges and manages customer service, evaluation measures and reporting
requirements can be specified in the contract.
System Function. Validating the proper function of the priced zone's system equipment (and
any need for potential operational changes) can require certain performance evaluation mea-
sures. Drawing from other forms of congestion pricing along with the findings for area or cor-
don pricing, applied measures could include system equipment availability (e.g., transponder or
license plate readers, cameras, and other vehicle detection and monitoring equipment), the num-
ber of system incidents (e.g., failures and errors), and the mean time to repair the result of the
incident. Collection of these measures can be built into the software that manages the systems and
reports produced as necessary.
Environment
Environment refers to aspects of the natural environment, such as air quality and noise, which
can be affected by transportation infrastructure. The "urban" environment, as may be evaluated
based on quality of life, is not explicitly included in this evaluation area. Measuring improve-
ments in the urban environment or quality of life is imprecise and depends on specific factors of
interest to scheme sponsors, stakeholders, and the public. Measures of improvement are cap-
tured across several evaluation areas, including the (natural) environment, traffic performance,
public perception, and economics.
Representative Environmental Goals. Area and cordon pricing schemes expected to result
in significant reductions in urban traffic levels often are accompanied by similarly aggressive
OCR for page 64
64 Evaluation and Performance Measurement of Congestion Pricing Projects
environmental goals. Targeted reductions in National Ambient Air Quality Standards (NAAQS)
criteria pollutants (e.g., NOx, CO, and particulates), volatile organic compounds (VOCs), green-
house gases (GHGs), and CO2 are primary goals. Reductions in ambient noise levels are others.
What Are the Environmental Measures? Measures of the identified pollutants and noise
levels are the environmental measures included in evaluation programs for area and cordon pric-
ing programs.
How Are Environmental Measures Applied? Calculating changes in air quality requires
using traffic performance data, including traffic volumes and speeds, as inputs to air quality
forecasting tools, such as EPA's MOBILE6 Vehicle Emission Modeling Software. Air quality
monitoring stations may already exist in the locations to be analyzed and should be incorpo-
rated into the scheme's performance evaluation program. Additional equipment can be
deployed as needed. Measuring air quality may require coordination with local, state, or fed-
eral environmental agencies. Reduction in noise requires deployment of targeted sound level
measurement equipment in areas of concern; collected data can be compared with data before
scheme implementation.
Transit
Transit refers to aspects of transit service that operate within the same region as the area or
cordon pricing zone, especially services that provide access to the zone itself. Both bus and rail
service are considered as alternate modes of travel to access the priced zone.
Representative Transit Goals. Goals related to transit service are of primary concern to
priced zone scheme sponsors because having alternate modes available is essential for successful
implementation. Increased ridership, a primary goal, is indicative of a successful mode shift from
personal vehicles entering the priced zone. Related goals focus on improving specific aspects
of service--frequency, timeliness, areas served, quality, and subjective indicators of customer
satisfaction.
What Are the Transit Measures? Aspects of transit ser-
vice include performance, ridership, finance (revenue), and
Example: Promoting Consideration for Transit quality of service (as measured attitudinally through customer
surveys). Research indicates that transit performance was
Singapore's Land Transport Authority (LTA) is
measured in two of the three schemes by examining travel
responsible for the country's roads and public
times, on-time rates, or excess wait times (delay); average
transportation systems, including heavy and
speeds; and ridership or boarding counts. Average vehicle
light rail, buses, and taxis. One main goal of its
occupancy; farebox revenue; and quality, satisfaction, and
Electronic Road Pricing (ERP) program is to
reliability as perceived by customers were also employed by at
encourage commuters to choose the most
least one pricing program. That more comprehensive metrics
appropriate transportation mode. ERP opti-
are used for priced zones than for variably priced managed
mizes the use of the city-state's constrained
lane or toll facilities indicates the greater role transit plays in
road capacity and strongly incentivizes public
successfully operating these schemes and the importance of
transportation, which has benefited from signif-
documenting the results.
icant investments in parallel with 35 years of
cordon pricing. LTA has set a target of making How Are Transit Measures Applied? If the sponsor of the
70 percent of all morning peak-hour trips on area or cordon pricing scheme also operates the region's tran-
public transport by 2020. Transit travel times sit service, acquiring transit performance data is not difficult.
from location benchmarks throughout Singa- Otherwise such data needs to be acquired (if such data exists)
pore have been established and are used to from individual transit agencies. Obtaining the performance data
monitor the reliability of service. sought, however, requires establishing a good working relation-
ship with that agency and coordinating data collection efforts.
OCR for page 65
Guidelines for Evaluation and Performance Measurement of Congestion Pricing Projects 65
Economics
Economics refers to a broad range of economic indicators and trends within the region affected
by an area or cordon pricing program. Equally, economics may include macro-level quantifi-
cations of economic health as well as individual examinations of effects to particular economic
sectors (e.g., specific businesses or urban activities).
Representative Economics Goals. Unlike with variably priced managed lane or toll facili-
ties, economics is likely to be an important consideration for area or cordon pricing projects
because the expected significant reduction of vehicular traffic within the zone could have a mea-
sured impact, perceived or otherwise, on economic activity--both at a macro scale and individ-
ually on certain sectors. Goals may include (1) having no net loss in economic activity or no
adverse effect on particular services and (2) an increase in economic activity because of improved
access, mobility, or the desirability of operating a business or conducting commerce in the zone
because of the improved urban environment.
What Are the Economics Measures? Economics measures are shown in Table 3-12 and
Table 3-13. Economic impacts at the macro level are measured by gross regional product or other
economic indices that quantify activity in the priced zone's region at an aggregate scale.
Specific economic impacts focus on businesses and property. General business performance
of commercial establishments, most easily captured through openings and closings, was found
in two of the three schemes examined. Measures of retail patronage and sales were also quanti-
fied. Other measures employed in the case of one scheme include a qualitative service-by-service
analysis of specific business sectors or worker populations, measures of business costs and prices,
and impacts to tourists. Property impact measures include residential and commercial values, as
well as sales and rental volumes.
Finally, unlike the findings from the other two forms of congestion pricing, research indicates
that area and cordon pricing have lent themselves to performing benefit-cost analyses. This
result may be indicative of the greater economic impact these schemes can have, as well as need
to further justify instituting such a marked change to managing traffic within a region. It may
also highlight the traditional lack of applying benefit-cost analyses to transportation improve-
ment projects in the United States.
How Are Economics Measures Applied? Applying and analyzing economic impact mea-
sures will require quantification of baseline economic activity level before scheme implementa-
tion, as well as control factors for other external impacts to the economy to attribute economic
impacts to the introduction of a priced zone definitively. Obtaining measures of economic activ-
ity may rely on data collected by a city's economic development organizations, departments of
revenue, and others. Specially designed surveys will be needed to target particular economic sec-
tors, businesses, or populations in order to focus on the effects of the pricing program. Economic
modeling can also be performed as a substitute or complement to selected quantitative findings.
Land Use
Performance measures to evaluate a priced zone's impacts on land have been used by one of
the three schemes examined. Tracking patterns of residential and commercial development may
be of interest to program sponsors. However, the results of such tracking would be a long-term
outcome, because land use patterns would require significant periods of adjustment before mea-
surable results could be achieved.
