Congestion pricing response: section I: literature review [Oct. 2008]

Congestion Pricing Response
Section I: Literature Review

Produced for Prepared by

Georgia Department of Transportation Center for Quality Growth and Regional Development & School of Civil and Environmental Engineering at the Georgia Institute of Technology

Catherine L. Ross, Ph.D., Director and Harry West Chair, CQGRD Principal Investigator
Randall Guensler, Ph.D.., Assistant Professor, CEE Co-Principal Investigator
Jason Barringer and Amy Danner, Research Scientists, CQGRD Molly Allen, Elise Barrella, Jessica Doyle, and Lyubov Zuyeva,
Graduate Research Assistants

October 2008

Congestion Pricing Response

Section I: Literature Review

Section Contents

List of Figures

iv

List of Tables

vi

Executive Summary

1

Glossary

4

Introduction

5

Making Congestion Pricing Feasible:

Advances in Technologies

7

Electronic Toll Collection (ETC)

7

ETC Technology Inoperability

10

Other Technologies

11

Consumer Responses to Technologies

14

Technology Evolution: The Case of Singapore 15

Conclusions

17

Public Acceptance of Congestion Pricing 19

Public Objections to Congestion Pricing

20

Equity Concerns

21

Political Difficulties in Implementing

Congestion-Pricing Programs

22

Public Education and Outreach

23

Conclusions

25

About Georgia Department of Transportation
The Georgia Department of Transportation (GDOT) plans, constructs, maintains and improves the state's roads and bridges; provides planning and financial support for other modes of transportation; provides airport and air safety planning; and provides air travel to state departments. For more information, visit www.dot.state.ga.us.
About School of Civil and Environmental Engineering
The School of Civil and Environmental Engineering (CEE) was established 1896 and encompasses three degreeawarding programs, including civil engineering, environmental engineering, and engineering science and mechanics. For more information, visit www.ce.gatech.edu.
About Center for Quality Growth and Regional Development
The Center for Quality Growth and Regional Development (CQGRD) is an applied research center of the Georgia Institute of Technology. The Center serves communities-- particularly those in the Southeast United States--by producing, disseminating, and helping to implement new ideas and technologies that improve the theory and practice of quality growth. For more information about CQGRD visit www.cqgrd.gatech.edu.
2008 by Center for Quality Growth and Regional Development (CQGRD). All rights reserved. Short sections of text, not to exceed two paragraphs, may be quoted without explicit permission provided that full credit, including the notice, is given to source. Available electronically at _________________________________ ____________________________.
Cover Credit: Amy Danner, CQGRD

ii

Case Studies

26

State Route 91 Express Lanes, California

26

Interstate 15 Express Lanes, San Diego

30

Houston QuickRide, I-10 and US-290,

Houston

32

MnPASS, Interstate 394, Minnesota

36

Interstate 25/US-36, Colorado

41

Conclusions

45

Summary Table

47

Other Examples of Congestion Pricing 48

Introduction

48

Within the United States

48

Outside the United States

52

Conclusions

59

Conclusions

60

References

61

iii

Congestion Pricing Response
Section I: Literature Review
List of Figures

Figure 1: FasTrak Transponder

8

Figure 2: Schematic of Automatic Vehicle

Identification Process

9

Figure 3: Complete DTECT System

12

Figure 4: Enforcement Gantry Light

13

Figure 5: MER Unit

14

Figure 6: SR-91 Express Lanes Map

26

Figure 7: SR-91 Express Lanes Toll Schedule,

July 2007

28

Figure 8: Average Daily Traffic and Revenue Growth,

SR-91, 19962006

29

Figure 9: I-15 Express Lanes Map

30

Figure 10: Current Configuration of Managed Lanes

on I-15

31

Figure 11: Map of Katy Freeway and Northwest

Freeway

33

Figure 12: Katy Freeway (I-10 West) HOT Lane

33

Figure 13: Motorist on Katy Freeway

35

Figure 14: I-394 HOT Facility

36

Figure 15: Line-Separated HOT Lane Entry

and Egress

38

Figure 16: Survey Results of Minneapolis Carpoolers 40

Figure 17: Map of I-25 HOV/Tolled Express Lanes 41

Figure 18: Configuration of I-25 HOT Lanes

43

iv

Figure 19: I-25 HOT Monthly Traffic Voumes,

June 2006March 2007

44

Figure 20: I-25 Projected and Actual Revenue,

First 10 Months of Operation

45

Figure 21: Map of HOT Lanes on I-15, Utah

48

Figure 22: Proposed Cordon Pricing for Manhattan 50

Figure 23: E-ZPass Operations

51

Figure 24: Map of Congestion Charging Zone,

London

53

Figure 25: Congestion Charging Cameras,

London

54

Figure 26: Toll Ring in Trondheim, Norway

56

Figure 27: 407 ETR Open Road Tolling System

58

v

Congestion Pricing Response
Section I: Literature Review
List of Tables

Table 1: Hours of Operation, Houston QuickRide

34

Table 2: Pre- and Post-MnPASS HOV Lane

Violation Rates

39

Table 3: Results from MnPASS User Surveys

40

Table 4: Citations, I-25 Express Lanes,

June 2006March 2007

44

Table 5: Net Proceeds from London Congestion

Charge, April 2002March 2006

55

Table 6: Projected Costs and Revenues, London

Congestion Charging Zone, 20002008

56

vi

Executive Summary
Traffic congestion is an increasing burden for American cities. Clogged, slow-moving freeways and interstate highways can delay truck transport and commuters, causing losses to local businesses and making the area as a whole less attractive to potential economic investors. Drivers suffer increased stress and the resultant negative health effects. Long delays in car travel leads to greater amounts of carbon dioxide and other pollutants being spilled into the atmosphere, increasing air pollution. In short, city and state transportation agencies have a strong interest in reducing congestion.
One strategy that has been introduced in the last several decades is "congestion pricing" or "value pricing." In short, congestion pricing is the practice of charging drivers to use a specific lane or enter a designated area, such as central London. The idea is to reduce demand, and thus the number of cars competing for space on the road, by making more explicit the costs of adding an additional driver to the lane or area. A refinement of this strategy is to vary the price to reflect demand--in many cases, the price is higher during peak travel periods and lower at other times of the day. The result is freer-flowing travel for those drivers willing to pay the toll.
This literature review explores advances in congestion pricing technologies, public acceptance of congestion pricing, case studies of five facilities, and a review of other strategies. It is intended to support the "Congestion Pricing Response" project currently being conducted by the Georgia Institute of Technology on behalf of the Georgia Department of Transportation (GDOT). The expectation is that this project, when completed, will help guide GDOT in the siting, evaluation, and implementation of future pricing strategies.
Congestion Pricing Technologies
New technologies now make it considerably easier to establish and enforce a congestion-pricing mechanism, whether the facility is limited to one lane or established as a cordon. The most widely-used new technologies include electronic toll collection via transponders located in individual cars. Technologies, such as license-plate reading, automatic vehicle occupancy detection, and enforcement gantry lights, are also being developed to assist in enforcement. Such technologies allow for the separation of a congestion-priced lane's users into those who do not have to pay the toll (e.g. a vehicle with two or three people inside) and those who do.
Audiences for electronic toll collection and transponder use have often responded favorably and adapted to using the new technology. However, there may be privacy concerns with some of the enforcement technology; users may react negatively to having their license plate read or their picture taken by a digital camera trying to determine occupancy.

Page 1 of 66

Congestion Pricing Response: Section I

Public Acceptance
Congestion pricing is still relatively new in the United States, and in some past cases has proven politically unpopular. While there is some evidence that American audiences are becoming slightly more comfortable with the idea of congestion pricing, the idea has not yet been matter-of-factly accepted. Familiarity seems to lead to more positive responses: people who have used a congestion-priced facility or a toll lane before seem to react more positively to the idea of a congestion-pricied facility than do those who have not. But potential users may react negatively if they believe that a "free" facility is being taken away from them. Turning an HOV (high-occupancy vehicle) lane into an HOT (high-occupancy toll) lane may be more acceptable to the majority of users, but it may provoke opposition from existing HOV users. Users may also cite equity as a concern, fearing that a congestion-free drive will be a privilege limited only to those who can afford it. All of these things can lead to difficulty in generating political support for projects.

There are certain things a public agency can do to mitigate these concerns. First, especially if this is the first proposed congestion-pricing facility in a region, prepare a detailed and comprehensive outreach program. Keep potential users informed as to the features and predicted consequences of the facility, and be willing to adjust the proposal based on their concerns. Second, anticipate heightened scrutiny in some areas, such as the potential equity issues and the planned destination of the resulting revenue. Finally, be able to explain, clearly and memorably, that a congestion-pricing facility will bring benefits to the surrounding area; be able to say what those benefits might be, and why they are worth incurring the costs of a new project.

Case Studies
To explore congestion-pricing experiences in the United States, five facilities that have applied different implementation approaches and experienced varying degrees of public acceptance have been selected for review. The oldest, State Route 91 (SR-91) in Orange County, California, which was opened in 1995, has relieved congestion in a high-trafficked area but came under public opposition as a result of the way the public-private financing and ownership of the facility was handled. Both the Houston QuickRide facility and the facility on Interstate 25 outside Denver, Colorado, were designed to take advantage of excess capacity on existing HOV lanes. The Express Lanes on Interstate 15 outside San Diego were originally conceived to raise money for transit, rather than to relieve congestion. The MnPASS facility to the west of Minneapolis/St. Paul has relieved congestion for suburban commuters into the city.

It is clear that there is no set formula for a congestion-pricing facility. However, we can note some similarities between the five cases. All five have barrier-separated sections; four have reversible sections. Four of the five are able to balance HOVs and SOVs in the same lane; while enforcement has been a

Page 2 of 66

Congestion Pricing Response: Section I

difficulty, it should be recognized that a congestion-priced lane can accommodate both carpoolers and single drivers. We can tentatively conclude that (with the exception of SR-91, which had the additional variable of a prominent public-private partnership coming under fire) all show a trend of consumer acceptance of the congestion-priced facility rising after it opened. This is true whether or not SOVs (single-occupancy vehicles) have been allowed to use the facility. Thus, it may be that the most difficult obstacles for a congestion-pricing project are faced before implementation begins.
Other Strategies
Other congestion-pricing projects are currently being considered in cities both inside and outside the United States. London, Singapore, and Trondheim, Norway have cordon-pricing schemes to price access to the central city; New York City is looking to implement a similar scheme for downtown Manhattan. The 38-mile-long, non-barrier separated HOT facility in Salt Lake City is the most recent and the longest addition to the list of U.S. HOT lanes. Both the northeast United States and Toronto have portions of toll lanes with variable pricing and transponders in use.
In contrast to the five case studies, the congestion-pricing projects in this section show a greater variety of congestion-pricing strategies, including the use of cordons, variable tolls, and monthly vehicle tags, and take different approaches to technology and enforcement. Together, they demonstrate that while congestion pricing remains a politically sensitive issue, there are now more options than ever for putting together a congestion-pricing project.
Conclusions
The biggest challenges to a congestion-pricing project, in terms of public acceptance, are familiarizing the public with the tolling mechanism and explaining the potential benefits, as opposed to the more evident costs. However, in most cases, public acceptance has increased once the congestion-pricing project has been implemented and shown to function smoothly.
A congestion-pricing project in metropolitan Atlanta would face similar obstacles. There may also be some public resentment based on distrust of state government and the perception that residents are being asked to pay for a previously "free" service. Furthermore, because of Atlanta's history of racial inequalities, equity questions might be even more politically volatile than they would be in another city. Any congestion-pricing project would need to be carried out with awareness of these issues and a wellstructured public information campaign with plenty of opportunities for public input and interaction with decision-makers.

Page 3 of 66

Congestion Pricing Response: Section I

Glossary
Automatic Vehicle Classification (AVC): when a transponder is used to classify the vehicle from which it is transmitting, usually by weight or number of axles.
Automatic Vehicle Identification (AVI): when electronic tags installed in the vehicle can communicate with roadside readers to identify vehicle ownership.
Congestion pricing: the practice of charging drivers to use a specific lane or enter a designated area. Also called "value pricing."
Electronic toll collection (ETC): a system that uses vehicle-to-roadside communication technologies to perform an electronic monetary transaction between a vehicle, passing through a toll station, and the toll agency.
Gantry: an structure (such as a sign) mounted on an overhead support under which cars traveling the highway must pass.
High-occupancy toll (HOT): a toll lane in which single-occupancy vehicles must pay but vehicles with two or more persons inside can travel for free.
High-occupancy vehicle (HOV): a passenger vehicle with two or more persons inside. License-plate recognition (LPR): when cars are identified by their license plates, read via video
recording. Managed lane: a toll lane in which the price of the toll varies with demand for the lane: the toll is higher
during peak travel periods. Mobile enforcement transponder reader (MER): a unit which allows a police officer to read the
transponders of passing vehicles, or to travel adjacent to a vehicle in the HOT lane and read the transponder. Radio frequency identification (RFID): the technology most frequently used in the United States for the communication between an in-vehicle transponder and a receiver. The frequency is usually 900 MHz. Single-occupancy vehicle (SOV): a passenger vehicle with only one person (the driver) inside. Super-low-emissions veichle (SLEV or SULEV): a vehicle which has been found to emit considerably fewer emissions than standard passenger vehicles (in California, the standard is that SLEVs are 90% cleaner than standard vehicles). SLEVs are allowed to use HOV lanes in many states even if occupied by only one person.

Page 4 of 66

Congestion Pricing Response: Section I

Introduction

State and federal departments of transportation are faced with the tasks of not simply increasing access to transportation, but battling congestion, mitigating the negative environmental impacts of personal transport, and maintaining existing infrastructure. Moreover, as America's population grows, infrastructure capacity has become a pressing issue for many cities and regions. Clogged, slow-moving freeways and interstate highways can delay truck transport and commuters, causing losses to local businesses and making the area as a whole less attractive to potential economic investors. Drivers suffer increased stress and the resultant negative health effects. Long delays in car travel leads to greater amounts of carbon dioxide and other pollutants being spilled into the atmosphere, increasing air pollution.

These increasing concerns about congestion and lack of infrastructure capacity have led to greater consideration of congestion pricing than in the past. In this case, congestion pricing refers to the act of making one or more lanes on a given stretch of public road a tolled facility. It is now possible to vary the toll as to increase or reduce demand so that speed in the toll lane (also known as a "managed lane"; see Glossary) can remain at a constant speed relative to the flow of traffic.

Introducing tolled lanes may allow for an increase in the capacity of the remaining lanes and thus a reduction in the overall congestion. Congestion pricing may also reduce congestion by reducing the number of cars on the road. A survey of Dutch car owners suggested that between 6% and 15% of car trips would be adjusted by drivers in response to tolls; of those adjustments, between 91.5% and 98.5% would be to alternatives to cars, such as public transport, non-motorized travel, or not making the trip at all (Ubbels and Verhoef, 2006). A similar study commissioned by the United Kingdom Department for Transport suggested that a charging scheme with 10 different price levels could result in a reduction of congestion of British roads by 46% (Bonsall et al., 2006).

In the United States, the opportunity for congestion pricing has in part come about from the perceived underutilization of existing high-occupancy vehicle (HOV) lanes on interstate highways. These lanes, whose creation was largely funded by federal grants, are meant to be used only by vehicles with one or more passengers in addition to the driver. HOV lanes are often marked separately from other lanes (see Figure 1), and solo drivers can be fined if caught driving in the HOV lanes.

HOV lanes, however, have been perceived as relatively ineffective tools of congestion management. They do not offer sufficient incentive for enough drivers to change their behavior to noticeably reduce congestion, and are often perceived as being underutilized. Moreover, since HOV lanes are only infrequently barrier-separated, enforcement can be difficult, further reducing drivers' incentives to carpool. Toll lanes can be more effective at reducing congestion than HOV lanes because they can be used by

Page 5 of 66

Congestion Pricing Response: Section I

more drivers and because, in the cases of barrier-separated toll lanes, they can be more easily enforced, and can promise better shielding from congestion caused by accidents, than can HOV lanes.
Because congestion pricing is still relatively new in the United States, this literature review will examine both recent technological advances in congestion pricing and what is known about public acceptability of congestion pricing. In addition, it will describe five of the best-known examples of successfully implemented congestion-pricing programs in the United States: the "QuickRide" program on Interstate 10 and US-290 outside Houston; the MnPASS program on Interstate-394 west of Minneapolis-St. Paul; the Express Lanes on Interstate 15 outside San Diego; the high-occupancy toll lanes on State Route 91 in Orange County, California; and the recently implemented toll facility on Interstate 25 outside Denver. This review also includes a brief summary of less typical examples of congestion pricing in the United States and abroad.
This literature review will thus provide an overview of the most recent advances in congestion pricing, and how public audiences have reacted to these advances. It is designed to help transportation planners who might be considering a congestion-pricing program to avoid past mistakes and learn from best practices.

Page 6 of 66

Congestion Pricing Response: Section I

Making Congestion Pricing Feasible: Advances in Technologies

While tolled roads have existed for centuries, variable pricing has only become feasible in the last two decades due to advances in traffic-monitoring technology. Electronic payments further facilitate variable pricing, as users can have money deducted automatically when their car crosses the tolled area. Toll facilities have also begun relying on cameras for vehicle identification, as is the case in London's congestion zone and on Toronto's E-470 tolled highway. The following section discusses the most recent advances in technology that might be useful when designing a congestion-pricing facility.

Electronic Toll Collection
Electronic Toll Collection (ETC) is a technology that allows for electronic payment of highway tolls and is essential to the high-occupancy toll (HOT) concept. ETC systems use vehicle-to-roadside communication technologies to perform an electronic monetary transaction between a vehicle passing through a toll station and the toll agency. ETC allows toll-collection transactions to be performed while vehicles travel at near highway cruising speed. Electronic Toll Collection is becoming a globally accepted method of toll collection, with the help of improvements in ETC technologies. ETC systems have the potential to:
reduce queues at toll plazas by increasing toll booth service rates; save fuel and reduce mobile emissions by reducing or eliminating deceleration, waiting times,
and acceleration; and reduce toll collection costs.

An ETC system typically includes the following components: 1. Automatic Vehicle Identification (AVI), using radio frequency identification (RFID) 2. Automatic Vehicle Classification (AVC) 3. Video Enforcement Systems (VES), using License Plate Recognition (LPR) and Barcode License Plate Recognition 4. Lane discrimination technology

Automatic Vehicle Identification / Radio Frequency Identification
Automatic Vehicle Identification (AVI) technology enables various ETC applications through its ability to accurately identify a specific vehicle at highway speeds. AVI entails the use of electronic tags installed in the vehicle, which communicate with roadside readers to identify vehicle ownership (Smith and Benko, 2007). As a vehicle passes under a toll-collection gantry (overhead sign assembly), its electronic identification, encoded a into transponder installed in the vehicle (see Figure 1), is read by a gantry-

Page 7 of 66

Congestion Pricing Response: Section I

mounted or roadside device. The vehicle identification is then used to deduct the applicable toll from the customer's preexisting account, or the customer is sent an invoice.
Figure 1: FasTrak Transponder
Source: SANDAG (2003) In the U.S., the transmission of an identification code between the transponder and a roadside reader is usually handled by a radio frequency identification (RFID) unit operating in the 900 MHz radio frequency band. The driver does not have to stop to pay the tolls, and no tollbooths are required. ETC also determines whether the cars passing are actually enrolled in the program, and can store the information on the vehicle in violation for further collection or enforcement action (IBI Group and Cambridge Systematics, 2006).

Page 8 of 66

Congestion Pricing Response: Section I

Figure 2: Schematic of Automatic Vehicle Identification Process

Source: IBI Group and Cambridge Systematics (2006)

Automatic Vehicle Classification
Traditionally, tolls have been differentiated by vehicle class. Automatic vehicle classification technologies installed in the roadway can determine a vehicle's class by its physical attributes, such as weight, length and number of axles. For the purpose of high-occupancy toll (HOT) lanes, vehicle classification is usually less important. Heavy commercial vehicles other than buses are generally excluded from the HOT lanes. Transponders used for AVI also allow for a quick and easy vehicle classification method. On Electronic Toll Route 407 (Canada), heavy vehicles which have weight over five tons are required to carry a transponder, which is set up to automatically charge a heavy vehicle rate (407 Express Toll Route, 2007). On SR-91 Express Lanes in California, high occupancy vehicles are required to carry a transponder, and use the facility at no charge or at a discounted charge, depending on the time of travel (Orange County Transportation Authority (OCTA)).

Vehicle Enforcement Systems, License Plate Recognition (LPR), and Barcode License Plate Recognition
Toll violators could be monitored by video enforcement systems, which use video imaging and license plate recognition to photograph their license plates in order to identify and fine the vehicle owner. The license plate recognition (LPR) technology allows the deciphering of license plate numbers.

Page 9 of 66

Congestion Pricing Response: Section I

Approximately 30 private companies currently offer LPR systems. When a toll violation is detected, the LPR system is activated to record the offending vehicle's license plate and store it for further transmission to a management center via standard telephone lines, cellular communication, radio transmission or Ethernet. Using automated violation processing, the license plate numbers can be used to automatically access the DMV records and find the owner's name and address. The driver can then be sent a citation by mail, much like a red light running ticket for violators caught on camera. Increasingly, this technology is becoming more reliable, with recognition rates up to 99.5%, even during severe weather conditions (Cothron, Skowronek et al. 2003).

LPR is the primary mode of tolling on Route 407 in Toronto. About 30% of the users have transponders installed. As the driver passes under a gantry, the system detects whether a valid transponder is present. If not, two separate cameras take a picture of the license plate, to ensure accuracy. The digital pictures are reviewed by a computer, and by a human eye in a small percentage of cases. The license plate numbers are linked with the DMV records, and the drivers are sent a "V-Toll" bill in the mail. There is a surcharge for being billed by V-Toll, rather than through using a transponder. In 2005, over 100 million trips were processed with an accuracy rate of 99.9%. (407 Express Toll Route)

Instead of LPR, it is possible to use barcodes printed on the back of vehicles. Barcode license plate recognition has been made possible by high-speed bar code readers, in combination with cameras with pulsed infrared illumination and a very fast shutter speed to capture crisp images at highway speeds. Developed computer software can analyze the video stream at 60 frames per second, and detect if a bar code image is present. If a bar code image is detected, that image can be separated from the video stream and used to automatically identify the vehicle.(Cothron, Skowronek et al. 2003) Currently, LPR is a more widely tested and used technology than barcode recognition. Barcode recognition systems might be most applicable in freight tracking and enforcement.

Lane discrimination technology
Lane discrimination technology is used to ensure that overhead transponder readers only collect signals from the AVI transponders in the appropriate lane. This would be important when a transponder-carrying vehicle is traveling in one of the regular lanes adjacent to a high-occupancy toll lane.

ETC Technology Interoperability
As more and more localities implement electronic tolling, the question of accommodating travelers from another area arises. With V-Tolls, as implemented on Route 407 in Toronto, out-of-state drivers are sent a V-Toll in the mail, thanks to the agreements between Ontario and other Canadian provinces. When the tolling system relies primarily on transponders, the interoperability of technology is an issue. In California, State legislature required that one single technology be used on all toll projects. Now, 79 miles of toll

Page 10 of 66

Congestion Pricing Response: Section I

roads and bridges in California are all using FasTrak electronic tolling technology. Because all transponders within the state are interoperable, a customer can use the same transponder on toll bridges in Northern California and on toll roads in San Diego. No national electronic tolling standard has been established at this point. FasTrak transponders cannot be used on East Coast tollways. Instead, EZPassTM technology is used in the North Eastern United States. As many older toll authorities consider open-road electronic tolling, creating a national tolling standard could become very important (Leahy 2005).

Other Technologies
In enforcing an HOV or an HOT facility, one of the primary difficulties is that only some of the users of the facility are required to pay a toll (e.g. carpoolers go free, or pay a discounted toll). One of the essential enforcement tasks, therefore, is differentiating between the vehicles that satisfy the occupancy requirements and are not required to pay a toll, and those that have to pay a toll. Various strategies exist in dealing with the differentiation process: separate lanes for HOVs and SOVs as they pass through the tolling station on I-25 in Denver; requiring both SOVs and HOVs to carry transponders, used on SR-91 in Orange County, California; enforcing occupancy requirement manually by patrolling officers, aided by public self-monitoring programs such as HERO, in Houston. This section will cover possible automated occupancy-detection technology and technologies available to assist officer enforcement.

Automated Vehicle Occupancy Detection
Obtaining accurate occupant counts is one of the main hurdles in automating HOT and HOV lane enforcement. Automated vehicle occupancy detection technology remains in the research and testing stage, and no full-scale systems have been implemented as of yet. However, it is important to keep track of technology development, as an automated vehicle occupancy detection system would be ready for implementation in the near future. The two main types of automated vehicle occupancy detection are remote and in-vehicle detection.

Remote occupancy detection uses systems outside of the vehicle and attempts to determine the number of people in the vehicle through manual or automated analysis of video and infrared images. A typical automated occupancy detection strategy would involve installing at least three cameras with artificial lighting sources: to capture the front windshield view, the side window view, and the rear license plate. A semi-automatic review process can detect when the number of occupants is less than required, electronically saving the images of the vehicle's interior and license plate for later manual confirmation and citation processing.

A semi-automated HOV occupancy enforcement system has been tested on the I-30 contra-flow HOV lane in Dallas, Texas, under the name of HOVER. Transformation Systems, Inc. (Transfo) of Houston,

Page 11 of 66

Congestion Pricing Response: Section I

Texas, collaborated with Computer Recognition Systems, Inc. (CRS), Texas Transportation Institute and other agencies in installing and configuring the HOVER system (Turner 1999). While the system was found to be effective for mailing educational information to suspected violators, the results also showed that actual enforcement would require better quality video cameras, reduced video signal transmission loss, additional camera views and better license plate recognition. There are additional difficulties in capturing images of children in a vehicle, or in dealing with tinted windows (Cothron, Skowronek et al. 2003).
The most successful automated occupancy system to date is the DTECT system, developed in the United Kingdom by Vehicle Occupancy Limited (Vehicle Occupancy Limited, 2007). DTECT illuminates the windshield area with two different wavelengths of infrared light and takes 2 digital infrared pictures of the windscreen at the instant of illumination. The output is a vehicle occupancy count that can be transmitted by an Ethernet link to a remote location within seconds. The complete DTECT system (see Figure 4) is contained within a single housing and mounted on a roadside support or on an overhead gantry. Tests on the A467 HOV lane in Leeds, England claim a 95% success rate in detecting real people and rejecting dummies (Poole 2006). Production is provisionally scheduled for the 4th quarter of 2007 (Vehicle Occupancy Limited, 2007).
Figure 3: Complete DTECT system is contained within a single weather and vandal-proof housing

Source: Vehicle Occupancy Limited (2007)
In-Vehicle Occupancy Monitoring
Due to the problems with automated detection methods because of inherent visibility limitations in the darkness, as well as the difficulty in seeing all of the seats in the vehicles, detecting the occupancy from

Page 12 of 66

Congestion Pricing Response: Section I

within the vehicle and then transferring that information to the monitoring agency could be considered as another potential solution. As of 2006, all new vehicles must have smart air bags, which include occupancy-detection systems in the front seats. This pre-existing technology could be readily adapted to occupancy enforcement if smart airbag sensor data could be transmitted to the gantry systems. The need for police to enforce occupancy in the field could be significantly reduced, once the managed lane users are restricted to vehicles equipped with an OMS. (Schijns, 2005)
There are concerns, however, about on-board OMS use for managed lanes occupancy. These concerns include privacy and civil liberty perceptions, cost of on-board unit to transmit the information, accuracy and reliability (very important if the ticketing of violators would be done by mail), and lack of political and public acceptance (Schijns, 2005)
According to Poole (2005), an in-vehicle approach for determining vehicle occupancy would not be a good enforcement strategy due to high costs and the fact that OMS applies only to the two front seats of the vehicle and would not help on HOV3+ facilities. Instead, he advocates outfitting eligible work carpools and vanpools with a transponder at no charge.
Technologies that Assist Officer Enforcement
Enforcement Gantry Light (I-15 San Diego, MnPASS) Enforcement gantry lights, or beacons, are installed on toll gantries. The light flashes when a vehicle with a valid transponder enters (see Figure 4). This simple technology allows a police officer following a vehicle to make a quick visual check as to whether the vehicle is in good standing.
Figure 4: Enforcement Gantry Light

Page 13 of 66

Source: Halvorson and Buckeye (2006) Congestion Pricing Response: Section I

Handheld Transponder Reader (MnPASS) The transponder readers allow enforcement officers to follow an SOV through a toll zone and determine if the SOV has a MnPASS account. When the SOV passes through the toll zone, the system will cause the enforcement transponder in the trailing officer's vehicle to issue an audible beep. If the officer does not hear the beep, the vehicle is in violation.
Mobile Enforcement Transponder Reader (MER) (MnPASS). Once installed in an enforcement vehicle, an MER unit allows an officer to read the transponders of passing vehicles, or to travel adjacent to a vehicle in the HOT lane and read the transponder (see Figure 5). The mobile unit provides the officer with the last date and time the transponder was read and the account status (e.g. valid, not valid). This technology can ensure users are not disengaging their vehicle's transponders as they pass under tolling gantries (Halvorson and Buckeye, 2006).
Figure 5: MER Unit

Source: Halvorson and Buckeye (2006)

Consumer Responses to Technologies
While the general public might have mixed views regarding electronic tolling technology, users of existing Express and HOT lane facilities are voting in favor by signing up for accounts and transponders in high numbers. For example, on State Route 91, the oldest Express lane facility in the U.S., additional 5,000 customers signed up for accounts in 2006, and 56% of all customers in 2006 have been users of the facility for seven years or more (Orange County Transportation Authority, 2006).

Page 14 of 66

Congestion Pricing Response: Section I

According to a NuStats report, the MnPASS program users in Minnesota reported a high satisfaction rating of the MnPASS subscription process and usage:
83% of MnPASS users were satisfied with the ease of opening a pre-paid MnPASS account. 83% of MnPASS users were satisfied with the ease of installing the MnPASS transponder. 87% of MnPASS users were satisfied with using a credit card to replenish their account. 93% of MnPASS users were satisfied with electronic operations of the system (NuStats
2006).

I-15 commuters (both full-time and part-time) felt that FasTrak's technology was working well and that it was relatively easy to purchase a transponder for the car (Godbe Research and Analysis, 1998). However, there was some confusion as to where transponders could be purchased. Respondents were asked questions regarding the transfer of information regarding the level of traffic on I-15, especially regarding the use of a website, a dedicated radio station, and variable message signs. Respondents suggested the creation of a website that would provide up-to-the-minute toll and traffic information, but then responded that they would not use the website due to the lapse in time between checking traffic information and using I-15. The creation of a radio station was a more favorable option to providing information regarding traffic and tolls on I-15. Full-time users were interested in knowing traffic information 1-2 miles before the entrance to the Express lanes, as well as the average speeds of traffic in general lanes compared to Express lanes. Part-time users were also interested in knowing the speed of traffic in the general lanes, as well as information about traffic problems and road closures on roads other than I15. Variable message signs were also suggested to communicate traffic information and were favored the most out of the three options. Full-time users preferred that the signs display the average speed of traffic in main lanes, while part-time users preferred that the signs display the toll and the average speed of traffic in the general lanes compared to Express lanes. However, all users wanted the signs in symbols rather than words as to not distract drivers and wanted the signs placed above the lanes, rather than on the side of the road (Godbe Research and Analysis, 1998).
Technology Evolution: The Case of Singapore*
Singapore's cordon-pricing project began as the Area Licensing Scheme (ALS) in 1975. As Singapore's pricing scheme is one of the oldest in the world, the technologies used to enforce the cordon have changed over time. In 1998 enforcement was switched from the ALS to electronic tolling.

Under the manual ALS, 150 persons were employed as operators of license sale booths and police officers in control booths at entry points. As the number of different types of licenses increased, police
* This section is drawn from Menon (2006).

Page 15 of 66

Congestion Pricing Response: Section I

had to pay special attention over longer periods of time. In 1992 when Singapore started looking for an electronic road pricing system, there were no comparable systems operating anywhere. Three international contractors took part in a competition to select the best system. In 1998, the manual system was replaced by electronic pricing. ERP gantries were installed at each entry point. The ERP system uses a short-range radio communication in the 2.40 GHz band. Vehicles are outfitted with In-vehicle Units (IU), with a slot for receiving a prepaid stored value smart card called the CashCard. The CashCard is a prepaid contact integrated circuit chip plastic card, and one can add money to the card at ATMs and gas stations. The CashCard can be used for car parking charges and regular purchases.
At each entry point, there are two sets of overhead gantries, 15 meters apart. They have the radio antennae that communicate with the IU of passing vehicles, optical sensors to detect the vehicles, and enforcement cameras to take picture of the rear license plates of vehicles found in violation. A local controller at each outstation houses the monitoring equipment for ERP gantries, and communicates with the central computer system via telephone lines. The main control center receives all the records of ERP transactions and digital photos of violating vehicles. The control center settles all the monetary transactions for the day, sends out violation fine notices, and notifies drivers of vehicles with malfunctioning IUs to take their vehicle in for a free inspection. The consortium of local banks manages the sale of CashCards and reimburses the authorities daily for the ERP transactions.
When a vehicle approaches the first gantry, the radio antenna communicates with the vehicle IU, determines its validity and the vehicle classification, and instructs the IU to deduct the appropriate ERP fee from the CashCard. When there is no valid ERP deduction, the enforcement camera takes a picture of the rear license plate, and the reason for the violation or the error is recorded.
When the motorist inserts a CashCard into the IU, the display shows the cash balance. When the vehicle goes under the ERP gantry, the IU displays the new balance. There is a low balance indicator which appears when the vehicle goes under an ERP gantry with less than S$5 (US $3.28) is on the CashCard. IUs are color-coded for different class of vehicles, to prevent swapping. Emergency vehicles and police cars are outfitted with special IUs that do not require CashCards. A visitor can rent a temporary IU at a gas station near the border. Alternatively, a foreigner can drive without an IU and pay an ERP fee of $5 for each day of driving in the RZ upon departing the country. The system does not conflict with the privacy of travelers, as it only photographs violators.
The ERP operates on weekdays in the RZ, and during peak periods on selected expressway links. There are no restrictions on Saturday. When the monitored speeds for half-hour intervals on the expressway fall below 45kph, or below 20 kph on the arterial roads, the ERP fees are raised for that half-hour period, and vise versa. Currently, the major roads within the RZ all operate within the desired range of 20-30 kph

Page 16 of 66

Congestion Pricing Response: Section I

during the ERP period of 7:30 a.m.-7 p.m. They do not drop for the half-hour before the congestion charge start, but speeds do drop for a 45 minute interval after the congestion pricing ends at 7 p.m. Unlike the ALS system, a driver is charged every time he or she enters the RZ, so multiple trips are limited. The automated system makes it easy to change fees and hours of operation. It is also relatively easy to change the area boundaries of the congestion pricing zone, by installing and removing ERP gantries. The half-hour increment fees are reviewed at three months' intervals and announced via local newspapers (Menon 2006).
Motorcycles constitute about 15 percent of the total vehicle population, and pay half the fee of a light vehicle. There are about 3 violations per 1000 ERP transactions, and the majority of those are due to drivers forgetting to insert the CashCard in the IU. Taxi commuters have to pay the ERP fee that their trip incurs. When a taxi enters the IU empty, the driver has to pay the fee. Taxi fares have been deregulated recently, and there are ample taxis in the city.
A massive publicity program was put out for a year prior to ERP implementation. Motorists were educated on how to use their IU and CashCards. The ERP gantries were in "test mode" for 3 months with zero charging, before the actual start date. That way, motorists could check that their equipment was working.
The annual revenues from congestion pricing account for 0.3 percent of total government revenue, and are only 8 percent of the annual costs of building, maintaining and operating the land transport infrastructure (including both rail and roads). Congestion revenues are not earmarked for transport related projects. The main purpose of new roads being built is to provide access to new areas of development, to fill in missing links in the road network, and to address congestion by widening in problem areas.
Investment in public transport has been an integral part of a successful ERP system. The heavy rail is anticipated to be expanded to 500 km by 2030. Two private companies have signed a long-term licensing and operating agreement to run the train services, without a government subsidy for O&M. The bus fleet, owned by the same two private companies, is benefited by bus lanes and special bus signals, and is the mainstay of public transport, providing a wide coverage without government subsidies. The rail and bus systems are well integrated, with a common ticket used on both, and careful positioning of stations within easy transfer distance of each other.
Conclusions
New technologies now make it considerably easier to establish and enforce a congestion-pricing mechanism, whether the facility is limited to one lane or established as a cordon, as in the case of Singapore. The most widely-used new technologies include electronic toll collection via transponders found in the individual cars. Technologies, such as license-plate reading, automatic vehicle occupancy

Page 17 of 66

Congestion Pricing Response: Section I

detection, and enforcement gantry lights, are also being developed to assist in enforcement. Such technologies allow for the separation of a congestion-priced lane's users into those who do not have to pay the toll (e.g. a vehicle with two or three people inside) and those who do.
Audiences for electronic toll collection and transponder use have often responded favorably and adapted to using the new technology. However, there may be privacy concerns with some of the enforcement technology; users may react negatively to having their license plate read or their picture taken by a digital camera trying to determine occupancy. Any agency trying to establish a new congestion-pricing facility would have to gauge the mood of the public carefully, as a negative reaction to a proposed new technology could lead to negative publicity for the project as a whole.

Page 18 of 66

Congestion Pricing Response: Section I

Public Acceptance of Congestion Pricing
In the past few years, a number of surveys have been conducted as to public perceptions of congestionpricing projects. While most of these were conducted via telephone, there are two examples of online surveys (Ubbels and Verhoef, 2006; Burris et al., 2007). The most obvious trend, when comparing the various surveys, is that those conducted after a specific congestion-pricing project found more favorable opinions of congestion pricing and tolls than those which were being conducted on the topic of hypothetical tolling projects. Appendix A has a table summarizing the structure and findings of several recent congestion-pricing-related surveys.
Before and after the opening of the MnPASS congestion-pricing facility on Interstate 394 near Minneapolis, NuStats conducted a series of surveys of consumer behavior in the area. Most of the 980 participants in the first of three "waves" of surveys also participated in the second and third "waves" (Zmud et al., 2007). To date the NuStats surveys seem to be the most comprehensive effort at tracking public reactions to a congestion-pricing project over time. The NuStats surveys are included in Appendix A.
Some studies have relied on interviews with participants in past congestion-pricing implementation programs (Evans et al., 2007) or with specific local stakeholders (Benjamin et al., 2007). A few have used focus groups (Godbe Research and Analysis, 1998; Texas Transportation Institute, 2005; Cook Research, 2004) to obtain more detailed user reactions.
How the public regards a travel-demand management measure, such as congestion pricing, depends on a number of factors. Non-coercive measures are more acceptable than coercive measures (Grling and Schuitema, 2007). The more effective the scheme is perceived to be at solving congestion or environmental problems, the weaker the probable public opposition (Grling and Schuitema, 2007). Meanwhile, in the United States, the nature of government is inherently biased against significant policy change, and large projects are vulnerable to "last-minute withdrawal" from political actors if public opposition mounts (Ungemah and Collier, 2007).
There is some suggestion that as congestion problems have worsened over time, and as managed-lanes projects in the United States have gained publicity, public attitudes towards congestion pricing have begun to warm. Benjamin et al. (2007), discussing the possibility of adding HOT lanes to a major arterial (Interstate 40) in a medium-sized city (Greensboro, North Carolina, with an estimated population of 237,000 in 2006) found that approximately as many leaders favored tolls for highways as opposed them. Yet American attitudes towards congestion pricing have been characterized as lukewarm at best (Swisher

Page 19 of 66

Congestion Pricing Response: Section I

and Ungemah, 2006), and previous polls have found little support for time-of-day variable pricing (Weinstein and Dill, 2007).
Public Objections to Congestion Pricing
It should be noted, first, that different existing users will have different attitudes towards a proposed congestion-pricing program. Those who already benefit from the free HOV lane, such as transit riders, drivers of super-low-emissions vehicles (SLEVs), and carpoolers, may balk at the idea of sharing the HOV lane with single drivers, even those paying a fee (Swisher and Ungemah, 2006). Regular users of the road on which the proposed project will be implemented may feel differently towards congestion pricing than occasional users.
Value pricing is widely recognized to be politically difficult because it adds a price to a public service previously perceived as free (Benjamin et al., 2007). A survey of travelers in Dallas and Houston found that the primary reason given for opposition to value pricing was dislike of the tolls; a feeling that taxes had already paid for the road was also frequently cited (Burris et al., 2007). The idea of value pricing as "double taxation" is also shared by many American political leaders (King, Manville, and Shoup, 2007). However, in public opinion studies support for tolls to fund roadway projects often increases when directly compared to raising gas taxes (Weinstein and Dill, 2007).
The political acceptability of congestion pricing may also depend on the plans for the revenue (King, Manville, and Shoup, 2007). One study suggested that revenue uses that benefit individual drivers, such as decreasing road or fuel taxes, was more likely to win public support than revenue uses meant to benefit society as a whole (Grling and Schuitema, 2007). Participants in focus groups for the San Diego I-15 FasTrak reported that while they were unsure of how revenues were being used, they felt that revenues should be used mainly for highway projects. Improving public transportation was cited as a positive use of revenues by some, but overall participants advocated using toll revenues for improving roads and constructing new express lanes (Godbe, 1998). Focus groups in Atlanta recommended that revenues first be used to cover the capital, operations, and maintenance costs associated with converting to or constructing HOT lanes (Meyer et al., 2006).
Several studies have found that prior familiarity with congestion pricing or managed lanes increases the likelihood that the user will support congestion pricing (Kockelman and Kalmanje, 2005; Burris et al., 2007). In the case of the variably-priced HOT lanes on I-394 in the Minneapolis/St. Paul metropolitan area, NuStats conducted three separate surveys; in November/December 2004, before the lane opened; between November 2005 and January 2006, six months after its opening; and a year after its opening, in May-June 2006 (Zmud et al., 2007). One subsequent observation, after a comparison of the results of the three surveys, was that

Page 20 of 66

Congestion Pricing Response: Section I

It seems that when an SP [stated preference] survey is done before respondents have any experience with the actual HOT lane context, their responses may tend to [be] "homogenized" to some extent. After the actual HOT lane system is introduced, on the other hand, respondents may have a much better idea of whether or not they would be willing to pay the toll in specific situations, so their responses will tend to show a wider variance (Zmud et al., 2007).
Moreover, those polled while living in areas without tolls were more likely to consider congestion pricing unfair (Podgorski and Kockelman, 2005).
An additional consideration is the way in which a pricing project is marketed. A study in Oahu found that when pricing was presented as "a time-of-day charge to manage congestion by inducing shifts to transit and travel times," it only received 15% public approval. But when it was presented as "a user fee wherein those using the facility the most pay the most and where fees go toward road development and maintenance", it garnered 42% acceptance (Ungemah and Collier, 2007).
Equity Concerns
Previous studies suggest that public opposition to congestion pricing decreases as the toll decreases (Gaunt et al., 2007). Part of this decline can be attributed to simple self-interest, as users may hope that if congestion pricing is actually implemented, the tolls will be minimal. But concerns over the absolute cost of the toll may also be related to the perception that managed lanes will function as "Lexus lanes," available only to the wealthy. In the second panel survey concerning the MnPASS managed lanes on Interstate 394, the most often cited objection to letting single-occupied vehicles (SOVs) use carpool lanes was that only the rich would benefit; the second-most popular objection was that carpool lanes should be free to all (Zmud, Peterson, and Douma, 2007). The "Lexus lanes" objection was also prominent in early discussions of the I-15 Express Lanes project (Evans et al., 2007).
Ungemah (2007) divides concerns about equity, as relating to congestion-pricing schemes, into five types; participation equity, opportunity equity, modal equity, geographic equity, and income equity; his analysis concentrates on the last two. Geographic equity concerns are similar to environmental-justice concerns in that a community may feel it is being asked to bear the burdens of a particular project seen as regionally beneficial--for example, the community may be host to roads which become more congested as drivers try to avoid tolled roads. Income-equity concerns frequently center around the question of whether an additional toll would be an unacceptable cost burden for low-income communities.

Page 21 of 66

Congestion Pricing Response: Section I

There is some evidence that higher-income potential users suffer less risk from a congestion-pricing scheme than lower-income potential users. Of the respondents to a survey about a potential congestionpricing scheme in the Netherlands, those with higher incomes were less price-sensitive (Ubbels and Verhoef, 2006). An early look at I-15 FasTrak users found they were more likely to have higher levels of income and education, and to own homes, than non-users (Hultgren and Kawada, 1999). But surveys have not found differences between higher- and lower-income users' attitudes towards congestion pricing (Weinstein and Dill, 2007).
Kockelman and Kalmanje (2005) and Gulipalli and Kockelman (2006) have discussed the possibility of credit-based congestion pricing, or CBCP. In this scenario, all travelers would receive an automatic credit to be applied to tolls. Modeling a CBCP policy in the Dallas-Fort Worth area using three different scenarios, Gulipalli and Kockelman (2006) suggest that a majority of users would receive welfare gains. But CBCP has not been implemented in any of the existing American congestion-pricing programs.
Political Difficulties in Implementing Congestion Pricing Programs
There have been a number of failed attempts to implement value-pricing schemes. Portland, Oregon was unable to convert existing HOV lanes to HOT lanes at several congested locations because the public saw the conversion as a removal of capacity (Sullivan, 2003). A variable-toll program for the Chesakpeake Bridge in Maryland was cancelled by the governor, partly on grounds of local concerns (Sullivan, 2003).
In the United Kingdom, although the flat congestion-charge toll has been successfully implemented in London, a road-user charging scheme was voted down in Edinburgh, Scotland, in 2005. The scheme would have set up two cordons around the city, with a one-time charge of 2 for crossing one or both cordons each weekday. Exemptions would have been given to taxis, buses, motorbikes, emergency vehicles, and the disabled, but not to residents of the city. The money would have been put towards transport investment. After a five-stage, three-year public-involvement process, the vote in February 2005 led to a 74.4% rejection of the proposed scheme (Gaunt et al., 2005). The Edinburgh scheme lacked national political support (Gaunt et al., 2005) and thus, as with the Maryland and California schemes, lacked a strong political champion.
Successful managed-lane projects need top-level political support (Swisher and Ungemah, 2006). King, Manville, and Shoup (2007) argue that congestion pricing has historically lacked strong advocates because it lacks a constituency to derive concentrated benefits that exceed the costs. While two groups--drivers for whom the time saved is worth more than the tolls paid, and people who already use transit--benefit from congestion pricing, it hurts those drivers who pay more than their time is worth, drivers who switch to a more convenient route to avoid tolls, and drivers on non-tolled routes who see

Page 22 of 66

Congestion Pricing Response: Section I

traffic increased. This explains why car users mainly voted against the Edinburgh scheme (Gaunt et al., 2005). King, Manville, and Shoup (2007) suggest awarding the revenues collected from congestion pricing to cities, so that city leaders will give their political support to the proposed scheme.

The Texas Transportation Institute (Kuhn et al., 2005), summarizing five years of research, made a series of recommendations as to improving the chance of public acceptance of a congestion pricing program:
Present the managed lane as an additional choice for commuters. Emphasize that managed lanes are not a short-term "band-aid" but one tool in a long-term
comprehensive plan. Explain that variable pricing increases will increase capacity. Assure users, especially users unfamiliar with electronic toll collection, that ETC will not
impede already-congested travel. Make sure an enforcement mechanism is in place so that users do not suspect that they
might pay while others break the law. Define and communicate how the resulting revenue will be used from the outset of the
project. Present pricing as a means to raise revenue for projects that might not otherwise be funded
(Kuhn et al., 2005).

Revenue use can become a particularly vexing question for users. In a focus group study with users of the I-15 Express Lanes (Godbe Research and Analysis, 1998), respondents said they did not know how the revenue from the HOT lanes was being spent, and were reduced to making guesses, including paying government employees, being used to build a new freeway, or putting "empty buses" on the road.

A second study, examining congestion-pricing projects in California (both successfully and unsuccessfully implemented), found that the public is more likely to accept a project where the sponsoring agency:
is responding to a serious congestion or environmental concern; keeps stakeholder groups appraised of project details, and solicits their input; anticipates effects of the project on surrounding areas, and develops mitigation measures
accordingly; and mounts a comprehensive public outreach campaign (Evans et al., 2007).

Public Education and Outreach
Support for tolling projects increases when respondents are provided more information about the topic (Weinstein and Dill, 2007). A directed and coordinated public education and outreach effort can provide the public with the necessary information to form opinions about a value pricing or managed lanes project. The successful implementation of I-394 MnPASS, after a decade and several failed attempts,

Page 23 of 66

Congestion Pricing Response: Section I

can in part be attributed to this phenomenon. After a proposal for I-394 HOV conversion to HOT was pulled due to public opposition, Minnesota resurrected the project in 2001 with a revised public outreach strategy for value pricing (supported by a grant from FHWA's Value Pricing Pilot Program) and was successful with project approval in 2003 (Munnich and Loveland, 2005).
Public education and political leadership were viewed by the MnPASS project team as crucial and so they had hired a communications consultant to help coordinate efforts and an engineering firm with to answer detailed questions and support the education component. Recognizing that there is more public trust for an initiative led by an academic institution rather than a governmental agency, the Humphrey Institute at University of Minnesota organized a Value Pricing Advisory Task Force of key, diverse community stakeholders. The public education effort focused on building strong stakeholder relationships. The outreach team held dozens of small group visits with legislators, interest group leaders, state government leaders, municipal officials, and transportation and transit advocates. They also held large group dialogues with civic groups and several public policy roundtable discussions between experts and the public, and conducted marketing research and extensive media outreach to disseminate information (Munnich and Loveland, 2005).
The MnPASS team, through failed and successful efforts, developed a set of "lessons learned" as a reference for other value pricing project teams. These lessons include:
It is difficult to maximize public outreach efforts without the support of higher-level officials who will share their advocacy with the public. Minnesota's governor participated in conversations with value pricing advocates.
A "Grasstops" Coalition of community leaders is needed. MnPASS's project team reached out to community leaders, discussed the concepts, and then asked supportive leaders to help contact their constituents and peers.
An unanswered question (or accusation) can become an accusation believed. Minnesota formed a public outreach team to quickly answer any questions from the public. Common public concerns included technical feasibility, equity, impact on HOV use, and public acceptance.
Constituents must understand the benefits that they will receive. Minnesota used customized messages (in addition to common themes) for each individual audience. For example, messages to businesses focused on reducing the cost of congestion and increasing reliability while messages to carpool advocates made assurances that they would maintain priority on the HOT lane and have more choices.
The project team should focus on the benefits offered by value pricing rather than the costs; in other words, use terms that accentuate the positive. Minnesota uses "express lanes" and

Page 24 of 66

Congestion Pricing Response: Section I

"MnPASS" rather than terms that emphasize user costs (like value pricing, congestion pricing, toll lanes, etc.) (Munnichand Loveland, 2005).
Ungemah and Collier (2007) offer additional "lessons" for public education approaches. These include: Educate citizens about the current system of transportation funding to serve as a comparison to congestion/value pricing. Develop a simple message to communicate the concept of congestion pricing/managed lanes to the public. For example, in a survey about Houston's I-10 HOT lanes, half of non-users were either unaware of QuickRide or misinformed about its logistics. Use initial and on-going marketing--it is the key to success. This includes branding the project early on to make it identifiable, as Minnesota did by referring to their project as MnPASS. Raise public awareness of why pricing is being pursued instead of using a "traditional" financing scheme, as it is often a means of more efficiently allocating transportation resources and of advancing the financial feasibility of a project. Be prepared to answer the revenue-spending question. Finally, accentuate the positive.
Conclusions
While there is some evidence that American audiences are becoming slightly more comfortable with the idea of congestion pricing, the idea has not yet been matter-of-factly accepted. Familiarity seems to lead to more positive responses: people who have used a congestion-pricing facility or a toll lane before seem to react more positively to the idea of a congestion-pricing facility than do those who have not. But potential users may react negatively if they believe that a "free" facility is being taken away from them. Turning an HOV lane into an HOT lane may be more acceptable to the majority of users, but it may provoke opposition from existing HOV users. Political support can be difficult to obtain. Users may also cite equity as a concern, fearing that a congestion-free drive will be a privilege limited only to those who can afford it.
There are certain things a public agency can do to mitigate these concerns. First, especially if this is the first proposed congestion-pricing facility in a region, prepare a detailed and comprehensive outreach program. Keep potential users informed as to the features and predicted consequences of the facility, and be willing to adjust the proposal based on their concerns. Second, anticipate heightened scrutiny in some areas, such as the potential equity issues and the planned destination of the resulting revenue. Finally, be able to explain how a congestion-pricing facility will bring benefits to the surrounding area; be able to say what those benefits might be, and why they are worth incurring the costs of a new project.

Page 25 of 66

Congestion Pricing Response: Section I

Case Studies
State Route 91 Express Lanes, California
Background
Figure 6: SR-91 Express Lanes Map

Source: FHWA (2003, Chapter 7)
When it opened in 1995, the tolled facility on State Route 91 (hereafter SR-91) was the only variablypriced road project in the world, and the first example of an Express lane/HOT lane facility in the U.S. Unlike later cases which involved an HOV to HOT conversion, this facility was built as a toll road that allowed free access to HOV 3+ vehicles. The 10-mile, four-lane HOT facility built in the median of SR-91 was completed in December 1995.
The Riverside/State Route 91 Freeway, in Orange County, California, is one of the most heavily congested highways in the United States. Prior to the opening of a tolled facility, peak-period delays of 20-40 minutes were common (Sullivan and Burris, 2006). The SR-91 congestion-pricing facility had a relatively difficult political birth, being approved after a statewide bond issue for highway improvements was narrowly voted down by California voters (Evans et al., 2007). It was eventually completed for $134 million, of which $82 million was leasehold and equipment costs (Sullivan and Burris, 2006).

Page 26 of 66

Congestion Pricing Response: Section I

Of interest is the fact that this Express lane facility started out as a for-profit, privately-operated venture. This was one of the four public-private partnerships made possible by AB 680 legislation in California. The state was not responsible for construction costs, and the right-of-way costs were negligible due to the prior availability of space for additional lanes in the median. Under the 35-year franchise agreement between the State of California and California Private Transportation Company (CPTC), the CPTC would construct and operate the facility, with the right-of-way leased from the state. The agreement stipulated that no improvements could be made to the general lanes on SR-91 (the non-compete clause), so that the profitability of the project would not be undermined. In 2002, Orange County Transportation Authority (OCTA) purchased the SR-91 Express lanes and the operational franchise agreement from CPTC for $207.5 million, at which point the goal of variable pricing on the facility could be focused on maximizing traffic flow, not revenue (FHWA, 2003, Chapter 7).
Implementation and Operations
The HOT lanes are separated from the general purpose lanes by a painted buffer and plastic pylons. All vehicles are issued a transponder, including 3+ occupant carpools. At the end of the 200405 fiscal year, there were over 172,000 transponders in circulation (FHWA, 2006).
The physical separation of SR-91 Express lanes from general purpose lanes makes enforcement easier than it would be for a striping-separated HOT facility. Carpool vehicles are required to carry a transponder. SR-91 has a contract with California Highway Patrol (CHP), which covers the costs for all CHP services 24 hours a day and involves monitoring of vehicle occupancies.
The SR-91 case is of utmost importance both because of the length of time it has been in operation and the number of changes that have been made to the tolling system over time. From January 1998 to 2003, HOV 3+ vehicles had to pay 50% of regular tolls (Sullivan, 2000). Beginning in May 2003 (following the change in ownership), HOV 3+ vehicles, motorcyclists, disabled license plate carriers and zero-emission vehicles travel free except when traveling Eastbound, Monday through Friday between the hours of 4:00 p.m. and 6:00 p.m., when they pay 50% of the regular toll. As of July 2007, tolls range from $1.20 to $9.50 (see Figure 7, below).

Page 27 of 66

Congestion Pricing Response: Section I

Figure 7: SR-91 Express Lanes Toll Schedule, July 2007

Source: Orange County Transportation Authority (2007)
Public Response
Use of the SR-91 Express Lanes has grown consistently since the facility opened in 1995. Figure 8 shows historical growth in average daily traffic and gross annual potential revenue (Vollmer Associates, 2007).

Page 28 of 66

Congestion Pricing Response: Section I

Figure 8: Average Daily Traffic and Revenue Growth, SR-91, 1996-2006

Source: Vollmer Associates (2007)
In the 200506 fiscal year, the SR-91 Express lanes serviced approximately 38,800 vehicles per day, resulting in $37.5 million in gross potential annual revenue (Vollmer Associates, 2007). One estimate put the value of time travel saved by the express-lane facility at $34.9 million in 2005 alone (Sullivan and Burris, 2006).
The SR-91 lanes are not being used exclusively by wealthy drivers, but by a broad swath of society (Evans et al., 2007). A travel profile of SR-91Express Lane users conducted immediately after its opening found approval from both low- and high-income households (Ungemah, 2007).
Public support for SR-91 did decline after its opening--not because of the addition of a toll, or because of perceived inequities in tolling, but due to the nature of the public-private partnership agreement and the non-compete clause (Collier and Goodin, 2002). Public opposition was also based on the idea that CPTC was making excess profits off high tolls. When OCTA bought the SR-91 facility from CPTC, those concerns subsided.

Page 29 of 66

Congestion Pricing Response: Section I

Interstate 15 Express Lanes, San Diego
Background
Figure 9: I-15 Express Lanes Map, San Diego

Source: FHWA (2003)

Unlike subsequent congestion-pricing programs in the United States, the I-15 Express Lanes were not created primarily to reduce congestion, but to create an additional source of revenue with which to fund public transit in the area (Evans et al., 2007). The idea of variably-priced lanes was first considered by the San Diego Association of Governments (SANDAG) in 1991. Jan Goldsmith, then mayor of Poway, a city northeast of San Diego, and member of SANDAG's board, would function as the Express Lanes project's champion (Evans et al., 2007; Hultgren and Kawada, 1999). For example, Goldsmith was able to have the California state legislature pass an exception to the rule allowing only HOV-2+ vehicles to use HOV lanes, so that the I-15 Express Lanes demonstration project could be built (Hultgren and Kawada, 1999).

The conversion from HOV lanes, implemented in 1988, to HOT lanes took place gradually. During the demonstration phase, which lasted from December 1996 to March 1998, solo drivers could purchase monthly passes to ride in the Express Lanes, while carpoolers could ride for free (much like the process currently in use on the HOT facility in Salt Lake City, Utah). FasTrak transponders were not issued until

Page 30 of 66

Congestion Pricing Response: Section I

the second phase, which began in March 1998 (Brownstone et al., 2002). Capital costs of converting HOV lanes to HOT added up to $1.85 million (Poole and Orski, 2000).
Implementation and Operations
The I-15 FasTrak is an eight-mile, 2-lane peak-period reversible HOT facility. HOV2+ vehicles may use the facility at no cost. There are only two entrance/exit points -- one at either end of the eight miles. There are two lanes at the center of the right of way that are separated from the general purpose lanes by permanent barriers. The lanes run eight miles southbound into San Diego from 5:30 am to 11:00 am. The lanes are then reversed and they run northbound out of San Diego from noon until 7:00 pm. The lanes are only closed for the short period (about 1 hour) that is required to reverse the direction of travel and on weekday evenings (7:00 pm to 5:30 am the following day). The lanes operate northbound at present on Saturdays and Sundays.
Figure 10: Current configuration of managed lanes on I-15

Source: FHWA and FTA (2006)
The goal of the I-15 facility is to manage congestion, while keeping the level of service (LOS) at C or better. The toll level is changed according to the congestion level. Loop detectors are used to measure the volume of vehicles on the lanes. The toll, which is set dynamically (based upon real-time traffic conditions), usually ranges from 50 cents to $4 (although it can be set as high as $8), and is updated automatically every six minutes.
The toll is charged only to single occupant vehicles (SOV). High occupancy vehicles (HOV), containing two or more passengers, are exempt from the toll. FasTrak transponder users are given special bags (static bags) in which to place the transponder when their vehicle has 2 or more occupants. The static bags prevent the transponders from being read and prevent charges from appearing on user accounts. Toll enforcement is an issue that is difficult to address and the San Diego region is still looking for better methods. Tolls are charged electronically and there are currently about 30,000 toll transponders in circulation. However, approximately 77% of the traffic is HOV vehicles. There was an increase in usage when FasTrak transponders were introduced to allow users to pay to access the lanes. If HOV and transit

Page 31 of 66

Congestion Pricing Response: Section I

vehicles ever produced LOS C (or worse) without the tolled vehicles, the HOV requirement would be raised to 3 occupants or more.
Revenue for the I-15 lanes is $2 million per year, which is split evenly between transit and facility operations. The $1 million designated for transit supports the Inland Breeze bus service. (FHWA and FTA 2006). A planned expansion, to be completed by 2012, will extend the Express Lanes to more than 20 miles and will include a bus rapid transit (BRT) system (FHWA and FTA, 2006).
Public Response
Members of a 1998 focus group said they were generally satisfied with the Express Lanes program, although regular users spoke more positively of the program than occasional users (Godbe Research and Analysis, 1998). According to the results of an 800-person telephone survey of I-15 Express Lane users completed in 2001, motorists of all income levels are able to recognize the benefits of HOT lanes:
91 percent of those surveyed approved of the travel time savings options provided by the I-15 Express Lanes;
66 percent of non-Express lane users support the Express Lanes concept; 73 percent of non-Express Lanes users agree that the HOT lane reduces congestion in the
corridor; 89 percent of Express Lanes users support the extension of the Express Lanes; and when considering the statement "People who drive alone should be able to use the I-15
Express Lanes for a fee," 80% of the lowest income motorists using the I-15 corridor agreed with it, and low income users were more likely to support the statement than the highest income users (FHWA, 2003). In effect, such results diffuse the equity concerns raised in regards to HOT lanes and their potential higher usage by high-income populations.
The case of San Diego has suggested that converting an HOV lane to an HOT lane is not as politically charged as creating a new congestion-priced lane, since SOV drivers gain, rather than lose, options (King, Manville, and Shoup, 2007).
Houston QuickRide, I-10 and US-290
Background
The QuickRide program has been implemented on two interstate highways that feed into Houston from the west, Interstate 10 and US-290. The former is also known as the "Katy Freeway" and the latter as the "Northwest Freeway" (see Figure 11). Both highways had existing HOV lanes prior to the HOT

Page 32 of 66

Congestion Pricing Response: Section I

conversion. The first HOV lanes in the Houston area were built in 1979; the Katy HOV lane opened in 1984 (Burris and Stockton, 2004). By the mid-1980s, state officials had learned that if vehicles with just two people inside (HOV-2 vehicles) were permitted to use the lane, it rapidly became congested during peak hours, but allowing only vehicles with three or more people inside (HOV-3+ vehicles) led to an inefficient use of existing capacity (Burris and Stockton, 2004). Thus the idea of allowing HOV-3+ users to use the lanes for free, while charging HOV-2 users, arose as an adjustment to observed traffic conditions over time.
Figure 11: Map of Katy Freeway and Northwest Freeway

Source: Texas Transportation Institute (2003a) Figure 12: Katy Freeway (I-10 West) HOT Lane

Page 33 of 66

Congestion Pricing Response: Section I

Source: Turnbull (2003)

Implementation and Operations
The Katy Freeway HOV lane was converted to an HOT lane in January 1998, and the Northwest Freeway HOV lane to an HOT lane in November 2000. In both cases, the lanes are reversible and restricted to use by HOV 2+ vehicles. HOV 3+ vehicles travel for free, while vehicles with two people must pay $2 during the congestion periods. As Table 1 shows, the hours of operation differ between the Katy Freeway and the Northwest Freeway.

Days Open

Table 1: Hours of Operation, Houston QuickRide (asterisk denotes application to Katy Freeway only)

Hours of Operation

Direction

Minimum Capacity

5 to 6:45 a.m.

inbound

2

6:45 to 8 a.m.

inbound 3

QuickRide

2

Weekdays

8 to 11 a.m. 2 to 5 p.m.

inbound

2

outbound

2

5 to 6 p.m.*

outbound 3

QuickRide*

2

6 to 8 p.m.

outbound

2

Saturday*

5 a.m. to 8 p.m.

outbound

2

Sunday*

5 a.m. to 8 p.m.

inbound

2

Source: Texas Transportation Institute (2003)

Enforcement on Houston HOV and HOT facilities is provided by METRO police officers, with the goal of providing safe and efficient operation. At least one METRO police officer is located in the HOT lane corridor during operational hours, responsible for patrolling and monitoring the corridor for violators of the occupancy requirement and other and regulations. Specific enforcement areas are set up to not interfere

Page 34 of 66

Congestion Pricing Response: Section I

with the flow of traffic, at the entrance points to the facility (see Figure 13 below). The presence of concrete barriers simplifies the enforcement task (Cothron, Skowronek et al., 2003).
A HERO self-enforcement program is in effect in the Houston area, where a dedicated phone number is available for motorists to call and leave a message if they notice a driver who violates the rules of an HOT or HOV facility. The reported violator receives a warning letter from the METRO police (Cothron, Skowronek et al., 2003). Such a program can be useful for mailing educational materials, even if it does not have the "teeth" to fine the reported violator.
Figure 13: A motorist passes through an enforcement zone while heading westbound on the I-10 Katy Freeway in Houston, TX

Source: Obenberg (2004)
Future expansion and changes to the QuickRide program are likely. One stated-preference survey, conducted in late 2003, suggested that single-occupancy vehicle drivers would pay to use the QuickRide facilities, given sufficient time savings. Based on the survey, Burris and Xu (2006) propose an off-peak toll schedule for SOVs that would allow approximately 2,000 more QuickRide users and generate approximately $4,500 in additional revenue per day. As of July 2007, however, QuickRide was not open to single-occupancy vehicles (Texas Transportation Institute, 2003).

Page 35 of 66

Congestion Pricing Response: Section I

Public Response
As of April 2002, over 1,500 transponders were in circulation for use on the two facilities, and an average of 160 users traversed the two facilities each day (FHWA, 2003). One estimate put the value of time savings over ten years of QuickRide use at $2.35 million, and fuel savings at $13,500 (Burris and Stockton, 2004).
Burris and Hannay (2003), surveying both users and non-users of QuickRide in 1998, found that while there were no significant differences in perceptions or usage of QuickRide amongst different socioeconomic groups, the surveyed users of QuickRide had significantly higher incomes than nonenrolled drivers.
MnPASS, Interstate 394, Minnesota
Background
Figure 14: I-394 HOT Facility

Source: Minnesota DOT (2007 B)
Interstate 394 runs 9.5 miles, with its eastern terminus in Minneapolis and its western terminus in Minnetonka. As such it serves as a route for commuters from the western suburbs of the Minneapolis-St. Paul metropolitan area to drive into downtown Minneapolis. Previously existing HOV lanes were converted to variably-priced HOT lanes in May 2005. The goal of the MnPASS HOT system is to maintain the free-flow nature of the managed lane and improve the overall effectiveness of the corridor (Douma, Zmud, and Patterson, 2005).

Page 36 of 66

Congestion Pricing Response: Section I

The managed lanes on Interstate 394 came about only after several setbacks. In 1996 a proposed public-private partnership to build a toll road on Minnesota Highway 212 (now incorporated into Highways 5 and 36) was abandoned after local opposition led to a city council veto. A year later, the state Department of Transportation (MnDOT) floated the idea of converting the existing HOV lane to an HOT lane; that, too, attracted local opposition, including the placement of full-page anti-HOT ads in newspapers by a local political leader (Munnich and Loveland, 2005). In the time between the rejection of the HOT plan in 1997 and its endorsement by Minnesota's then-governor in 2003, MnDOT and the Hubert H. Humphrey Institute of Public Affairs at the University of Minnesota used a Value Pricing Advisory Task Force to solicit information from stakeholders, and embarked on an education campaign for the public (Munnich and Loveland, 2005).
One advantage for the MnPASS system was that it was financed not by a public bond issue, but by a loan from a downtown parking ramp fund. The state legislation which authorized the adaptation of the existing HOV lane to an HOT lane requires that a portion of any excess revenue will go to transit improvements in the corridor (Munnich and Buckeye, 2007).
Implementation and Operations
The MnPASS facility consists of a 3-mile section east of I-100 with 2 barrier-separated reversible lanes (eastbound 6 a.m.1 p.m.; westbound 2 p.m.5 a.m.) and an 8-mile section west of Interstate 100 with one HOT lane in each direction, with the HOT lane separated from the general-purpose lanes by a 2-footwide double white line. Using plastic pylons was not feasible on the project due to incompatibility with snow plows during winter conditions, and building a concrete barrier is not in the plans due to engineering constraints (Halvorson, Buckeye, et al., 2006). Toll revenue is re-invested in the corridor (Douma, Zmud, and Patterson, 2005).
The 8-mile section separated from the general traffic by double white line allows for multiple entry points. Each stripe is eight inches wide, with an eight inch space between the lines, for a total width of two feet. While the double-white strip buffer has not been previously used in Minnesota, it has been successfully used at other U.S. locations to delineate HOV lanes. It is illegal to cross the double white stripe, and violators are subject to fines (Halvorson, Buckeye et al. 2006).
At designated entry and exit points, the double white stripe is replaced by a dashed line, which is legal to cross (see Figure 15, below). Most access points are over one half mile long (with at least a quarter mile required) (Halvorson, Buckeye et al. 2006).

Page 37 of 66

Congestion Pricing Response: Section I

Figure 15: Line-Separated HOT Lane Entry and Egress

Source: Halvorson, Buckeye et al. (2006)
Because the facility has multiple entry and exit points, tolls vary not only by time of day but also by distance traveled. The enforcement presents a special challenge, due to the possibility of illegal weaving in and out of the lane, as well as occupancy violation. Drivers caught crossing the double white line receive a $142 moving violation fine (Minnesota DOT 2007b). The Minnesota State Patrol oversees the enforcement, assisted by the City of Minneapolis Police Department, Metropolitan Transit Police and the City of Golden Valley Police Department. While police officers rely on visual verification, technological advances, such as overhead gantry lights and MERs, discussed previously, help the better monitor the electronic tolling aspect. The annual cost of enforcement amounts to about $200,000 (Buckeye, 2007).
Compared with pre-MnPASS violation rates in the HOV facility, the violation rates on I-394 have decreased. In particular, in the stripe-separated section of the corridor violation rates fell from 20% to 9% (Cambridge Systematics, Inc. et al, 2006). Table 2 (below) illustrates the comparison, with an example of violation increase on a non-MnPASS corridor.

Page 38 of 66

Congestion Pricing Response: Section I

Table 2: Summary Comparison of Pre- and Post-MnPASS HOV Lane Violation Rates

Source: Cambridge Systematics, Inc. et al. (2006)
When interviewed, representatives of three of the four transit providers (Metro Transit, Plymouth Metrolink, and Prior Lake Laker Lines) indicated that the implementation of MnPASS on I-394 has had a negligible effect on their operations and travel times (Cambridge Systematics, Inc. et al., 2006).
The traffic volume in the MnPASS has increased after implementation of the program, yet the travel speeds have not been negatively affected. It appears that the pricing algorithm worked well to maintain the speeds in the HOT lane, with a minor exception at one observed location. In the general purpose lanes, the traffic volumes have slightly decreased, and the travel speeds have experienced a minor increase. This relative increase is mostly felt on days with highest traffic volumes, or when incidents occur. It appears that MnPASS program has been effective in mitigating the delay on the worst travel days, and decreasing the travel time variability in the corridor (Cambridge Systematics, Inc. et al. 2006).
Public Response
When the variably-priced HOT lanes opened on May 17, 2005, about 4,000 electronic transponders had been leased. By the end of 2005, that number had more than doubled, to 9,300 (Halvorson and Buckeye, 2006). In 2007, MnPASS reported more than 11,100 transponders leased (Minnesota Department of Transportation, 2005). Munnich and Buckeye (2007) report that the MnPASS users use the lane about twice a week on average, less often than expected.
Public response to the MnPASS lanes seems to have been largely favorable. By 2006 nearly 60% of the surveyed public in Minnesota supported the option to pay a fee and bypass congestion (Halvorson and Buckeye, 2006). Despite continued concerns that the I-394 lanes would disproportionately benefit wealthier drivers, drivers of all income levels use the lane (Munnich and Buckeye, 2007). A 2007 Wall Street Journal article on the MnPASS lanes included positive quotes from drivers, although a representative of the American Automotive Association's Minneapolis branch expressed the organization's position that the lanes should be available to all drivers at all times (Machalaba, 2007).

Page 39 of 66

Congestion Pricing Response: Section I

The NuStats surveys found that support for the project was strong, with nearly two-thirds of those polled saying that allowing SOV drivers in the HOV lane for a fee was a good idea (Zmud et al., 2007).

Through surveys conducted after the MnPASS implementation, it was found that MnPASS users have a positive view of HOT lanes performance (see Table 3, below).

Table 3: Results from MnPASS User Surveys

Percentage of those surveyed...

Who reported satisfaction with traffic speeds in the HOT 88%

lanes

Who reported satisfaction with dynamic pricing

65%

Who reported satisfaction with safety of merging

65%

Who described their travel experience as "enjoyable"

68%

Source: NuStats (2006); Berman (2007); Minnesota DOT (2007a)

In addition, the approval of HOT lanes is widespread across various income groups (Figure 16). Sixty-five percent of respondents to the survey conducted in spring of 2006, a year after MnPASS implementation, thought that MnPASS was a good idea (NuStats, 2006).

Figure 16: Percentage of Minneapolis Consumers Surveyed Who Approve of Allowing SingleOccupancy Drivers to Use the Carpool Lane for a Fee

100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%

Lower-Income

Mid-Income

Higher-Income

Source: Berman (2007)

Fall 2004 Fall 2005 Spring 2006

One early setback actually may have increased the credibility of the MnDOT in managing the MnPASS lanes (Munnich and Buckeye, 2007). Soon after the MnPASS facility opened, westbound (reversecommute) commuters in one portion of I-394 began experiencing greater congestion in the general purpose lanes. The immediate public outcry was answered by an adjustment in the hours of operation for

Page 40 of 66

Congestion Pricing Response: Section I

the MnPASS, giving more access to general-purpose users. The addition of an auxiliary lane in the fall of 2005 further reduced congestion, and the negative feedback quickly subsided.

Interstate 25/US-36, Colorado

Background

Figure 17: Map of I-25 HOV/Tolled Express Lanes

Source: Colorado DOT (2007 C)
The Colorado Department of Transportation (CDOT) opened reversible, high-occupancy vehicle lanes (HOV) on the 7-mile stretch of I-25 between Denver and US-36 in 1994. However, the lanes had significant unused capacity because they carried fewer cars than the adjacent general-purpose lanes (Stegman, 2007). As in Houston, the HOT program evolved as a way to take advantage of that unused capacity.

Page 41 of 66

Congestion Pricing Response: Section I

In the summer of 2002, the Federal Highway Administration awarded 12 pilot projects in value pricing, including the possible conversion of HOV lanes on I-25 in Denver to HOT lanes (Engineering News Record, 2003). The conversion of HOV lanes to HOT lanes would include the installation of plastic pylons, message signs, and transponder equipment.
The North I-25 Front Range EIS, the three-year study contracted by the CDOT, was undertaken in 2004 to determine the "effect of adding various transportation improvements in northern Colorado on the lives of residents and commuters in the area," focusing on the addition of lanes and safety features on I-25 among others. The study area included 70 miles of I-25--north from Wellington, south to Denver, east from US-85 and west to US-287. This area incorporated seven counties and two metropolitan planning groups (McCombs, 2005).
The Denver HOT lanes were influenced by a Denver metropolitan area resident survey that determined that 67.9% of residents believed that the creation of these lanes was an effective way for funding additional highway lanes. The study also found that 74.4% of residents preferred tolled lanes over increasing taxes.
On June 1, 2006, express lanes and high-occupancy toll lanes along 7-mile stretch of I-25 between Denver and US-36 opened. CDOT received a Federal Value Pricing Grant of $2.8 million from the US Department of Transportation to start the program. The program was implemented in partnership with transportation agencies of the area, including Denver's Regional Transportation District (RTD), the Denver Regional Council of Governments (DRCOG), the City and County of Denver, the Federal Highway Administration, and the Federal Transit Administration. The total cost of the program was approximately $9 million and included two feasibility studies, technology components, construction, and a reserve fund for two years of maintenance and operation costs. Originally funded using the grant and taxpayer money, the HOT program is now completely funded by toll revenue.
Implementation and Operations
The facility consists of a 6.6-mile-long stretch of 2 reversible, barrier-separated lanes and one tolling station. The transponder used is the same as the one used on E-407 (outside Toronto) and in Houston. One of the lanes is reserved for HOV vehicles, while the other lane is meant for vehicles paying a toll. An additional 15-foot enforcement lane allows police car access for enforcement. Camera enforcement is used in addition to police patrols. Drivers pay between $0.50 and $3.25 per trip. Figure 18 illustrates the configuration of the I-25 HOT Lanes.

Page 42 of 66

Congestion Pricing Response: Section I

Figure 18: Configuration of I-25 HOT Lanes

Source: Colorado DOT (2007 B)
Due to the barrier separation, limited number of entry and egress points, and space for a 15-foot-long enforcement lane, the enforcement efforts on I-25 Express lanes have been quite successful. As the cars pass through the tolling point, they separate into two lanes by status (SOVs and HOVs). Self-declared HOVs go free, and SOVs pay a charge via a transponder. A police officer can monitor from a location adjacent to the tolling point. If a carpool vehicle goes through the toll lane by mistake, the driver will either receive a charge on their transponder (if present), or a violation ticket in the mail, just like a SOV driver without a transponder. If an SOV driver tries to use the HOV lane illegally to avoid toll, the police officer monitoring the facility can issue a citation (Colorado DOT, 2007b).
Table 4, below, shows the number of citations issued on Interstate 25 between June 2006 and March 2007. Both toll citations and HOV citations have declined since the first month of operations, suggesting that users might be adjusting to the new facility rules.

Page 43 of 66

Congestion Pricing Response: Section I

Table 4: Citations, I-25 Express Lanes, June 2006March 2007
Source: Colorado DOT (2007 A) Figure 19: I-25 HOT Monthly Traffic Volumes, June 2006March 2007

Source: Colorado DOT (2007a)

Page 44 of 66

Congestion Pricing Response: Section I

Public Response
Figure 20: I-25 Projected and Actual Revenue, First 10 Months of Operation

Source: Colorado DOT (2007)
Unlike the MnPASS project, early revenue collection from the I-25 HOT lanes have exceeded projections. Between the opening in June 2006 and late March 2007, revenues of $1.58 million had been collected; the Colorado Department of Transportation (CDOT) had expected to collect $800,000 the first year (Cada, 2007). Peggy Catlin, the Deputy Executive Director of CDOT and the Acting Director of the Colorado Tolling Enterprise, testified in a Congressional hearing in June 2007 that the I-25 congestion pricing project was "off to a solid start" but added, "The project's success has been and is still largely dependent upon public perception and partner relationships" (Catlin, 2007).
Conclusions
The five case studies presented here differ in terms of age of project, initial purpose, method of enforcement, and problem addressed. San Diego hoped to raise more money for transit; Houston and Colorado were trying to put their HOV lanes to more efficient use; the SR-91 and MnPASS projects were undertaken for congestion relief; the Colorado project was in part spurred on by increased FHWA interest in congestion pricing. It is clear that there is no set "formula" for a congestion-pricing facility.

Page 45 of 66

Congestion Pricing Response: Section I

However, we can note some similarities between the five cases. All five have barrier-separated sections; four have reversible sections. Four of the five are able to balance HOVs and SOVs in the same lane; while enforcement has been a difficulty, it should be recognized that a congestion-priced lane can accommodate both carpoolers and single drivers. We can tentatively conclude that (with the exception of SR-91, which had the additional variable of a prominent public-private partnership coming under fire) all show a trend of consumer acceptance of the congestion-priced facility rising after it opened. This is true whether or not SOVs have been allowed to use the facility. Thus, it may be that the most difficult obstacles for a congestion-pricing project are faced before implementation can begin.

Page 46 of 66

Congestion Pricing Response: Section I

Summary: Case Studies

Project State

Project Website

SR-91 Express CA Lanes

http://www.91expresslanes.com/

I-15 Express CA Lanes

http://www.sandag.org/index.asp?classid =29&fuseaction=home.classhome

Houston Quickride

TX

http://www.quickride.org/

I-394 MnPASS

MN

http://www.mnpass.org/

I-25/US36

CO

http://www.dot.state.co.us/CTE/ExpressL anes/index.cfm

Type of Congestion
Pricing
Barrier-separated HOV-3+/HOT lane
Reversible, barrier-separated HOV-2+/HOT lane
Reversible, barrier-separated HOV-3+/ HOT-2 lane HOV-2+/HOT lane, with one barrier-separated reversible section and one section separated by striping Reversible, barrier-separated area with separate lanes for HOV-2+ and SOVs.

Main Challenges
Implementation was politically difficult; public reacted strongly against public-private partnership after lane had gone into use. Toll enforcement, and distinguishing between HOVs-2+ (who do not pay the toll) and SOVs (who do). Balancing access between HOV users and toll payers for maximum efficiency. Initial political opposition to congestion pricing had to be overcome; enforcement remains an issue.
Coordination between the multiple municipal and county governments that stood to be affected by the facility.

Public Response
Generally positive once the lane was returned to public control.
Less of an initial negative reaction than in other cases, since the lane was converted from an existing HOV lane. Somewhat positive, but because use is limited to HOVs-2+, number of users is lower than with other cases.
Has been increasingly positive as users have become more familiar with the system.
Revenue collection has been higher than predicted. Public surveys before facility was built expressed a preference for HOT lanes over increased taxes.

Page 47 of 66

Congestion Pricing Response: Section I

Other Examples of Congestion-Pricing Projects
Introduction
So far this review has concentrated mainly on five case studies. However, as congestion has increased in cities throughout the world, congestion-pricing projects have attracted worldwide attention. In this final section, we review innovative pricing projects both inside and outside the United States. While these projects may be harder to draw universal lessons from, because they are new (or proposed) or because the circumstances are so unique, they give an idea of how congestion-pricing strategies can be adapted to different environments.
Within the United States
Salt Lake City
Figure 21: Map of HOT Lanes on I-15, Utah

Source: Warburton (2006)
The 38-mile-long, non-barrier-separated HOT facility in Salt Lake City is the most recent and the longest addition to the list of HOT lanes in the United States. Converted from previously existing HOV lanes, I-15 Express Lanes (one in each direction) are double solid white line separated, with 16 access points

Page 48 of 66

Congestion Pricing Response: Section I

marked by a white dotted line. Each access point is 3,000 feet long. HOV-2+, motorcycles, emergency vehicles, buses, and clean-fuel vehicles can use the lanes for free.
The pricing and enforcement scheme is fairly simple: SOVs can buy a monthly decal for $50, and each month's decal is in a different color. Once a driver signs up for the program, he or she is automatically issued a new decal each month. Utah Highway Patrol officers are responsible for HOT lane enforcement. A solo driver using the lanes without a decal is issued a citation with a fine of $92 in Salt Lake County and $82 in Utah County (Utah DOT, 2006).
One lane on I-15 is estimated to have a capacity of 1,500 cars, and can preserve a minimum speed of 55 mph during peak travel times. Before conversion, the HOV Lanes carried between 650 and 750 vehicles per lane per hour. Currently, up to 1350 solo drivers per month can purchase the decal and use the Express Lanes. Monitoring of the lane ensures free-flow conditions. The price and number of decals sold are subject to change to ensure that speeds in the Express Lanes do not drop below 55 mph. The facility may be converted to electronic tolling in the future, at an estimated cost of around $15 million (Utah DOT, 2006). While this pricing scheme can be seen as a first step towards introducing variable pricing in combination with electronic tolling, as it was done on I-15 Express Lanes in California, at this point the system cannot be classified as variable pricing. If the facility were to become congested under the current rate structure, implementing a different rate structure would require a significant time lag (at least a month, until drivers renew their decals).
Manhattan
On April 22, 2007, Michael Bloomberg, the mayor of New York City, announced a proposed cordonpricing scheme for Manhattan. Figure 22 shows the proposed cordon area (the green rectangle is Central Park). Based on London's cordon scheme, the congestion zone would be in effect on weekdays between 6 am and 6 pm. Cars would be charged $8 daily, and trucks $21, to enter, leave, and move within the zone. Cars with handicapped license plates, taxis, emergency vehicles, and transit buses would be exempt (City of New York, 2007).

Page 49 of 66

Congestion Pricing Response: Section I

Figure 22: Proposed Cordon Pricing for Manhattan

Source: Office of the City of New York (2007)
The mayor's report predicted that congestion within the charged zone would decrease by 6.3% and traffic speeds would increase by 7.2%; moreover, it predicted, only 1.4% of travelers would refuse to enter the zone at all in order to avoid paying the fee (City of New York, 2007). Nonetheless, political opposition has run strong in New York's state legislature, which would have to approve any plan. The strongest opposition has come from representatives of suburban commuters. Charles J. Fuschillo, Jr., a New York state senator who represents parts of Nassau and Suffolk counties, told the New York Times, "It's just another version, in my opinion, to hit Long Island residents with a significant yearly fee" (Confessore, 2007). Although, in Bloomberg, the New York cordon-pricing project has the strong political advocate that many congestion-pricing projects have lacked in the past, the particular nature of New York state politics may delay or completely inhibit the implementation of a Manhattan cordon.
New York / New Jersey
E-ZPass is an electronic toll collection system which uses a transponder to record and deduct toll costs from prepaid accounts as drivers pass through the toll lanes. The E-ZPass program allows for one

Page 50 of 66

Congestion Pricing Response: Section I

E-ZPass account to be used on toll roads in the Northeast area (Maine, New Hampshire, Massachusetts, New York, New Jersey, Pennsylvania, Delaware, Maryland, West Virginia, and Virginia) and parts of the Midwest (Illinois and Indiana).
Figure 23: E-ZPass Operations

Source: E-ZPass New Jersey Customer Service (2007)

There are two types of accounts: individual and business. Individual accounts may have up to four tags per account and are for cars, vans, pickup trucks, motorcycles and RVs. The business account applies to tractor trailers, auto transporters, pickup/other trucks, buses, vans, cars, and motorcycles and may have an unlimited amount of tags per account. The E-ZPass also allows drivers to receive an automatic discount when using the E-ZPass lanes compared to general toll payment.

The New Jersey Turnpike (NJTPK) is a 148-mile toll road with 29 interchanges and is one of the most densely traveled roadways in the United States (700,000+ vehicles/day). There are 187 E-ZPass toll lanes operating on the turnpike. 92% of the NJTPK revenue is derived from tolls, 35% of which is from out-of-state traffic. The toll is determined by time-of-day pricing program which encourages peak-period commuters to alter their travel times to off-peak times to reduce congestion.

Page 51 of 66

Congestion Pricing Response: Section I

The E-ZPass program on the NJTPK was implemented in two stages. Stage 1 was implemented in 2000 and introduced the E-ZPass program's time-of-day pricing to the Turnpike. Stage 2 occurred in 2003 and increased the toll levels for each time period and each vehicle type (5% for E-ZPass off-peak, 10% for E-ZPass peak, and 17% for cash payers).
Despite the implementation of the toll program, there is a continual increasing trend in annual traffic, suggesting that time-of-day pricing does not have an impact on the increasing traffic congestion on the NJTPK. Also, the increase in tool price from Stage 1 to Stage 2 pricing and the differences between peak and off-peak periods were not substantial enough to have a statistically significant impact on NJTPK traffic (Ozbay, Yazman-Tuzel, and Holgun-Veras, 2006).
Outside the United States
London
Implemented in February 2003, the Central London Congestion Charging program serves the dual purpose of mitigating the congestion on the streets of London and generating additional revenue for the transit system. The congestion charging zone covers a 21-square kilometers area of inner London, and the charge (currently set at 8 [$16.17], up from 5 per day in the beginning), applies from 7 a.m. to 6:00 p.m. (originally 7 a.m.6:30 p.m.) on weekdays, excluding holidays (Transport for London, 2007). Taxis, alternative fuel vehicles, motorcycles, buses and emergency vehicles, among others, are exempt from the charge. Residents of the Congestion Charging zone only pay 10% of the full fee. For vehicles that are simply passing through, there are a number of routes that allow drivers to cross the zone during charging hours without paying (Litman, 2006).

Page 52 of 66

Congestion Pricing Response: Section I

Figure 24: Congestion Charging Zone (orange and purple areas), London, after western extension

Source: Transport for London (2007)
The geography of Central London was very well suited to pricing implementation. The street network at the core has not changed much since the medieval ages. Heavy travel demand resulting in severe congestion, and a wide variety of other transportation choices available (walking, taxi, bus and subway services) have created optimal conditions for adoption of congestion pricing. (Litman, 2006)
The system has experienced several changes since the implementation. The charge amount was increased from 5 (then $8.72) to 8 (then $13.95) in July of 2005 (Transport for London, 2006). On February 19, 2007, the Congestion Charging zone was extended to the west, and the charge hours were shortened by half an hour, to 7:00 a.m.6:00 p.m. The cost of implementing the western expansion was 118 million (Transport for London, 2006).

Page 53 of 66

Congestion Pricing Response: Section I

Figure 25: Congestion Charging Cameras, London

Source: Wiss (2006)
The technology enabling congestion charging is based on video imaging and license-plate recognition. More than 200 cameras are used to capture vehicle registration plates at the entry point and store the vehicle information in a database until it can be matched to a payment. Strict enforcement and monitoring are necessary to ensure the effectiveness of the program (Turner, 2003). Non-payment rates were high during the first few weeks, probably due to driver confusion and vehicle plate number recognition errors. It is of interest that the payment methods preferred by the drivers are technology-intensive: increasingly, drivers pay the congestion charges by mobile phone text message (Litman 2006). In the future, using RFID transponders for an automatic payment by some of the drivers is a possibility.
In addition to cellular text messaging, motorists can pay the Central London Congestion Zone charge through payment machines located in the area, over the Internet, and at select retail outlets. Weekly, monthly and annual passes with a 15% discount are also available. Motorists that have not paid the charge by the end of the next business day are assessed a 80 fine, reduced to 40 if paid within two weeks, and increased to 120 if not paid within a month (Litman, 2006).
According to Litman (2006), about one million people enter the Central London District every day. Prior to congestion pricing, 12% of them would use private cars for the trip. Within a few months of congestion pricing introduction, the number of private vehicle drivers declined to 10% of the total number of people entering the zone. Approximately 110,000 motorists a day pay the congestion charge (98,000 individual drivers and 12,000 fleet vehicles). After six months of operation, 60,000 cars fewer per day were entering the congestion charging zone, and 110,000 persons per day were paying the congestion charge (Litman, 2006). According to Turner (2003), travel time to, from and across the priced zone were down by 14%, while the time spent moving at below 10 kilometers per hour had decreased by 25%. The express bus

Page 54 of 66

Congestion Pricing Response: Section I

routes serving the congestion zone have decreased the waiting time by 33%, and revenues for 2003 and 2004 were projected at 68 million. The annual net benefits, apart from the toll revenues, were forecast to be 50 million, and included time savings, fuel savings and benefits from transport reliability. The mayor has directed toll revenues for investment in public transport (Turner, 2003).
As of March 2006, net revenues of 303 million have been generated, excluding the implementation costs of 162 million (covered from Transport for London's General Fund; see Table 5, below). A further 620m was projected to be raised over the next four financial years. The revenues are being directed towards bus network improvements; extending accessibility improvements; interchange improvements to aid the integration of the transport network; contributing to the costs of developing possible tram or segregated bus schemes; safety and security improvements; accelerating road and bridge maintenance programs; restructuring fares on public transport; increasing late night public transport; improvements to the walking and cycling environment and other specified goals (Transport for London, 2006).
Table 5: Net Proceeds from London Congestion Charge, April 2002March 2006

Source: Transport for London (2006)
Although very successful, the congestion pricing scheme has several drawbacks. The charge is not timevariable or location-variable. The transit system is somewhat crowded and requires further investment to support an additional influx of customers. The charge system has fairly high overhead costs. As indicated in Table 6, below, the annualized start-up and operating costs of running the Central London Congestion Charging program take up more than half of the revenue collected.

Page 55 of 66

Congestion Pricing Response: Section I

Table 6: Projected Costs and Revenues, London Congestion Charging Zone, 20002008

Source: Litman (2006)

Political considerations play an important role when dealing with a wide-reaching congestion-pricing scenario. In 2000, the political system of London was restructured to give the elected Mayor new powers to manage the city's transport system and raise taxes to fund transport improvements. Ken Livingstone, who won the election, ran with a platform that included congestion pricing implementation, with revenues to be used for public transit improvements (Litman, 2006). In the United States, it is not clear whether the constituency of major cities will be as receptive to congestion pricing as the citizens of London, although New York might be better suited to such a congestion pricing scheme than most.

Trondheim, Norway

Figure 26: Toll Ring in Trondheim, Norway

Page 56 of 66

Source: Amdal (2003) Congestion Pricing Response: Section I

Trondheim, Norway is a city of about 140,000 inhabitants, with about 60% of the population living in the center of the city. In 1991 the city officials implemented an electronically operated cordon pricing system. The tolls are slightly higher during the morning peak hours, and are not assessed after 5:00 p.m. and during the weekends. There are no monthly passes, and the motorists are charged a toll for each entry (except when making repeated trips within the same hour, or after 75 crossings per month for cars with a special tag). The regular toll is 12 Norwegian Kroner (NKr), or about $2.07, which is 10% of the average hourly earnings for a Norwegian blue-collar worker. Heavy vehicles (over 7700 lbs, or 3500 kg) are charged double price. With the implementation of the Toll Ring, there was a 10% decrease in traffic both during the peak and non-peak charge period. Traffic increase in the evenings and on weekends was a little over 8% (Langmyhr 1999).
The Trondheim Toll Ring Project was well marketed before the opening. Today about 95% of the motorists entering the city center use the electronic payment system. The current revenues, around 150 million NKr annually, are used to finance new roads, improved public transit and new pedestrian and bicycling facilities. In the first year after opening, weekday bus travel increased by 7%.
In 1998 the system was adjusted to cover more traffic in the urban area. The city was divided into six sectors, and vehicles crossing from one sector into another were required to pay toll. As a result, the traffic situation in the city center became significantly less congested than it had been ten years earlier. More recently, the Toll Ring was expanded again, with additions including six new charging points and an increase in the base price. The current system is estimated to produce about 200 million NKr ($34 million) per year of toll revenue, with operating costs of 17 million NKr per year (less than 10% of revenue). Resulting inflow of revenue was sufficient to finance the latest round of investments in Trondheim's surface transportation infrastructure in 2005 (Amdal, 2003).
In 2001, Trondheim introduced a new toll charging technology called AutoPASS, with the goal of ensuring interoperability between the toll system in Trondheim and other Norwegian cities. Norway is supporting AutoPASS as a basis for standardization in Europe for electronic fee collection systems. As of 2003, drivers in South Trondelag County are able to use a common payment card, called the t:kort, for almost all the services in public transport (Porter, Kim et al., 2004). Overall, the system has been a success, although continuous challenges and improvements are to be expected. The issue of interoperability in toll paying technology between various jurisdictions is becoming more apparent in Europe, as it likely soon will in the U.S.

Page 57 of 66

Congestion Pricing Response: Section I

407 ETR, Toronto

Figure 27: 407 ETR Open Road Tolling System

Source: Road Traffic Technology (2007)
Toronto's 407 Express Toll Route (ETR) is the world's first "open-road" (no cash tollbooths), all-electronic highway. Developed by Raytheon Systems Company and in operation since 1997, this electronic tolling system incorporates infrared and video technology to monitor vehicle usage and automatic toll collection.
407 ETR stretches over 67 miles (108 km) across the north side of the Greater Toronto area. The project was built in stages, with the first 36 km opened in 1997, and the final section completed in 2001 (Road Traffic Technology, 2007).
The toll route is equipped to collect tolls from transponder-equipped vehicles, as well as cash customers, without using toll plazas. Overhead tolling gantries record transponder-equipped vehicles as they enter and exit the facility (see Figure 27). Tolls vary by vehicle class and distance traveled. Heavy vehicles, over 5 tons (4500 kg) are required to carry a transponder that charges a special heavy vehicle rate (407 Express Toll Route) .
Close to 30% of the vehicles using the facility are not equipped with transponders. The license plate numbers of those vehicles are recorded electronically, and a bill is sent to the owner in the mail (such a toll is called a V-Toll). An additional charge applies for V-Tolls. Thanks to the agreement with neighboring Canadian provinces and some of the states in the U.S., drivers outside of the region can receive a bill in

Page 58 of 66

Congestion Pricing Response: Section I

the mail. Local vehicle owners with outstanding accounts who fail to pay their bill have their information sent to the Registrar of Motor Vehicles, where renewal of vehicle registration can be denied until all tolls are paid (Cothron, Skowronek et al., 2003).
A License Plate Recognition system identifies about 80% of the vehicles correctly. Digital images of the remaining 20 percent are checked by human eyes. About 6% remain unbilled as a result of an inability to read the license plate number, or the Ontario government's not having an extradition agreement with the vehicle owner's home state (Cothron, Skowronek et al., 2003). In 2005, over 100 million trips were processed with an accuracy rate of 99.9% (407 Express Toll Route).
I-407 has dedicated safety vehicles that patrol it seven days a week. In addition, the Ontario Provincial Police (OPP) and Ministry of Transportation enforcement officers also patrol I-407. While toll collection and enforcement is an automated process, traffic offenses such as speeding are enforced by OPP (Cothron, Skowronek et al. 2003).
Even though this system has been quite successful, it remains to be seen whether the currently possible accuracy rate is acceptable for implementation of future similar projects in the U.S. (such as on Toll Highway 101 in Tampa, Florida).
Conclusions
Bedeviled by congestion, cities around the world are increasingly experimenting with tolling via newly introduced technologies, to increase the cost of entering or traversing the city via car. This has given rise to the congestion-charging cordons in Tronheim, Singapore, and London, which have in turn inspired the proposed scheme in Manhattan. The Utah, E-407, and EZPASS systems, by contrast, are aimed at facilitating traffic flows on highways. This gives some idea of how congestion pricing can be used by different agencies to address different aims.
Of the six different congestion-pricing projects featured in this section, only two use variable pricing. This may be because variable pricing introduces a new element of potential confusion and uncertainty to users, especially users not previously familiar with toll facilities or electronic payment. In the long run, it might be easier to begin a congestion-pricing program with a set toll and later, if desired, convert it to a variable toll.

Page 59 of 66

Congestion Pricing Response: Section I

Conclusions
It is only in the last few years that advances in technology have made congestion pricing, especially variable pricing, possible. The increased interest in congestion pricing means that further refinements in existing technology may be available to new facilities opened within the next decade. There are now more options than ever for creating, managing, operating, and refining a congestion-pricing project. The five case studies, which opened between 1995 and 2006, are all still in operation, and while they have not escaped criticism, they have generated positive feedback, especially in the case of the MnPASS and I-15 Express Lanes facilities.

This is not to suggest that implementing a congestion-pricing project can or will be easy for other state departments of transportation. The public remains skeptical of congestion pricing. Moreover, the less exposure a person has had to electronic tolling mechanisms, congestion pricing, or variable pricing, the more likely he or she is to dismiss it as a form of congestion mitigation or revenue generation; and for all the advances of the five case studies, those facilities have been use by only a tiny fraction of the American driving population. A belief that the freeway capacity is "already paid for" will increase hostility to the proposed project unless the proposing body has a clear purpose for the generated revenue. Even then, as King, Melville and Shoup (2007) have pointed out, some members of the public will have more costs than benefits. Thus supporters of a new congestion-pricing project should be prepared for political opposition.

Finally, the concern that variably-priced lanes will allow the wealthy to buy their way out of congestion, further contributing to inequitable access to smoothly-flowing traffic, cannot be dismissed as resulting from public ignorance. The majority of equity-concerned surveys have suggested that managed lanes are used by both high- and low-income users, although the Houston QuickRide survey data (Burris and Hannay, 2003) suggests that users might have, on average, higher incomes than non-users. There may be a difference in access to transponders, or in trust in the enforcement system, between lower- and higher-income users.

The successful congestion-pricing projects featured in this review have been shaped over time by changes in policy and strategy. The SR-91 Express Lanes suffered a severe setback when public opinion turned against the private operator, and the MnPASS system was only implemented after several previous political failures. The I-15 Express Lanes and London congestion-pricing cordon may never have been implemented without the strong backing of key political figures (Jan Goldsmith and Ken Livingstone, respectively). In short, in order to overcome potential opposition and implement a congestion-pricing project that could ultimately benefit the region it serves, the backing body must be flexible, politically astute, able to communicate well with the public and future users, clear in its goals for the facility, open to new advances in technology, and confident in congestion pricing as a useful tool.

Page 60 of 66

Congestion Pricing Response: Section I

References
407 Express Toll Route. (2007, April 17). "Frequently Asked Questions." Retrieved April 17, 2007, from http://www.407etr.com/About/qas.htm#q5.
Amdal, E. (2003, November 19-22). Lord of the Rings: Trondheim, Norway. Paper presented at the International Perspectives on Road Pricing, Key Biscayne, Florida.
Benjamin, J. M., Sakano, R., McKinney, B., Khattak, A. J., Rodriguez, D. A., and Gaskins, C. (2007). An Analysis of HOT Lanes in North Carolina. Paper presented at the Transportation Research Board 86th Annual Meeting.
Berman, W. (2007, April 4, 2007). "Presentation 5: Public and Political Outreach. Involving the Public and Creating Political Champions Case Study. ." April 4, 2007-Congestion Pricing and Urban Partnerships Webcast. Retrieved August 1, 2007, from www.ntoctalks.com/webcast_archive/to_apr_04_07/to_apr_04_07_part2.ppt
Brownstone, D., Ghosh, A., Golob, T. F., Kazimi, C., and Van Amelsfort, D. (2003). Drivers' willingnessto-pay to reduce travel time: evidence from the San Diego I-15 congestion pricing project. Transportation Research Part A: Policy and Practice, 37(4), 373-387.
Burris, M. W., and Hannay, R. L. (2003). Equity Analysis of the Houston QuickRide Project. Paper presented at the Transportation Research Board 82nd Annual Meeting.
Burris, M. W., Sadabadi, K. F., Mattingly, S. P., Mahlawat, M., Li, J., Rasmidatta, I., et al. (2007). Reaction to the Managed Lane Concept by Various Groups of Travelers. Paper presented at the Transportation Research Board 86th Annual Meeting.
Burris, M. W., and Stockton, B. R. (2004). HOT Lanes in Houston: Six Years of Experience. Journal of Public Transportation, 7(3).
Burris, M. W., and Xu, L. (2006). Potential Single-Occupancy Vehicle Demand for High-Occupancy Vehicle Lanes: Results from Stated-Preference Survey of Travelers in High-Occupancy Toll Corridors. Transportation Research Record(1960), 108-118.
Cada, C. (2007, May 1). "No Free Roads": Colorado pushing for more tolls to raise revenue and ease congestion. Colorado Construction, 10.
Cambridge Systematics Inc., Short-Elliott-Hendrickson Inc., et al. (2006). I-394 MnPASS Technical Evaluation Final Report. Minneapolis, MN, Minnesota DOT.
Catlin, Peggy. Testimony before House Committee on Transportation and Infrastructure, Subcommittee on Highways and Transit, United States Congress, June 7, 2007.
Collier, T., and Goodin, G. D. (2002). The Funding and Financing of Managed Lanes Projects (Report No. FHWA/TX-03/4160-9). College Station, Texas: Texas Transportation Institute.
Colorado DOT (2007). I-25 Express Lanes Monthly Progress Report March 2007. Monthly Express Lane Reports

Page 61 of 66

Congestion Pricing Response: Section I

Colorado DOT. (2007, May 18, 2007). "About the I-25 HOV/Tolled Express Lanes." Retrieved May 18, 2007, from http://www.dot.state.co.us/cte/expresslanes/about.cfm.
Colorado DOT. (2007, August 7, 2007). "I-25 HOV/Tolled Express Lanes Map." Retrieved August 7, 2007, from http://www.dot.state.co.us/cte/expresslanes/maps.cfm.
Cook Research and Consulting Inc. (2004). MnPASS Focus Groups. Confessore, N. (2007, July 27). Governor Signs Bill to Put Congestion Pricing Plan on Track. New York
Times. Cothron, A. S., D. A. Skowronek, et al. (2003). Enforcement Issues on Managed Lanes, Texas
Transportation Institute, Texas A and M University System; Available through the National Technical Information Service. DeCorla-Souza, P. (2005). A New Financing Approach for Transportation Infrastructure Expansion, TRB Paper No. 06-1300. Douma, F., Zmud, J., and Patterson, T. (2006). Pricing Comes to Minnesota: Baseline Attitudinal Evaluation of the I-394 HOT Lane Project. Paper presented at the Transportation Research Board 86th Annual Meeting. Engineering News Record. (2003). Variably Priced Express Lanes Gaining Interest Nationwide. Engineering News Record, 250, 17. March 17. Evans, A., Gougherty, M., Morris, E., and Smirti, M. (2007). Politics, Public Opinion, and Project Design in California Road Pricing. Paper presented at the Transportation Research Board 86th Annual Meeting. E-ZPass New Jersey Customer Service. (2007). E-ZPass Customer Reference Guide. E-ZPass. Retrieved May 17, 2007, from http://www.e-zpassny.com/static/info/index.shtml#. FHWA (2003). A Guide for HOT Lane Development, Chapter 4. Washington, D.C. Retrieved March 4, 2007 from http://www.its.dot.gov/JPODOCS/REPTS_TE/13668_files/chapter_4.htm FHWA (2003). A Guide for HOT Lane Development, Chapter 7. Washington, D.C. Retrieved March 4, 2007 from http://www.itsdocs.fhwa.dot.gov/jpodocs/repts_te/13668_files/chapter_7.htm FHWA (2006). Value Pricing Quarterly Report October-December 2006. FHWA. (2007). "Converting High-Occupancy Vehicle (HOV) Lanes to High-Occupancy Toll (HOT) Lanes." Value Pricing Pilot Program Retrieved March 12, 2007, from http://ops.fhwa.dot.gov/tolling_pricing/value_pricing/projtypes/hovhotlanes.htm. FHWA and FTA (2006). Peer Exchange Report -- Value Pricing: Managed Lanes on Interstate 15 in San Diego County. The Transportation Planning Capacity Building Program. Florida Department of Transportation. Florida's High Occupancy Vehicle Lanes. Retrieved August 1, 2007, from http://www.dot.state.fl.us/TrafficOperations/OtherLinks/HOV/HOV.htm Grling, T., and Schuitema, G. (2007). Travel Demand Management Targeting Reduced Private Car Use: Effectiveness, Public Acceptability and Political Feasibility. Journal of Social Issues, 63(1), 139153.

Page 62 of 66

Congestion Pricing Response: Section I

Godbe Research and Analysis. (1998). Focus Group Research for I-15 Commuters. San Diego, California.
Gulipalli, P., and Kockelman, K. M. (2006). Credit-Based Congestion Pricing: A Dallas-Fort Worth Application. Paper presented at the Transportation Research Board 85th Annual Meeting.
Halvorson, R. and K. R. Buckeye (2006). "High-Occupancy Toll Lane Innovations: I-394 MnPASS." Public Works Management and Policy 10(3): 242.
Hultgren, L., and Kawada, K. (1999). San Diego's interstate 15 high-occupancy / toll lane facility using value pricing. Institute of Transportation Engineers. ITE Journal, 69(6), 22.
IBI Group and Cambridge Systematics. (2006). "Background Paper #8: Toll Technology Considerations, Opportunities, and Risks." Washington State Comprehensive Tolling Study Final Report-Volume 2 Retrieved May 25, 2007, from http://www.wstc.wa.gov/Tolling/FR1_WS_TollStudy_Vol2_Paper08.pdf.
King, D., Manville, M., and Shoup, D. (2007). The political calculus of congestion pricing. Transport Policy, 14(2), 111-123.
Kuhn, B., Goodin, G. D., Ballard, A., Brewer, M., Brydia, R., Carson, J., et al. (2005). Findings from Texas: Five Years of Managed Lanes. College Station, Texas: Texas Transportation Institute, Texas Department of Transportation, and Federal Highway Administration.
Kockelman, K. M., and Kalmanje, S. (2005). Credit-based congestion pricing: a policy proposal and the public's response. Transportation Research Part A: Policy and Practice, 39(7-9), 671-690.
Langmyhr, T. (1999). "Understanding innovation: the case of road pricing." Transport Reviews 19(3): 255271.
Leahy, M. (2005). Operating The Toll Roads: A National Perspective. San Joaquin Hills TCA, Foothill/Eastern TCA. Retrieved June 1, 2007 from http://www.thetollroads.com/home/images/innovation_perspective.pdf.
Litman, T. (2006). London Congestion Pricing: Implications for Other Cities. Victoria, British Columbia, Canada: Victoria Transport Policy Institute.
Machalaba, D. (2007, June 21). Paying for VIP Treatment in a Traffic Jam. Wall Street Journal. McCombs, B. (2005, June 11). I-25 panel narrows focus. Greeley Tribune. Menon, G. (2005). Evaluation of Singapore's Electronic Road Pricing System. TRB Conference
Proceedings 34: International Perspectives on Road Pricing. Menon, G. (2006). Congestion Pricing: The Singapore Experience. Street Smart: Competition,
Entrepreneurship, and the Future of Roads Meyer, M., Saben, L., Shephard, W., and Drake, D. (2006). Feasibility Assessment of Metropolitan High-
Occupancy Toll Lane Network in Atlanta, Georgia. Transportation Research Record, 1959, 159167. Minnesota Department of Transportation. (2005). About MnPASS. Retrieved July 31, 2007, from http://www.mnpass.org/

Page 63 of 66

Congestion Pricing Response: Section I

Minnesota DOT. (2007a). "MnPASS Express Newsletter. January 2007" Retrieved August 7, 2007, from http://www.mnpass.org/ExpressNewsletter/MnPASSExpressJanuary07.pdf.
Minnesota DOT. (2007b, May 18, 2007). "I-394 MnPASS Background." Retrieved May 18, 2007, from http://www.mnpass.org/394/index.html.
Munnich Jr., L. W., and Buckeye, K. R. (2007). I-394 MnPASS High-Occupancy Toll Lanes: Planning and Operational Issues and Outcomes. Paper presented at the Transportation Research Board 86th Annual Meeting.
Munnich Jr, L. W., and Loveland, J. D. (2005). Value Pricing and Public Outreach: Minnesota's Lessons Learned. Transportation Research Record, 1932, 164-168.
NuStats (2006). Attitudinal Panel Survey Wave 3: Final Report. MnPASS Evaluation. Minneapolis, MN, Humphrey Institue of Public Affairs, University of Minnesota. Retrieved June 4, 2007 from http://www.mnpass.org/pdfs/MnPASSFinalReport%2027NOV06.pdf
G. J. Roth (Ed.). Oakland, California, The Independent Institute. Obenberg, J., P.E. (2004). "Managed Lanes." Public Roads 68(3). Odeck, J., and Brthen, S. (1998). The Planning of Toll Roads--Do Public Attitudes Matter?
Transportation Research Record, 1649, 72-80. Office of the Mayor of the City of New York. (2007). PlaNYC 2030: A Greener, Greater New York. New
York, New York. Olshan, J. (2007). "Bell May Toll for Booths on Hudson". New York Post. New York, NY. May 16, 2007. Orange County Transportation Authority (2006). The Model of Success: 91 Express Lanes Fiscal Year
2006 Annual Report. Retrieved August 4, 2007 from http://www.91expresslanes.com/generalinfo/91annualreport.pdf Orange County Transportation Authority. (2007, July 1). 91 Express Lanes -- Toll Schedules. Retrieved August 3, 2007, from http://www.91expresslanes.com/tollschedules.asp# Ozbay, K., Yanmaz-Tuzel, O., and Holgun-Veras, J. (2006). Evaluation of Combined Traffic Impacts of Time-of-Day Pricing Program and E-ZPass Usage on New Jersey Turnpike. Transportation Research Record, 1960, 40-47. Podgorksi, K., and Kockelman, K. M. (2005). Public Perceptions of Toll Roads: A Survey of the Texas Perspective. Paper presented at the Transportation Research Board 84th Annual Meeting. Poole Jr, R. W. and C. K. Orski (2000). "HOT Lanes: A Better Way to Attack Urban Highway Congestion." Regulation 23(1): 15-20. Poole Jr., R.W. (2005) Automated HOV Enforcement? Don't Hold Your Breath. Surface Transportation Innovations April 2005 (22). Retrieved May 4, 2007 from http://www.reason.org/surfacetransportation22.shtml. Porter, J. D., D. S. Kim, et al. (2004). "Electronic Road Pricing Systems: Capabilities, and Existing and Needed Technology Solutions." 83rd Annual Meeting of the Transportation Research Board, Washington, DC.

Page 64 of 66

Congestion Pricing Response: Section I

Pratt, R. H., K. F. Turnbull, et al. (2000). Traveler Response to Transportation System Changes: Interim Handbook. TCRP Web Document 12, Transit Cooperative Research Program.
Road Traffic Technology. (2007, June 27). "407 Express Toll Route (ETR), Ontario." Retrieved July 11, 2007, from http://www.roadtraffic-technology.com/projects/407/.
SANDAG. (2003, August 18). "I-15 FasTrak Online: Photo Album." Retrieved March 12, 2007, from http://fastrak.sandag.org/photos.html.
Schade, J., and Schlag, B. (2003). Acceptability of urban transport pricing strategies. Transportation Research Part F: Traffic Psychology and Behaviour, 6(1), 45-61.
Schijns, S. and McCormick Rankin Corp. (2005). Automated Occupancy Monitoring Systems for HOV/HOT Monitoring and Enforcement. In 12th International HOV Systems Conference: Improving Mobility and Accessibility with Managed Lanes, Pricing, and BRT. FHWA, Office of Operations.
Smith, L., and Benko, M. (2007, June 20). Services and Technologies: Electronic Toll Collection. Retrieved August 3, 2007, from http://www.calccit.org/itsdecision/serv_and_tech/Electronic_toll_collection/electron_toll_collection _report.html
Stegman, S. (2007, February). Tolling for lane privileges. The American City and County, 122, 42. Sullivan, E. (2000). Continuation Study to Evaluate the Impacts of the SR-91 Value-Priced Express
Lanes: Final Report. Sacramento, California DOT Traffic Operations Program HOV Systems. Sullivan, E. C. (2003). Implementing Value Pricing for U.S. Roadways. European Journal of Transport
and Infrastructure Research, 3(4), 401-413. Sullivan, E., and Burris, M. (2006). Benefit-Cost Analysis of Variable Pricing Projects: SR-91 Express
Lanes. Journal of Transportation Engineering, 132(3), 191-198. Swisher, M., and Ungemah, D. (2006). So You Want to Make a HOT Lane? The Project Manager's
Guide for an HOV-to-HOT Conversion. Paper presented at the Transportation Research Board 85th Annual Meeting. Ungemah, D. (2007). This Land Is Your Land, This Land Is My Land: Addressing Equity and Fairness in Tolling and Pricing. Paper presented at the Transportation Research Board 86th Annual Meeting. Texas Transportation Institute. (2003a). Houston Value Pricing Pilot Program. Texas Transportation Institute. (2003b). QuickRide: About QuickRide. Retrieved July 31, 2007, from http://www.quickride.org/about_quickride.stm Texas Transportation Institute. (2005). Results of San Antonio Focus Groups: Public Acceptability of Express Lane Options for I-35. Transport for London. (2006). Four Year Programme. London, United Kingdom: Mayor of London. Transport for London. (2007). Congestion charging... where and when does it operate. Retrieved May 11, 2007, from http://www.cclondon.com/infosearch/dynamicPages/WF_ZoneCheck_W.aspx

Page 65 of 66

Congestion Pricing Response: Section I

Turnbull, K. F. (2003). Houston Managed Lanes Case Study: The Evolution of the Houston HOV System. College Station, TX, Texas Transportation Institute.
Turner, D. (2003, November 19-22). Central London's Congestion Charging Scheme: Has It Achieved Its Objectives? Paper presented at the International Perspectives on Road Pricing, Key Biscayne, Florida.
Turner, S., Video Enforcement of High-Occupancy Vehicle Lanes: Field Test Results for I-30 in Dallas. In Transportation Research Record: Journal of the Transportation Research Board, No. 1682. TRB, National Research Council, Washington, D.C., 1999 (pp. 2837).
Ubbels, B., and Verhoef, E. (2006). Behavioural Responses to Road Pricing: Empirical results from a survey among Dutch car owners. Paper presented at the Transportation Research Board 85th annual meeting.
Ungemah, D., and Collier, T. (2007). I'll Tell You What I Think! A National Review of How the Public Perceives Pricing. Paper presented at the Transportation Research Board 86th Annual Meeting.
Utah DOT (2006). "Express Lanes Frequently Asked Questions." Retrieved June 13, 2007, from https://secure.utah.gov/expresslanes/action/public/faq.
Vehicle Occupancy Limited (VOL) (2007). DTECT: Product Outline Retrieved May 3, 2007 from .http://www.vehicleoccupancy.com/pages/prod_outline.html. Accessed May 3, 2007.
Vollmer Associates. (2007). 91 Express Lanes Traffic and Revenue Update. Memorandum prepared for Orange County Transportation Authority.
Warburton, N. (2006). Utah commission OKs stretch of I-15 toll lanes. Deseret Morning News. Salt Lake City, UT. Saturday, April 22, 2006. Retrieved May 14, 2007 from http://deseretnews.com/dn/view/0,1249,635201642,00.html
Weinstein, A., and Dill, J. (2007). How to Pay for Transportation? A Survey of Public Preferences. Paper presented at the Transportation Research Board 86th Annual Meeting.
Wiss, D. (Artist). (2006). Photo: 83 of 88, Album: England 2006 [Photograph].

Page 66 of 66

Congestion Pricing Response: Section I