Value Pricing Pilot Program: Lessons Learned – Appendix B

3.0 Variable Pricing on Existing Toll Facilities

Table 3.1 Pricing on Bridges and Variable Tolls for Heavy Vehicles in Lee County, Florida

Pricing on Toll Facilities	Pricing on Bridges and Variable Tolls for Heavy Vehicles in Lee County, Florida
Operations	In August 1998, Lee County, Florida, in cooperation with the Florida DOT and FHWA, implemented value pricing on two toll bridges – the Cape Coral and Midpoint Memorial Bridges. Under terms of the program, bridge users are offered a toll discount during selected off-peak hours as an incentive to change trip times from peak to off-peak (less congested) hours. The discount program offered a 50‑percent toll discount for two-axle vehicles traveling during times either 30 minutes before or two hours after the 7:00-9:00 a.m. peak traffic period and two hours before or 30 minutes after the 4:00-6:30 p.m. peak traffic period. This amounted to a $0.50 toll discount for users paying the regular toll of $1. For bridge users who took advantage of the option of paying a $40 annual fee to obtain a bridge toll rate of $0.50, the variable pricing option provided a $0.25 saving for travel during the discount periods. The toll discount is limited to bridge users who pay using an electronic transponder and the discount program applies only to weekday traffic. Of the drivers who obtained a variable toll discount, approximately 94 percent received a $0.25 discount and 6 percent received a $0.50 discount. Approximately 25 percent of all bridge users participated in the variable toll discount program. Approximately 2.2 million vehicle trips took advantage of the toll discount program from October 2005 through March 2006, nearly a 25 percent increase over the same period a year earlier (URS Report). Effective November 1, 2007, the toll structure on these bridges was changed to a one-way toll (for a one-year trial period), with tolls for two-axle vehicles raised to $2 but collected only in the westbound direction (heading into Cape Coral). The shoulder peak discount is $1, continuing a 50 percent toll discount. The Leeway tolling system is operated by the Lee County Department of Transportation which is overseen by a 5-member Board of County Commissioners. The Heavy Vehicle program extends the off-peak toll discount program available to light duty vehicles to vehicles with 3‑or‑more axles. The program became operational in December 2003. Vehicles with 3‑or‑more axles pay a discounted toll when they use the Cape Coral or Midpoint Memorial Bridges during variable pricing hours. The discount was initially 50 percent. The one-way toll rates established in November 2007 continue to provide a 25 percent discount for travel during the shoulder of the peak.
Cost, Finance, and Revenue	Lee County received FHWA VPPP funding of $604,000 in FY 2000, $428,000 in FY 2001, and $200,000 in 2005. The shoulder-of-the-peak discount program does not generate positive cash flow. (TCRP Report 95) However, the intent of the program was to spread traffic from peak to shoulder times, thereby postponing by years the need for expensive bridge capacity enhancements in this rapidly growing area.
Policy/Institutional	The discount program was designed to make better use of existing bridge capacity. The Lee County Board of Commissioners has the administrative authority to implement the projects. The Board decided to continue the value pricing program on these bridges after completion of FHWA’s initial funding support, including the expansion to heavy vehicles.
Outreach/Acceptance	Lee County engaged stakeholders by establishing three citizen advisory committees consisting of local bridge users and businesses to assist in project planning, design and implementation (Ward). The focus group technique was used extensively to gauge citizen reaction to introduction of the program. Focus groups were held in March and July 1996. Initial focus group reaction was mixed to negative. Persons with limited time availability were most interested, while retirees and students expressed the least interest. Resistance to automated debiting of checking or credit card accounts was expressed, although the idea of replacing the existing bar code scanning with the use of electronic tolling was universally accepted. The idea of variable tolling by increasing tolls during peak hours was rejected as penalizing those with inflexible schedules, and violating a promise by the County Commission not to raise tolls after an increase in 1994. The idea of introducing the program with off-peak discounts was accepted. Participants indicated that toll reductions of at least 50 percent would be needed to entice people to shift travel to the shoulder of the peak. Marketing techniques included paid radio and print media advertisements, advertising billboards, point-of-sale displays, newsletters, press releases and media kits, presentations to community groups and employers, and a web site. Newspaper articles and advertisements were found to be the most effective means of communicating the program. Direct mailings and newsletters also were found to be effective. Extensive advertising campaigns provided information about the variable toll program. A 1999 telephone survey indicated that 90 percent of travelers in Lee County were familiar with the toll discount program (OPS/FHWA). Marketing efforts for the extension of variable pricing to heavy vehicles began prior to implementation of pricing discounts and increased as the implementation date (December 2003) approached. Efforts focused on informing heavy vehicle owners/operators and fleet managers of the program. Public awareness campaign included direct mail distribution of: customer letters, a program newsletter, a self-mailing brochure and application, and press releases to the media. Posters were placed near the bridge lanes. After the LeeWay tolling system was installed, an additional public awareness program was launched, including direct mailings, newspaper advertising, posters and press releases.
Technology	To implement the variable toll discount program, Lee County switched from the use of optical scanning labels to a transponder (LeeWay) system with prepaid toll accounts. Amtech was chosen to supply the Radio Frequency Identification (RFID) system with toll plaza readers and vehicle transponders using “off the shelf” Amtech technology. This would allow future interoperability with the Statewide SunPass system used on other Florida facilities (this occurred in 2004) (URS Report). As of February 2005, heavy vehicles (vehicles with 3‑or‑more axles) were required to use the attended lanes on all bridges so attendant could manually indicate the number of axles when the toll transaction was processed. Axle counting and vehicle separation equipment was being installed at the Midpoint Bridge that would allow these vehicles to use electronic tolling lanes. The Cape Coral Bridge is scheduled for rebuilding in the near future at which point axle counting and vehicle separation equipment will be installed so vehicles with 3-or-more axles could be allowed in all lanes. As of January 4, 2005, 653 transponders had been issued to three-or-more axle vehicles.
Equity/Environmental	Females were more likely than males to take advantage of the discount program (Center for Urban Transportation Research, University of South Florida cited in TCRP Report 95). The percentage of eligible users not changing their travel time to take advantage of the toll discount was highest for travelers aged 35-54. (Burris, Pendyala and Swenson, 2002 cited in TCRP Report 95). Both of these findings may be associated with the finding that users making work trips were less likely to take advantage of the toll discount program than were users making nonwork trips. Since the value pricing program had no significant impact on vehicle speeds, queue lengths at toll plazas, average vehicle occupancy, transit ridership, or accidents, there is no significant environmental impact. Equity concerns were not expressed in either surveys or focus group conducted for this project. The heavy vehicle program is not likely to have a significant effect on environmental quality since vehicles with 3 or more axles represent less than one percent of total bridge traffic at each toll facility.
Impacts	Changes in traffic patterns show clearly that drivers alter their travel behavior in response to variable toll pricing, with drivers having more discretion in travel times more likely to change travel behavior. Approximately 38 percent of eligible participants indicate that the variable pricing program had caused them to alter their tripmaking behavior since the program began(TCRP Report 95). A 1999 survey indicated that half of respondents either “always” or “sometimes” considered the discounts when planning a trip across the bridges. Surveys indicate that of those travelers who altered their time of travel, 71 percent did so at least once per week. However, that was only 32.4 percent of bridge travelers. So, for total bridge travelers, its 23 percent use it at least once per week (Burris, 2001). Survey respondents’ primary reason for obtaining a LeeWay account was to save time (43.7 percent), followed by obtaining variable pricing toll discounts (27.5 percent). Greatest relative traffic shifts occurred during the pre-morning peak, with the elasticity estimate (percent change in traffic/percent change in toll cost) at the Midpoint Bridge being in the upper ranges of demand elasticities found in the literature. Elasticity estimates for other periods are much lower. Elasticity estimates range between -0.36 and -0.03 depending on the time of day. Alternative calculations using log arc elasticities show ranges from-0.02 to -0.24 (TCRP Report 95). There was no detectible change in vehicle speeds, queue lengths at toll plazas, average vehicle occupancy, transit ridership, or accidents. There was tremendous growth in traffic at the bridges. Therefore, no change in peak-period queuing should be viewed as a success. The objective of making better use of existing capacity, thereby postponing the need for additions to capacity, was successfully achieved. The initial intent of the heavy vehicle project was to reduce queue waiting times (heavy vehicles “can be a substantial portion of the queue”). However, since the technology was not initially available to allow heavy vehicle access to the more efficient automated lanes, queues were not reduced during the period of the evaluation study. “In fact, the overwhelming majority of changes in queuing were increases.” (Expansion of Variable Pricing). Traffic patterns on the Cape Coral Bridge reflect a reduction in eastbound AVI trips during the a.m. peak (2003 to 2004). For instance, trips at the 8:00 a.m. peak decreased by roughly 25 percent. Cash trips were nearly identical in the two periods, leading study analysts to conclude that variable pricing led to some reduction in peak trips by three-or-more axle vehicles. There was a similar percentage decrease in peak AVI traffic on the Midpoint bridge, although the relationships were not as pronounced. Similarly, comparison of westbound traffic on both bridges showed a similar pattern though not enough to suggest a definitive impact of variable pricing. In response to the survey question, “do you adjust the routes or timing of trips due to variable pricing, 50 percent of respondents indicated they do not use the ETC system (so chose “not applicable”). Of the remaining 50 percent, 4 percent indicated they frequently adjust trips and 15 percent indicated they adjust trip “occasionally.” Study analysts conclude these responses tend to confirm the modest response to variable pricing by three-or-more axle vehicles. An additional survey was conducted in February 2005 to determine two-axle vehicle bridge users’ response to the expansion of the variable pricing program to heavy vehicles. In this survey, drivers who indicated they usually travel outside the peak period were asked the reason for usually traveling outside rush hour. Just over 10 percent of respondents indicated they shifted travel due to variable pricing (this was before the variable pricing program was discussed). After an explanation of the variable pricing program, 18 percent of respondents indicated they do, at least occasionally change their trip time to take advantage of the off-peak discounts.
Evaluation	The Center for Urban Transportation Research at the University of South Florida conducted the project evaluation for the Lee County Department of Transportation. In most cases data collected before implementation (August 1, 1998) were compared with data collected after variable pricing went into effect to determine if differences were statistically significant (at the 5 percent level). Sources of evaluation data included traffic volume counts, speed measurements, toll plaza queue lengths, travel times, bridge user surveys, focus groups (two in 1996-before implementation, one in 1999-after implementation), and random telephone surveys of bridge users, including those eligible for the bridge toll discounts (i.e., participants in electronic toll collection) and users who were not eligible for the discount program (did not participate in electronic toll collection). Data on vehicle occupancy and transit ridership also were collected. For the heavy vehicle program evaluation, the data base for evaluation of traffic pattern impacts included all bridge trips taken in the year before and the year after implementation of variable tolling for heavy vehicles (data collected for toll collection purposes-both cash and AVI).

Sources:
Burris, M.W., Pendyala, R.M., and Swenson, C., Discrete Choice Models of Traveler Participation in Differential Time of Day Pricing Programs, Transportation Research Board, Washington, D.C., 2002.
Burris, M. W., Lee County Variable Pricing Project: Evaluation Report, Tampa: Center for Urban Transportation Research – College of Engineering, University of South Florida, January 2001.
Burris, M.W. and Swenson, C., Planning Lee County’s Variable Pricing Program, Transportation Research Board, January 1998.
Burris, Mark W., Cain, Alasdair, and Swenson, Chris, Toll Price-Traffic Demand Elasticity Analysis on Variable Priced Toll Bridges, Tampa: Center for Urban Transportation Research – College of Engineering, University of South Florida, July 1999.
Burris, Mark W., Application of Variable Tolls on Congested Toll Road, Journal of Transportation Engineering, ASCE/July/August 2003.
Lee County, Florida, www.leewayinfo.com.
Lee County, Florida Department of Transportation, Expansion of Variable Pricing to Heavy Vehicles: Final Report, CRSPE, Inc, and Cella Assoc., February 2005.
Lee County, Florida Metropolitan Planning Organization, 2030 Long-Range Transportation Plan, December 2007, page E-11.
Office of Operations, Federal Highway Administration, Mitigating Traffic Congestion – The Role of Demand-Side Strategies: Lee County Variable Bridge Tolls (ops.fhwa.dot.gov/publications/mitig_traf_cong/lee_case.htm).
URS Corporation, Traffic Engineer’s Semi-Annual Report on the Lee County Toll Facilities for the Period ended March 31, 2006.
Ward, Jennifer, Value Pricing: A Synthesis of Lessons Learned, Hubert Humphrey Institute of Public Affairs, University of Minnesota, Transportation Research Board, 2002.

Table 3.2 Illinois Tollway Value Pricing Pilot Study, Illinois

Pricing on Toll Facilities	Illinois Tollway Value Pricing Pilot Study, Illinois
Operations	The Illinois State Toll Highway Authority implemented new tolls for trucks and cars in January 2005; Tollway operated program (operates 274 miles of toll roads in northern Illinois, most serving Chicago area). Truck tolls increased by 100-220 percent (depending on class, plaza, time of day), with average increase 160 percent, but discounts from 25 to 33 percent were offered to trucks paying with I‑Pass transponders and traveling off-peak; night travel (10-6) discounted the same amount regardless of payment method; e.g., large trucks went from typical flat mainline rate of $1.25 to $4 daytime peak (I‑Pass and cash); $3 daytime off peak (I‑Pass) and $4 cash; and $3 overnight (I‑Pass and cash). Passenger car tolls for cash customers also were changed: increased by 100 percent (from about $0.03 to $0.06 per mile), with no change for passenger cars paying with I‑Pass transponders; overall increase across cash and electronic paying vehicles was 25 percent since 75 percent paid electronically (first 6 months 2005).
Cost, Finance, and Revenue	FHWA VPPP funding of $750,000 was awarded in FY 2006. Revenue generation was an important impetus for the study phase; Tollway had not increased tolls since 1983 and wanted revenues for capital plan (rehab, widening, extension); Tollway also interested in boosting electronic toll payment for possible cost saving, efficiency, and revenue gains. Wilbur Smith Associates models used to predict revenue changes from electronic payment and varying toll schedules, predicted increase in revenues by about 50 percent under planned increase in truck rate depending on vehicle size, and with doubling of toll for cash paying passenger cars, electronic payment unchanged.
Policy/Institutional	No new legislative authority or clearances required for Tollway to make toll changes. Interstate toll policy issue: after a period of negotiation, a resolution approved by the Indiana tolling authority allows I‑Pass customers to get the same 40-percent discount as Indiana’s I‑Zoom drivers (once the electronic tolling is operational); similarly, I‑Zoom customers would get the same 50-percent discounts when using Illinois’ Tollway system; Indiana drivers with an I‑Pass transponder not required to get new I‑Zoom to get the Indiana discounts (“Indiana resolves Tollway IPass dispute with Illinois”).
Outreach/Acceptance	Focus Groups – Before price change, RGS contractor ran six focus groups of I‑Pass and cash frequent and infrequent users of facility, in large part to assess feasibility and acceptance of possible HOT and BRT options; findings suggest new HOT lanes far preferred to pricing existing lanes, travel benefits must be substantial, neither carpooling or BRT held much attraction in HOT. Truckers – Perceptions before price change assessed via interviews with company operators/managers; 23 interviewed on payment method and time-of-day charges; result showed preference for electronic payment and about 20 percent prefer to shift time of trip under time variable pricing but by small amount (15 minutes). Public Hearing – Tollway held public hearing prior to adopting new tolls in September 2004. Communication Campaign – “Aggressive sales campaign” by Tollway to promote use of electronic payment and inform users and potential users of new toll structure before and after toll change; Tollway met with “dozens of civic groups and editorial boards” to convey that congestion relief would be forthcoming if enough drivers switched to electronic payment; this message was featured on electronic Tollway billboards starting in mid-November 2004 and highlighted in various media outlets; Tollway also partnered with grocery chain, Jewel-Osco for volume retailer for transponders; between November 2003 and December 2005, Jewel-Osco accounted for nearly three-quarters of total transponder sales, as the number of active I‑PASS holders increased from 1.1 million to 2.5 million (Chicago Federal Letter, April 2006).
Technology	Pricing via vehicle mounted transponder and signal sent to transponder; toll deducted from pre-paid account; some windshields and cars with special features such as GPS require an exterior transponder (license plate tag); about three million transponders currently in use. Refundable deposit of $10 and $40 in pre-paid tolls is charged at the time of purchase; auto-pay replenishment option available on credit card, with minimum balance and replenishment amounts calculated based on the average usage during the previous six months. Illinois Tollway’s goal was to convert all facilities from a traditional barrier system to an end-to-end open road automated tolling system (RFID technology to allow payments at highway speeds). This has been accomplished recently.
Equity/Environmental	Evaluators speculate loss of truck trips observed “likely came” more from the smaller independent firms which cannot easily pass on increased cost to customers. A study of changes in I‑Pass ownership trends before/after price changes and the corresponding public awareness campaign concluded there was a “boost in I‑PASS ownership ratios across all income groups. In fact, by February 2005 the number of I‑PASS accounts nearly equaled the number of the most intensive Tollway users-those likely to be using it for their everyday commute-in the lowest-income group” However, in terms of absolute proportions of I‑Pass users, income is a major determinant. The study says, “… high-income ZIP Codes (median-income among working households above $80,000) have more than twice as high a share of likely Tollway drivers as low-income ZIP Codes (less than $60,000).” (Chicago Federal Letter). No air quality impacts estimated.
Impacts	Use of Electronic Payment – Advance publicity regarding the advantage to electronic payment and new rate structure itself appears to have increased electronic payment from about 50 percent to over 70 percent in first week after rate change (January 1, 2005); absolute increase went to 1.9 million. Passenger Car Traffic Diverted from Tollway – Estimated at 3 percent for passenger cars (about 58,940 vehicles per day); impact on passenger car traffic was small since tolls did not go up for cars paying electronically is large proportion of car traffic; using “weighted average” of toll increase on passenger vehicles, the calculated elasticity is ‑0.11. Diversion of Trucks from Tollway – Estimated at about 11 percent for trucks; using “weighted average” of toll increase, the calculated elasticity is ‑0.07. Truck Diversion from the Peak Period – Was not significant; “modest time-of-day pricing incentives were not effective in shifting the time of commercial vehicle trip-making,” according to evaluators; and, “distribution of truck trips by time of day changed very little”; after pricing truck company interviews (limited number – see below) indicate 51 percent of trips in the peak before, and 56 percent after, an increase in peak-period travel (Appendix F). Trucker Reactions/Perceptions – Interviews with truck companies after toll changes (November 2005) indicated “little change” in overall use of the Tollway or the hourly trip making patterns”; reasons included “inflexibility of delivery times and ability to pass on extra costs … of tolls to their customers”; after interviews indicate “no substantial change in proportion of companies using I‑Pass with main reason being much of business is through contractors or owner-operators who make the choice, though there was an increase of 13 percent in overall I‑Pass transactions among trucks.
Evaluation	Before/After Transaction Data – Toll transactions from first 6 months of 2005 compared to the same six months in 2004 before toll change to assess impacts of toll change; last six months of 2005 data not used because construction and substantial gas price changes would cloud findings. Truck Before/After Survey – Interviews of selected trucking companies (see next point) were carried out before (January 2004)/after (November 2005) pricing, assessing changes in tolls paid, reactions to toll increase. Trucker Survey Challenge – Among 56 truck companies selected for interview before pricing changes, representing cross section of sizes, area terminals and types of shipments; only 23 interviews were completed, even with offer of $30 and telephone, fax, e‑mail possible response methods; in after pricing survey, the 23 companies responding in the before survey were contacted, but only 15 participated, obliging surveyors to add 5 new companies; evaluators indicate stated-preference surveys were feasible to administer to car passengers but not possible to administer to truckers.

Sources:
Illinois Tollway Value Pricing Pilot Study, Final Report, Wilbur Smith Associates, January 2007.
Illinois State Toll Highway Authority (2005). Press Release 03-Jan-05 and 10-Feb-05.
http://www.illinoistollway.com.
Illinois State Toll Highway Authority web information:
http://www.illinoistollway.com/portal/page?_pageid=133,1392806&_dad=portal&_schema=PORTAL.
Chicago Fed letter reference:
http://findarticles.com/p/articles/mi_qa3631/is_200604/ai_n17174749.
Indiana resolves tollway IPass dispute with Illinois, Mike Smith AP Writer, Chesterton Tribune, 6/15/07. See: http://chestertontribune.com/Northwest%20Indiana/6155%20indiana_resolves_tollway_ipass_d.htm.

Table 3.3 Variable Tolls on the New Jersey Turnpike, New Jersey

Pricing on Toll Facilities	Variable Tolls on the New Jersey Turnpike, New Jersey
Operations	The New Jersey Turnpike Authority operates a 148-mile facility with 28 interchanges. Average daily trips exceed 500,000 vehicles. The variable pricing program was initiated in September 2000 in conjunction with a general toll increase. The Turnpike Authority introduced E‑ZPass toll technology along with the first stage of its time-of-day (TOD) pricing program. In September 2000, the first stage of the TOD pricing program was introduced by increasing toll levels for cash users (e.g., cash tolls for passenger cars were increased by 20 percent) and peak E‑ZPass users (by 8 percent), while charging off-peak E‑ZPass users the same toll that was paid in 1991. The differential between E‑ZPass peak- and off-peak tolls for passenger cars amounted to an average of $0.35, or about a 7 percent reduction for off-peak travel. The differential between cash and E‑ZPass off-peak averaged about $0.90, or about a 16 percent toll reduction (for adopting E‑ZPass and traveling in off-peak hours). In January 2003 second stage, the general toll level was increased, but time-of-day toll differentials were maintained. The differential between E‑ZPass peak- and off-peak tolls now averaged $0.60, or an 11 percent reduction for off-peak travel. The differential between cash and E-Z Pass off-peak was $1.60, or a 25 percent toll reduction. On average, across all interchanges the passenger peak-hour tolls in 2003 were 15 percent higher than off-peak-hour tolls. Beginning in January 2006, peak-period toll discounts for E‑Z Pass users were eliminated, but the off-peak E‑Z Pass toll discounts were continued. Thus, the differential between peak- and off-peak E‑Z Pass users is now the same as shown for cash users, above.
Cost, Finance, and Revenue	FHWA VPPP funding of $477,468 was awarded to the project in FY 2001. TOD pricing was intended to improve use of capacity, not as a revenue enhancing measure. The combination of higher peak tolls and off-peak toll discounts was use to reach revenue targets.
Policy/Institutional	The introduction of TOD tolls had the strong support of the Executive Director of the New Jersey Turnpike Authority who wanted to develop marketing strategies that would reduce traffic flows during congested peak periods and make better use of existing Turnpike capacity. The initiative taken by the New Jersey Turnpike Authority stimulated other agencies in the Northeast to consider, and, in some cases, adopt variable toll strategies. Existing toll agencies can successfully implement variable tolling strategies by implementing off-peak toll discounts along with scheduled toll increases.
Outreach/Acceptance	There was considerable advertising of the pricing initiative and the program was introduced with minimal opposition from the public or various stakeholders, although there was initial opposition from truckers and the Automobile Club of New Jersey. Focus groups, public hearing, media announcements, hand outs, newsletters, all were utilized leading toward the program implementation. Strong support was provided by several environmental/transit advocacy groups. The TOD initiative shared media attention with the toll increase and the introduction of E‑ZPass, thereby diluting the public notice it got. A 2004 survey (general public, facility users?) found that the majority of respondents (64 percent) were not aware of any discounts on the New Jersey Turnpike and only 10.5 percent were aware of discounts associated with the time of day. Support by top-level Turnpike Authority (and other state) officials were key ingredients in the successful introduction of TOD pricing.
Technology	E-Z Pass toll technology was introduced in conjunction with the TOD pricing program.
Equity/Environmental	Systemwide emissions costs were increasing between 2001-2002 and 2002-2003. After the simultaneous introduction of the first phase of TOD pricing and the E‑ZPass deployment, there was a substantial reduction in vehicle emissions costs at the toll plazas as compared to vehicle emissions costs in 2000. It was not possible to isolate the emissions reducing effects of E‑ZPass deployment from the effects of TOD pricing, but reductions in delays (and emissions) at toll plazas are much more likely to be related to E‑ZPass deployment. Between 2000 and 2001, there was a reduction in emissions levels as high as 10.7 percent. However, emission levels increased between 2001 and 2003, likely due to increases in overall traffic levels. User profile by income surveys, the number of survey respondents indicating change in behavior due to pricing was so small that no meaningful results were obtained.
Impacts	Despite the toll increases that took place when time time-of-day (TOD) pricing was introduced, there was a significant increase in overall traffic demand on the Turnpike. The rate of change in traffic demand was increasing both before and after the introduction of the new toll schedule. Over the period before and after the first phase of TOD tolls, absolute traffic volumes during the off-peak periods increased at a faster rate (10 percent) than absolute traffic volumes during peak periods (6 percent for the a.m. peak and 4 percent for the p.m. peak). The increase in the percent share of off-peak traffic (1.1 percent) was statistically significant, as was the decrease in the percent share of peak-period traffic (1.7 percent for the a.m. peak and 3.7 percent for the p.m. peak). After implementation of the second phase of the TOD pricing program the trend in traffic flows was reversed, with absolute traffic flows during peak periods increasing at a faster rate (15 percent and 10 percent) than the traffic increase in the off-peak period (9 percent). The percent share of peak traffic increased while the off-peak percent share decreased. Analysts posit that the initial shift to the off-peak may reflect response to lower off-peak tolls, but the resulting increase in traffic may have led to slower traffic times, which, in turn, may have reduced the attractiveness of off-peak travel. A disaggregate analysis was conducted to further explore the reasons for the change in travel patterns. This analysis used traffic and travel-time information for each Turnpike entry-exit pair (tracked same actual users panel style before/after or used test cars – big difference) for the period between three months before and three months after the second stage of the pricing program. The analysis indicated that commuters responded more to congestion (lower travel times) than the slightly higher tolls during the peak “…given the small differential between peak and off-peak period toll levels, most of the users prefer peak periods with lower travel times and higher tolls instead of peak shoulders with higher travel times but lower tolls.” Descriptive analysis of the impacts of TOD pricing (based on 513 passenger surveys collected in June-July 2004) indicate that 7 percent of individuals changed their travel behavior after the first phase of the program. There was huge reduction in toll plaza delays, though this could be largely attributed to large increase in E‑ZPass use.
Evaluation	Evaluation study conducted by a team from Rutgers University and the Rensselaer Polytechnic Institute. Areas covered were traffic impacts, estimation and analysis of traffic delays and emissions (using a microscopic traffic simulation model of the Turnpike), impacts of TOD pricing on travel behavior, economic value of travel-time savings, demand elasticity, and media/decision-maker reactions to the TOD pricing initiative. Simultaneous introduction of TOD pricing and E‑ZPass toll technology made it difficult to isolate the effects of TOD pricing. The small number of passenger survey respondents indicating a change in travel behavior due to the TOD pricing initiative made it impossible to draw meaningful conclusions about demographics or behavioral choices. TOD toll differentials can influence travel behavior, but relatively small differences in toll rates may not have a significant and long-lasting effect, particularly if time savings are realized by traveling in the peak.

Sources:
Evaluation Study of New Jersey Turnpike Authority’s Time of Day Pricing Initiative, Final Report (FHWA/NJ-2005-012, May 31, 2005).
New Jersey Turnpike Authority web site, www.state.nj.us/turnpike/index.htm.
Value Pricing Project Quarterly Report, Office of Operations, Federal Highway Administration, July-September 2007.

Table 3.4 Variable Tolls on the Port Authority of New York and New Jersey (PANYNJ) Bridge/Tunnel Crossings

Pricing on Toll Facilities	Variable Tolls on the Port Authority of New York and New Jersey (PANYNJ) Bridge/Tunnel Crossings
Operations	Following approval by the PANYNJ Board of Commissioners, variable toll program introduced on March 25, 2001 at the two tunnels and four bridges that connect New Jersey and New York City. The Governor of each state appoints six commissioners to the PANYNJ Board, each appointment subject to the approval of the respective state senate. The Governors retain the right to veto the actions of the Commissioners. The Board appoints an Executive Director to carry out its policies and manage day-to-day operations. Only users of the electronic toll collection system (E‑ZPass) eligible for variable toll program. Peak periods are weekdays 6:00 a.m. to 9:00 a.m. and 4:00 p.m. to 7:00 p.m., and Saturdays and Sundays 12 noon to 8:00 p.m. Variable toll program based on high $6 round-trip cash toll rate combined with varying E‑ZPass electronic toll discounts depending on time of day. Passenger vehicles using E‑ZPass enjoy a $2 off-peak discount (from the cash toll rate) and a $1 peak-period discount. A $1 carpool rate is available for passenger vehicles carrying 3 or more people. (The latest tolls doubled the dollar discount to $2 between peak and off-peak hours by eliminating E-ZPass peak period discounts. Now, cash and E-Zpass pay $8 peak rate, but E-ZPass holders pay $6 off-peak). Trucks receive a $1 E‑ZPass discount per axle in the off-peak hours and a $2.50 discount during weekday overnight hours (midnight to 6:00 a.m.). These crossings carried average daily eastbound traffic of 352,000 vehicles, or more than 126 million eastbound (New York bound) vehicles in 2004 (115.2 million autos, 3.1 million buses, 8.2 million trucks). Tolls are collected in the eastbound direction only. In 2006, average daily eastbound traffic on these crossings totaled just over 127 million vehicles (115.5 million autos, 3.1 million buses, 8.4 million trucks).
Cost, Finance, and Revenue	In 2000, The PANYNJ established an incremental revenue goal of $150 million annually to sustain the $14 billion in projects included in its five-year capital investment program (Muriello and Jiji). It established a two-tiered toll increase (tolls raised more in the peak than in the off-peak period) as part of the program to meet this revenue goal. However, the pricing reform was viewed less as a revenue enhancing measure than as a way of improving the use of capacity when the toll increase was enacted.
Policy/Institutional	The PANYNJ Board of Commissioners adopted a variable toll strategy for users of the electronic toll collection system in March 2001.
Outreach/Acceptance	Initially, Port Authority leadership wanted to implement a straight toll increase to meet revenue goals. Staff of the Tunnels and Bridges unit developed analyses of both straight toll increases and a toll increase incorporating variable tolls. Analyses demonstrated that objectives of variable tolling, such as shifting traffic to off-peak periods, thereby reducing congestion costs and air quality concerns, could be met without detriment to expected revenues. This gained internal champions for the cause of variable tolling, including key PANYNJ leaders and state officials in New York and New Jersey. Once internal agreement had been reached, the variable toll plan was made public. The external debate, as reflected primarily in news coverage and public hearings, was much more focused on the magnitude of the toll increase than on the variable tolling aspect. Newspaper coverage reflected some positive reaction to the use of congestion pricing, but considerable opposition to the size of the toll increase. Public statements by PANYNJ to large employers and messages placed in media outlets stressed the value of flexible working hours in allowing employees to take advantage of the toll discount program. The New York University Wagner Rudin Center for Transportation Policy and Management assessed the acceptability of the variable toll program to opinion leaders and decision-makers before and after implementation of the new toll schedule. Two sources of information were used in making this assessment: 1) interviews with key stakeholders, opinion leaders, and decision-makers involved in establishing the new toll schedule; and 2) intense review of media statements prior to and immediately following implementation of the new toll schedule. In general, the opinion assessment concluded that opponents and proponents of the change in toll schedule maintained their pre-implementation views after implementation.
Technology	E‑ZPass is a toll collection system that uses RFID (radio frequency identification) technology to allow a driver to pass through a toll booth and pay the toll without stopping the vehicle. The driver typically pre-pays a month’s worth of tolls in advance and is issued a transponder that mounts inside the windshield. E-ZPass transponder does not send account information back to the roadside antennae. A unique transponder identification number is transmitted at roadside, and all account management and financial reconciliation is performed privately in the back office. The PANYNJ (and NJ Turnpike) charge tolls for commercial vehicles via electronic toll collection both via treadle equipment in traditional toll lanes and in highway-speed open-road tolling through road loops and advanced vehicle classification software.
Equity/Environmental	Combined effects of E‑ZPass use and the off-peak toll discount program would be expected to have some positive effect on pollution levels through reduction of peak-period congestion and waiting times at toll collection sites, though air quality impacts have not been quantified.
Impacts	Research found a statistically significant shift in weekday peak-period traffic percent share to the hours just before (5:00 a.m. to 6:00 a.m., 3:00 p.m. to 4:00 p.m.) or after (9:00 a.m. to 10:00 a.m., 7:00 p.m. to 8:00 p.m.) the peak toll rates are in effect for both autos and trucks (Evaluation Study). Traffic shifts resulted in travel-time savings and an earlier end to the morning peak by as much as 20 minutes at certain crossings. Overall traffic demand was dampened in 2003 due to sluggish economic conditions in New York City. This led to some shift back to the now less-congested peak periods by off-peak motorists. This may indicate that the effectiveness of a relatively small off-peak toll discount in shifting traffic out of the peak is highly correlated with the level of congestion during the peak period (Muriello and Jiji). Weekend car and truck traffic percent share did not have a statistically significant effect on peak shoulder traffic. Study results indicate that 7.4 percent of passenger trips and 20.2 percent of truck trips (including those that increased shipping charges or switched to E‑ZPass) changed behavior because of time-of-day pricing. Time-of-day pricing resulted in an increase on the percent share of peak shoulder traffic for both trucks and cars during weekdays, and short-term pre-peak elasticities are higher than post-peak elasticities during both a.m. and p.m. periods on weekdays for almost all of crossings.
Evaluation	Evaluation study conducted by a team from the Rensselaer Polytechnic Institute, Rutgers University and New York University. Three main focus areas of evaluation research: Disaggregate behavioral impacts, aggregate impacts on traffic and transit use, and public reactions to the variable toll initiative. Focus groups (four with automobile drivers, two with truck dispatchers) and surveys were conducted with passenger car users and truck dispatchers to assess behavioral effects. A total of 505 passenger car users and 200 carriers were surveyed, using computer aided telephone interviews. Traffic count data classified by vehicle type and hour of day at PANYNJ toll facilities before and after implementation of the new toll schedule were used to quantify effects of the variable toll program on traffic patterns, E‑ZPass usage and time of day. To avoid using traffic data affected by the impacts of 9/11/2001 and/or the various operational restrictions placed at PANYNJ facilities after 9/11, the analyses focused on the time from April-August 2001 for the period after the new toll schedule went into effect. Media coverage of the toll increase and variable toll program was intensively reviewed.

Sources:
Holguin-Veras, Jose, et al, Evaluation Study of Port Authority of New York and New Jersey’s Time of Day Pricing Initiative, Final Report (FHWA/NJ-2005-05, March 2005).
Mark F. Muriello and Danny Jiji, The Value Pricing Toll Program at the Port Authority of New York and New Jersey: Revenue for Transportation Investment and Incentives for Traffic Management, Transportation Research Record, Volume 1864 (2004).
The Port Authority of New York and New Jersey, Comprehensive Annual Financial Report (2004, 2006).
Rudin Center for Transportation Policy and Management, NYU Robert F. Wagner Graduate School of Public Service, Evaluation Study of the PANYNJ’s Value Pricing Initiative, Task 5: Monitoring of Media and Decision-Makers’ Reactions (January 2004).
U.S. Department of Transportation, Federal Highway Administration, Report on the Value Pricing Pilot Program through March 2004.
Value Pricing Project Quarterly Report, Office of Operations, Federal Highway Administration, July-September 2007.
www.panynj.gov/CommutingTravel/CustomerRelations/pdf/E-ZPass_fact_sheet.pdf.

Table 3.5 Variable Tolls on the Pennsylvania Turnpike, Pennsylvania

Pricing on Toll Facilities	Variable Tolls on the Pennsylvania Turnpike, Pennsylvania
Operations	The Pennsylvania Turnpike Commission completed a study of value pricing strategies on urban interchanges in the Philadelphia and Pittsburgh metropolitan areas. The objectives of the study were to analyze the potential of pricing toll facilities to: 1) provide an economic incentive to shift traffic out of peak periods; 2) provide an economic marketing incentive to encourage use of electronic toll collecting; 3) promote the safe and efficient movement of traffic on the Turnpike; and 4) enhance traffic and revenue growth on the Turnpike to help meet forecasted revenue needs. A value pricing feasibility study began in 2002 and focused on the PA Turnpike’s urban ticket interchanges in Pittsburgh and Philadelphia. Two independent analyses were conducted to address specific concerns of users. The first evaluated the potential to shift existing truck traffic from PA Route 41 to the Turnpike. The second involved the application of an across the board discount for motorcycles using E‑ZPass. The motorcycle discount was eventually implemented. A detailed profile of weekday traffic conditions were developed to provide a baseline for analysis of the impacts of value pricing. Seven variables were identified which formed the basis of all alternatives studies and included: 1) peak-period hours of application, two or three hours; 2) area of application, urban interchanges or full Turnpike; 3) discount method, fixed or variable; 4) method of time delineation, time of entry or exit; 5) days of application, weekdays or all days; 6) vehicle applicability, passenger cars or trucks; and 7) amount of toll differential between peak and off-peak periods and between cash and E‑ZPass users. A short list of scenarios was identified for further study by PTC, and included: Scenario 1 – Two hours during peak periods in urban areas with a fixed discount on weekdays applicable to all vehicles. For cash users rates of peak, off-peak, and night tolls rose, E‑ZPass rates increased only during peak hours, other rates did not change. Scenario 3 – Two hours per peak period on the full Turnpike with a fixed discount on weekdays applicable to all vehicles. For cash users rates of peak, off-peak, and night tolls rose, E‑ZPass rates increased only during peak hours, other rates did not change. Scenario 6 – Two hours per peak period in urban areas with a fixed discount on weekdays, applicable to all vehicles. For cash users rates of peak, off-peak, and night tolls rose, E‑ZPass rates increased only during peak hours, other rates did not change. Scenario 9 – Two hours per peak period in urban areas with a fixed discount on weekdays, applicable to all vehicles. For cash users rates of peak, off-peak, and night tolls rose, E‑ZPass rates increased during peak hours but at a lower amount than scenarios 1,3, and 6, other rates did not change. Scenario 15 – All hours on the full Turnpike with a fixed discount on all days of the week applicable to all vehicles. For cash users rates of peak, off-peak, and night tolls rose, E‑ZPass rates did not change. Scenario 17-1 – Two hours per peak period in urban areas with a fixed discount on weekdays, applicable to all vehicles. For cash users rates of peak, off-peak, and night tolls rose, E‑ZPass rates increased during peak hours, other rates did not change. Scenario 17-9 – Two hours per peak period in urban areas with a fixed discount on weekdays, applicable to all vehicles. For cash users rates of peak, off-peak, and night tolls rose, E‑ZPass rates increased during peak hours but at a lower amount than other scenarios, other rates did not change. Based on the above pre-study of value pricing, the PTC applied for FHWA VPPP funding in 2002 to implement value pricing. The planned phase 2 of this second study effort is to involve implementation of short-term (approximately 12 months) pilot variable pricing programs for both commercial vehicles and passenger vehicles, with emphasis on congestion management in the highly congested Philadelphia region. The Commission has in place an existing discounted commercial hauler fare collection system for most major trucking firms that utilize the Turnpike system on a recurring basis.
Cost, Finance, and Revenue	In FY 2001, $776,000 was awarded to study value pricing concepts on the Turnpike. In FY 2002, $800,000 was awarded for implementation of value pricing. The total VPPP funding level was $1,576,000.
Policy/Institutional	The five-member Pennsylvania Turnpike Commission was created in 1937 by the Pennsylvania Legislature, Act 211, with the authority to construct, finance, operate and maintain a toll highway. Concurrent with the value pricing study, the Pennsylvania Turnpike Commission (PTC) implemented electronic toll collection (E‑ZPass) for travel between the ticket interchanges on its mainline system. The PTC-equipped additional lanes with E‑ZPass. This work was accomplished without Federal funds.
Outreach/Acceptance	Turnpike patrons’ and stakeholders’ opinion were assessed during two focus groups in Philadelphia and two groups in Pittsburgh. Most participants expressed dislike of variable time-of-travel toll rates. Commuters felt they were exercising their flexibility to travel to the maximum amount possible already. Noncommuters indicated greater likelihood to shift travel time to avoid peak-period tolls. Each focus group was presented with 18 value pricing messages to be used in future marketing campaigns. The top three messages were: “EZ Pass saves me time at toll plazas. Now, with value pricing I can save money too.” “Using the Turnpike is less stressful than traveling on other more congested highways.” “Value pricing is an idea whose time has come. It makes sense to use financial incentives to manage traffic congestion.” A series of interviews were conducted with regional stakeholders from PennDOT, the Turnpike Commissioners, Transportation Management Association, City officials. The results of these interviews were significant. Support dropped as the level of knowledge increased. Those with knowledge of value pricing and were close to the proposals expressed concerns of implementing a project. Support increased in direct proportion to stakeholders distance from project. Stated-preference surveys were administered to drivers using the Turnpike within the designated VP areas. 1,800 surveys were collected and 25 trucking companies were contacted. It was found that relatively large toll differentials would be required to alter motorist travel times. Trucking companies were somewhat more favorable to the value pricing concept, as between 15 and 35 percent of respondents indicated they would shift the scheduling of trucks to take advantage of off-peak discounts or to avoid peak surcharges.
Technology	Not evaluated.
Equity/Environmental	Not addressed in this study.
Impacts	Of the 11 scenarios tested the three highest scoring were number 3, 17-1, and 17-9. These scored high on revenue impact and impact on interchange and mainline operations at 2002 and 2012 levels. Scenario 3 was shown to have an annual revenue impact of an increase between 16.7 and 20.7 percent ($62,717,000 to $77,416,000) and total a.m. peak traffic impact of a percent decrease between 16.4 and 21.8 percent. Scenario 17-1 was shown to have an annual revenue impact of an increase between 19.9 and 24.5 percent ($74,696,000 to $91,657,000) and total a.m. peak traffic impact of decrease between 16.4 and 21.8 percent. Scenario 17-9 was shown to have an annual revenue impact of an increase between 19.4 and 24.0 percent ($72,757,000 to $89,873,000) and total a.m. peak traffic impact of decrease between 14.3 and 20.2 percent. An analysis of commercial vehicle only night-time discount value pricing scenarios was completed. Under all scenarios tested, the net revenue loss to the Turnpike was minimal, under the highest percent estimated net revenue loss amounted to 1 percent of total system toll revenue, the lowest percent loss was 0.3 of total system revenue. The net impact on commercial volumes in night-time periods ranged from an increase of 0.2 percent to 0.7 percent as a result of an off-peak discount. An analysis of the potential for shifting heavy truck traffic from PA Route 41 to the PA Turnpike using discounted toll rates for commercial traffic was completed. Findings showed that the additional time and distance truck drivers would incur using the Turnpike cost them nearly $20 more than using the current Route 41 trip. Thus, after offering toll rate reductions of up to 50 percent on the Turnpike, less than 50 trucks per day were estimated to shift current routing. The net toll revenue impacts was estimated to be a loss of $160,000 at 2003 levels. An estimate of the toll revenue impacts of providing discounts of 15 to 50 percent for E‑ZPass motorcycle patrons was completed. At the highest level of discount (50 percent) the net revenue loss was estimated to be 0.1 percent of total system revenue. As a result, a motorcycle E‑ZPass reduction of 25 percent was implemented on July 1, 2003. Despite the prediction of favorable results the turnpike decided not to adopt variable tolls.
Evaluation	A detailed logit model was developed based on stated-preference surveys and used to determine the shift potential (of travel to off-peak periods and from cash to E‑ZPass users). A regional Turnpike model also was developed to determine toll sensitivity of motorists.

Sources:
Pennsylvania Turnpike Commission, Summary Report: Pennsylvania Turnpike Value Pricing Study. Prepared by Wilbur Smith Associates. March 2004.
http://knowledge.fhwa.dot.gov/cops/hcx.nsf/All+Documents/750C4F311CB4924A85256DC500657FE0/$FILE/Summary PA Turnpike Final Report.pdf.

previous | next