Traveler Response to Transportation System Changes Handbook, Third Edition: Chapter 5, Vanpools and Buspools (2005)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Vanpool subsidies, particularly within employer-sponsored programs or partnerships of employ- ers and third-party operators, should be taken in context with benefits. A number of expanding companies report savings in parking space requirements and reduction of localized traffic con- gestion among other benefits. In a self-assessment covering 160 corporations, a majority of employ- ers rated their vanpool programs as definitely cost effective, even when objective analyses showed that most employers did not achieve positive or even break-even revenue returns. Sixty percent of the firms paid less than $10,000 per year to support ridesharing programs, including administra- tion (Wegmann, 1989). In the circa 1980 third-party vanpool demonstrations, it was found that among other cost savings for vanpoolers themselves was the ability to sell a household vehicle or defer purchase of a new one. In Norfolk, 5 percent of vanpool passengers and 21 percent of drivers sold a vehicle, with 28 and 29 percent, respectively, claiming that they had deferred purchase of a new vehicle. Percentages for Knoxville and the Golden Gate Corridor were lower but still substantial (Heaton et al., 1981; Dorosin, Fitzgerald, and Richard, 1979). ADDITIONAL RESOURCES The Comsis Corporation and Institute of Transportation Engineers report, Implementing Effective Travel Demand Management Measures: Inventory of Measures and Synthesis of Experience, prepared for the Federal Highway Administration and the Federal Transit Administration, provides a compre- hensive review of vanpooling as a strategy, its market and cost effectiveness, and parametric esti- mates of travel and traffic impact potential. Published as report DOT-T-94-02, the document includes case studies of both employer based and third-party vanpooling (Comsis and ITE, 1993). Vanpooling—A Handbook to Help You Set Up A Program At Your Company is available on the FTA web- site at http://www.fta.dot.gov/fta/library/planning/VANPOOL/vanpool.html. This manual, prepared by Commuter Transportation Services, Inc., provides program design and implementa- tion procedures targeted toward employee transportation coordinators who are in charge of van- pooling efforts and choose to lease vanpool vehicles (Commuter Transportation Services, 1993). The Texas Transportation Institute report, Transit-Operated Vanpools in the United States: Selected Case Studies provides encapsulated case-by-case and summary survey findings with emphasis on the insti- tutional and funding aspects of transit provider vanpool operations (Higgins and Rabinowitz, 2002). The on-line TDM Encyclopedia at www.vtpi.org examines vanpooling in the context of ridesharing/ TDM overall with references and periodic updates (Victoria Transport Policy Institute, 2004). Chapter 19, “Employer and Institutional TDM Strategies,” of this TCRP Report 95, “Traveler Response to Transportation System Changes” Handbook will also cover vanpooling and buspooling in the over- all TDM program context. CASE STUDIES The 3M Company Employer Based Vanpool Program Situation. The 3M Company, St. Paul, Minnesota, in 1973 began an experimental vanpooling program for employees not conveniently served by transit. The 3M Center involved consisted of 20 buildings housing approximately 10,000 administrative and laboratory employees, located on 5-41

a 400 acre site at the eastern edge of St. Paul. The center had facilities to park 8,000 vehicles. A 1970 Home-Work Travel Survey showed only 43 persons using transit, and a 1.24 average auto occupancy. Actions. Standard 12-passenger vans were purchased by the 3M Company and provided to van- pools formed on the basis of a special pilot program questionnaire. Drivers were 3M employees willing to pick up and drive at least 8 other employees to and from work. Vehicle maintenance and preferential parking for the vans were provided by the 3M Company. Drivers’ responsibilities included picking up and delivering passengers on a set schedule, arranging for service and main- tenance of the van, keeping at least 8 paying passengers in the vanpool, and providing for standby drivers. In exchange for their responsibilities, the drivers were not required to pay the approxi- mately $20 to $30 monthly fare charged other passengers, were given personal use of the van dur- ing non-work hours for a reasonable mileage rate, and could keep the fares for any passengers over the minimum of eight. Analysis. The 3M Company undertook and made available detailed evaluations of the 3M van- pool and overall ridesharing program in the initial years. In April 1974 and August 1976, survey questionnaires were given to all participants in the Commute-A-Van program. Responses were obtained from 437 and 566 users respectively. The full array of employee mode shares for the jour- ney to work was tracked through 1985. Results. The 3M vanpooling endeavor began as a 6 van pilot project in April, 1973. As a result of the success of the original experiment, the number of vans was gradually increased to a total of 86 carrying over 800 riders as of January 1977, the date of the second status report. When surveyed in 1974 each van was carrying an average of 11.36 persons for an average monthly fare of $23.72 and an average round trip distance of 49 miles. The operating ratio (total operating costs divided by operating income less amortization) was 0.88. The 86 vanpools recorded in 1977 reduced the demand for parking by 735 spaces and saved well over 2,250,000 vehicle miles of travel and 190,000 gallons of gasoline per year. Responses were virtually identical for both the 1974 and 1976 surveys of vanpool users. Of those who responded, 49 percent previously drove to work alone, 7 percent drove with a passenger, 23 percent were in a rotating carpool, 16 percent were a carpool rider, 4 percent were dropped off at work, and 1 percent rode transit. Eighty percent of the respondents found the vanpool more con- venient than their former means of getting to work and 97 percent intended to continue using the vanpool on a permanent basis. The average travel time for vanpoolers before using the van was 28 minutes compared with 38 minutes afterwards. One quarter traveled over 20 minutes longer after joining the vanpool. Vanpool program benefits were numerous and well distributed. Participating commuters saved money, reduced the tensions associated with commuting, and freed a car for use by other family members. Non-users benefited from the reduction in congestion and parking demand in and around the 3M Company. The Company itself was able to expand without adding more roadway and parking capacity. More . . . For use in vanpool planning a Utility Ratio was derived, defined as the passenger-pickup time divided by the line-haul time. It was anticipated that the larger the ratio, the more difficult it would be to form and operate a vanpool. Some problems were encountered in forming vanpools where the Utility Ratio was greater than 1.0, but ultimately many operating vans fell into this cat- egory. In 1974, when 52 vans were operating, the average Utility Ratio was 1.18 and the Utility Ratio breakdowns were as recorded in Table 5-19. 5-42

Table 5-1 in the main “Response to Vanpool and Buspool Programs” section documents the effec- tiveness of the 3M program over time. Peak vanpool usage was circa 1980, with effects of 1970s energy shortages not yet worn off, when 10.3 percent of all employees commuted by vanpool. Vanpool usage at the site dropped from 135 vans in 1980 to 105 vans in 1985. Company managers speculated that high employee turnover, relocations, and the introduction of flextime were to blame for the decline. In 1995, of the nearly 13,300 employees, 525 or 3.9 percent used 68 vanpools to travel to work. There is another possible interpretation, although lacking recent data on pick-up and line-haul time and mileage breakdowns, it involves considerable speculation. Two rules of thumb applied as indicators of likely vanpool attractiveness are that the passenger-pickup time to line-haul time ratio should be less than 1.0, or that the line-haul distance should be at least 20 miles. Either of these criteria suggests that only a third of the 1970s 3M vanpool users, as broken down in Table 5-19, were vanpooling under inherently attractive circumstances. One-third of the peak 1980 vanpool share of 10.3 percent is 3.4 percent, very close to the 3.9 percent achieved in 1995. This may be purely coincidental, or it may validate the rules of thumb, suggesting that 3M vanpooling was operating in a “supersaturated” mode in the 1970s, perhaps by virtue of the energy crises com- bined with a corporate vanpooling ethic and enthusiasm that may ultimately have proved hard to sustain. Sources. Owens, R. D., and Sever., H. L., The 3M Commute-A-Van Program: Status Report. Reprinted by the Federal Highway Administration, U.S. Department of Transportation, Washington, DC (May, 1974). Also Status Report II. 3M (January, 1977). • Kuzmyak, J. R., and Schreffler, E. N., “Evaluation of Travel Demand Management (TDM) Measures to Relieve Congestion.” Prepared by Comsis Corporation and Harold Katz and Associates for the Federal Highway Administration, Washington, DC (February, 1990). • Comsis Corporation and the Institute of Transportation Engineers, “Implementing Effective Travel Demand Management Measures: Inventory of Measures and Synthesis of Experience.” Prepared for the Federal Highway Administration and Federal Transit Administration, Washington, DC (September, 1993). • Minnesota Mining & Manufacturing Co., “3M Center Fact Sheet: Year-end 1995.” Maplewood, MN, http://www.mmm.com/profile/looking/ center.html (Webpage accessed July 9, 1998). • Bhatt, K., and Higgins, T., An Assessment of Travel Demand Management Approaches at Suburban Activity Centers: Final Report. Prepared by K. T. Analytics for the Transportation Systems Center, Cambridge, MA (July, 1989). • Certain interpre- tations added by Handbook authors. 5-43 Table 5-19 3M Vanpool Utility Ratio Breakdowns for 1974 Utility Ratio Average Pick-up Time Average Line Haul Distance Number of Vans Percentage of Vans 0.35-0.75 15.9 minutes 21.5 miles 10 19% 0.76-0.99 26.6 22.2 7 13 1.00-1.20 25.8 16.4 14 27 1.21-1.60 29.0 14.8 11 21 1.61-2.40 33.4 13.2 10 19

Golden Gate Vanpool Transportation Demonstration Project Situation. The Golden Gate Vanpool Transportation Demonstration Project grantee, the Golden Gate Bridge, Highway and Transportation District, was and is a multi-modal transportation agency that operates fixed route buses and passenger ferries, sponsors club buses, and controls the Golden Gate toll bridge. The project area is the congested corridor north of San Francisco, with an exclu- sive, toll-free HOV lane leading via the toll bridge toward the San Francisco employment center. The project was designed to test the feasibility of a public sector transportation agency promotion of vanpool group formation, and of “seeding” owner-operator vanpool groups via transition from initial third-party operation, after a 6 month introductory period. An overarching objective was to decrease vehicle demand on the bridge without requiring further expansion of the District’s deficit financed transit service. The demonstration ran for 33 months, October 1977 through June 1980. Results of the latter part of the project were impacted by the 1979 gasoline shortage and price increases, which had a positive effect on the demand for ridesharing. Analysis. Conclusions were based on preexisting data bases (more complete for corridor travel than for intra-suburbs travel), bridge vehicle and occupancy counts, vanpool application form data, initial (at time of joining) and supplementary vanpooler surveys, and on-board trip logs. The ini- tial 9 months were analyzed and reported in detail, with further conclusions developed at the end of the project. Actions/Results. A variety of methods were used in a promotional campaign launched at the beginning of the project to attract vanpooling applications. Toll booth handouts proved the most cost-effective at $11 per application generated, followed by bus handouts ($13), employer con- tacts ($17), and downtown street demonstrations ($17). The least cost-effective strategies were shopping center demonstrations ($100), fair booths (no applications), and the following approaches which cost over $200 per application (in increasing order of expense): take-one holders in public places, newspaper advertising, free rides, community meetings, and kiosks (3 kiosks for one appli- cation). Not measured were the effect of news releases and synergistic effects. Of the 1,350 appli- cants for vanpool membership in the first 9 months, 287 (21 percent) became active vanpoolers. The corresponding 33 month totals were 3,926 applicants with 804 (20 percent) becoming active vanpoolers. Half of all applicants submitted their applications following a telephone contact with staff, rather than in response to a specific marketing activity. Fifteen percent of all vanpoolers in project vanpools never went through the formalities of application submission, and another 20 per- cent did not submit applications because they were in project-assisted vans not furnished by the project. Many came through employer coordinators, driver efforts, and word of mouth. Driver incentives were a free commute and limited personal use of the van for 17.5 cents per mile (price as of 3/1/80). Thirty vanpools were formed in the first 9 months, with an average occu- pancy of 9.6 persons. Five of these were terminated because of inability to achieve full ridership (3 vans), inability to replace riders transferred to another work site, and end of a school year (State College destination). Luxury vans with airline type seats were initially in greater demand than bench seat vans, despite a 60 mile round trip monthly fare of $44 versus $36, but this preference dissipated later in the project. Initial demand split into two markets, the San Francisco commute (20 vans less 2 terminations) and intra-suburbs (10 vans less 3 terminations). In May 1978, of 40,400 inbound Golden Gate Bridge commuters between 6 and 10 AM, 59.4 percent used 1 or 2 occupant autos, 26.6 percent used public transit, 35 percent were in 3+ carpools, and 0.5 percent used project vanpools. Project vanpoolers constituted 0.1 percent of the intra-suburbs market. A socio-economic profile of May 1978 vanpoolers, with comparison to bus and ferry commuters, is given in Table 5-20. 5-44

By the end of the 33 months, 148 project and project-assisted vanpools had been formed. Their apportionment among markets served was more evenly distributed than for the initial vanpools, with 53 percent serving San Francisco employers, 40 percent traveling suburbs to suburbs, and 7 percent serving reverse commuting out from San Francisco. Reasons for the shift to suburbs-to- suburbs orientation included a change in marketing emphasis, greater public awareness, and a 53 percent increase in the cost of gasoline, which made vanpooling cost effective for shorter dis- tances. Of vanpools formed, 25 percent terminated prior to the end of the project. Vanpools suc- cessfully transitioned to other third-party or owner-operator status totaled 34 percent at the end of the project; 24 percent were still operating as project vanpools, and 17 percent were project- assisted vanpools that had never been formal project vanpools. At the conclusion of the demon- stration, the 111 operating vanpools were carrying 1,232 commuters (804 from applications, 428 accepted without formal application). Although nearly half of the early vanpoolers previously used public transit, Golden Gate transit bus ridership increased throughout the vanpool project. No direct transit service was available for the markets covered by 44 percent of the vanpools. Prior modes of vanpoolers at both the beginning and the end of the project were given in Table 5-15 of the “Related Information and Impacts” section. The average round trip for the San Francisco commute was initially 79 miles for bench seat vans and 93 miles for luxury vans. The corresponding intra-suburbs averages were 70–73 miles. The average round-trip distance steadily decreased as the program matured, gasoline prices increased, and the market shifted. In the initial 9 months, prior to the 1979 oil crisis, vanpooling was found to be always less expensive than one or two occupant auto commuting, less expensive than bus or three occupant carpool commuting for round trips of over 30 miles or so, and occasionally less expensive than five occupant carpool commuting. These cost comparisons take into account that the average vanpooler was found to ride only 4 out of 5 days, thus increasing the effective van- pool user cost. Travel time averaged only a minute longer than for prior modes, but former tran- sit riders saved an average of 9 minutes while former auto commuters added nearly 11 minutes. Thirteen percent of all vanpoolers sacrificed 20 minutes or more. Riders themselves ranked van- pooling faster than bus or club bus, slower than driving alone, and equivalent to a carpool. At the end of 9 months, survey results suggested that 8 percent of all vanpoolers had deferred replacing an auto, 7 percent had avoided buying an auto, 1 percent had sold a vehicle, and 4 percent planned to. 5-45 Table 5-20 Socio-Economic Characteristics of Golden Gate Vanpoolers, Bus Riders, and Ferry Passengers Socio-Economic Parameters (1978 Dollars) Marin County - San Francisco Bus Riders Larkspur Ferry Passengers Golden Gate Vanpoolers Income under $15,000 30% (sic) 24% 14% $ 15,000 - $24,999 31 (sic) 24 40 $25,500 or over 29 (sic) 52 45 Male/Female 63/37 73/27 63/37 No auto 0 2 0 1 auto 47 41 33 2 or more autos 53 57 67 Source: Dorosin, Fitzgerald, and Richard (1979), as presented in Pratt and Copple (1981).

More . . . Little progress was made during the initial months on the demonstration effort to tran- sition individual vanpool groups from the project’s vanpool incubation period, intended to be only 6 months, to owner-operator status. By the end of the project, however, 51 vanpools, 42 percent of all vanpools formed, had been transitioned into either owner-operator (or leased) vanpools, van- pools of the Bay Area RIDES third-party operation, or employer sponsored vanpools. It would appear that the transition was greatly facilitated by the 1979 oil crisis. In any case, the demonstra- tion project as a whole was found worthy, and was transitioned into a permanent Ridesharing Division within the Bridge District. Sources. Dorosin, E., Fitzgerald, P., and Richard, B., Golden Gate Vanpool Demonstration Project. Prepared by Crain & Associates, Inc. for the Urban Mass Transportation Administration, Wash- ington, DC (July, 1979). • Dorosin, E., Golden Gate Vanpool Transportation Project: Final Report. Pre- pared by Crain & Associates, Inc. for the Urban Mass Transportation Administration, Washington, DC (September, 1982). Connecticut’s Easy Street® Vanpool Program Situation. The Connecticut Department of Transportation and the non-profit rideshare brokerage The Rideshare Company, serving greater Hartford and Eastern Connecticut, operate the vanpool program Easy Street®. The state subsidizes the commuter service, which is available for trips beginning or ending in Connecticut. Easy Street® is a new mode of operation for The Rideshare Company, developed in response to ridership losses in the mid-1990s. Business relocation to the suburbs, workforce reductions, and policy changes that led companies away from subsidizing alternative transportation all combined to cause a reduction in Rideshare Company vanpools from a high of 200 in 1993 to 155 in the fall of 1995. Actions. The Easy Street® repackaging of The Rideshare Company’s vanpool operations was imple- mented in October 1995. Instead of anonymous white, Easy Street® vans each have a green, yellow, and purple decal brand design on a white background, along with the toll free number to attract potential riders. An automated voice mail system offers detailed information. Easy Street® provides a free commute to the driver along with 40 free personal miles monthly. Easy Street® takes care of maintenance, gasoline charges, and 24 hour roadside assistance with no out-of-pocket expense to the driver. Fares are structured to cover costs. The service includes a guaranteed ride home program. Easy Street® offers predictable prices by setting the fares across the board, based on round trip mileage. Calculated in 5 mile increments, the fares range from $70 to $100, with vacation rebates. Part time and daily fares are also offered. TransitChek vouchers, purchased and to varying degrees subsidized by employers, can be used to help pay the fare. The state of Connecticut pays for about a third of an employer’s TransitChek voucher’s cost and also subsidizes empty seats in new van- pools. The minimum group to start or continue a van is eight passengers and a driver. A sliding scale for 4 to 6 months determines a gradually decreasing subsidy. Analysis. The Easy Street® vanpool program has not been analyzed in depth. Instead, the avail- able information has been culled from brochures, newsletters, briefs, and the Internet, as indicated under “Sources.” The reported mileage and air quality reduction benefits were apparently esti- mated assuming that all vanpool passengers would otherwise be driving alone, and thus may well be overstated. Results. Phone calls inquiring about vanpool service increased from 74 in October 1995 to 143 in January 1996. After The Rideshare Company’s count of 155 vans in the fall of 1995, Easy Street® 5-46

vans in service increased to 163 in January 1996, 183 in March, and 203 in November. November 1996 riders totaled 1,787, up by nearly 400 riders from Easy Street’s® inception. The vanpool fleet was reported to have been driven 3.5 million miles in calendar year 1996. Easy Street® was recog- nized as one of the U.S. EPA’s “Transportation Partners: 1997 Way to Go! Award Winners” for reduction of vehicle traffic while preserving or enhancing transportation choices and quality of life in the community. The vehicle traffic kept off of Connecticut highways was estimated at 28 mil- lion single occupant miles. The corresponding air quality emissions savings were reported as 10,670 tons of Carbon Dioxide, 55 tons of Carbon Monoxide, 3 tons of nitrogen oxides, and 4 tons of hydrocarbons. More . . . The ridership increases are attributed in large measure to the improved visibility that the self-promoting branded vans provide. Even the decals themselves received national praise—Easy Street® won the annual Commercial Fleet Graphics Contest sponsored by the Commercial Carrier Journal and the National Private Truck Council. Note that 2001–2002 statistics and service infor- mation on Easy Street® operations are provided within the “Response to Vanpool and Buspool Programs” section under “Third-Party Vanpool Programs”—“Third-Party Vanpool Program Evolution.” Sources. The Rideshare Company, “Easy Street® Is The Convenient New Van Service Available In Your Area.” 2Plus, Cary, NC, http://www.easystreet.org (Webpages accessed July 8, 1998). • Metropool, Inc., “Vanpooling Proves Test of Time.” Commuter Connections. Vol. 7, No. 4. http:// www.metropool.com (Web document dated 1997). • 2Plus, “Easy Street® commuter service grows, attracts new riders.” Commuters’ Register, Connecticut Edition, Vol. 2, No. 3. Cary, NC (March, 1996a). • 2Plus, “Easy Street® system wins national award.” Commuters’ Register, Connecticut Edition, Vol. 2, No. 10. Cary, NC (October, 1996b). • 2Plus, “Branding boosts ridership to new high.” Commuters’ Register, Connecticut Edition, Vol. 3, No. 5. Cary, NC (May, 1997). • Renew America, “EPA’s Trans- portation Partners: 1997 Way To Go! Award Winners.” Washington, DC, http://www.crest.org/ environment/renew_america/wtgo97.html (Webpages accessed July 16, 1998). Pace Vanpool and Subscription Bus Programs in Suburban Chicago Situation. Pace, the Chicago Regional Transportation Authority’s suburban bus division, provides service to an area of six counties and 264 municipalities that is nearly the size of Connecticut. The population and employment have grown to substantially exceed those of Chicago. This 3,446 square mile suburban area had a 1990 population of 4,454,300 and employment of 2,163,600, including 40 percent of the Chicago region’s office space. Approximately 48 percent of Pace fixed route bus riders are making suburb to suburb trips. However, of the more than 55 million square feet of office space built in the suburbs since 1975, the majority is poorly accessible to transit patrons. One of the biggest challenges faced by Pace was serving the 5,000 employee Sears Merchandise Group during and following its 35-mile relocation in November 1992 from the Sears Tower in downtown Chicago to Hoffman Estates on the fringe of suburbia. Actions. To serve small groups of commuters in the diverse and changing suburban market, Pace in 1991 established what it calls its vanpool incentive program (VIP). The VIP service provides pas- senger vans to groups of 5 to 15 people. Vanpools may be initiated through an employer or inde- pendently. Pace plans the route, provides the van and insurance, pays for fuel and maintenance, sets the fare, bills riders individually, and offers a Guaranteed Ride Home. Vanpool drivers ride free and get up to 300 personal use miles per month. The monthly fare is calculated rider by rider, based on mileage increments, ranging, in January 1998, from $47 for a 14-passenger vanpool rider with 20 round trip miles or less, to $126 for a 4-passenger vanpool rider traveling 131 to 140 miles 5-47

($55 for a 14-passenger vanpool rider). A Commuter Club Card may be requested, which is effec- tively a Pace fixed route bus pass. For a surcharge, a Pace/CTA Universal Monthly Pass may be obtained. These passes facilitate use of connecting buses, and also rail rapid transit in the case of the universal pass. Employers may subsidize vanpools through Pace’s Transit Check program. Around 1994, the vanpool program was expanded to add ADvAntage vans intended to provide a transit alternative to individuals with disabilities who commute regularly to either worksites or rehabilitative workshops. The ADvAntage vans are available to human services organizations, workshops, and agencies providing such work-related transportation services. The van driver must be an employee of the human service entity or a relative of a rider. The monthly fee as of 1999 for an entire van was $325 for minivans, or for any available van type if more than half the trips served are made by ADA certified clients; otherwise the fee was $650 per month. The Pace Suburban Bus Service worked closely with Sears for 3 years prior to their 1992 move on employee surveys and the development of transportation alternatives. A mix of fixed route ser- vices, subscription bus services, and vanpools was designed. The fixed routes connected with the pre-existing transit network, with service levels ranging from 2 to 9 trips in both the morning and evening peak periods. Subscription bus service was designed for areas with a significant concen- tration of Sears employees, but no suitable fixed route service. A minimum of 30 passengers was required for a buspool. Ten routes were established using thirteen motorcoaches operated by pri- vate contractors. Each route served a park-and-ride lot an hour or more from the worksite, and charged a monthly fare of $75 to $94 (later standardized at $80). Smaller groups of employees were offered the option of vanpools, with preferential parking at the employment site. Initially 44 van- pools were formed. Analysis. Pace vanpool and buspool fleet and usage statistics have been culled from reports, papers, fact sheets, and the Internet, as indicated under “Sources.” Information on Pace VIP pro- gram participants, their travel and opinions, and support provided by their employers, was obtained in a Spring 1993 survey. This self-administered survey was sent to 671 riders and drivers. A 48 percent return was obtained, 87 percent from riders and back-up drivers, and 13 percent from primary drivers. Results. Officially reported Pace VIP vans in maximum service for 1994 through 1997 were 162, 205, 231, and 291, respectively. Some 80 to 90 percent were serving the suburbs-to-suburbs market in 1994–96, with the remainder serving the city-to-suburbs reverse commute. Vanpooler unlinked trips grew from 558,100 in 1994 to 969,900 in 1996. The corresponding weekday vanpool loadings grew from 6.9 to 8.6 passengers, including driver, per vanpool. Trip length, relatively stable over the 1994 through 1996 period, averaged 38.6 miles one way or 77.1 miles round trip. Although the vanpool program is structured to achieve an 80 percent cost recovery ratio minimum, it has in prac- tice typically achieved over 100 percent, exclusive of ADvAntage vans. (Additional cost recovery detail is provided in the “Revenue/Cost Considerations” section under “Related Information and Impacts.”) Compared to a 17 percent recovery ratio minimum performance standard for fixed routes, and 35 percent for the system as a whole, Pace subscription bus services are required to maintain a 60 percent recovery ratio. ADvAntage vans totaled approximately 20 circa 1994, and 55 in 1996. The mix of fixed route, subscription bus, and vanpool service provided to the relocated Sears Merchandise Group employees was initially successful in retaining a 30 to 35 percent transit and paratransit share, compared to 92 percent at the Sears Tower site. The fixed plus subscription route recovery ratio was 36.6 percent for 1992, not counting a short-term subsidy provided by Sears, which raised it to 55 percent. In January 1993, Sears transit and paratransit use was roughly equally 5-48