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.
56 The literature review in chapter two synthesized information from a great number of sources to provide information on the results of fare-free demonstrations or feasibility analyses conducted by public transit agencies. This appendix provides a summary of many of the reports that were used to produce the literature review. Many of these sources were developed specifically for particular transit agencies and would not be available through normal research channels. Project panel members and members of listservs were able to provide infor- mation that led to the identification of such reports. This bibliography summarizes reports and articles that describe the results of: â¢ Fare-free demonstrations that were discontinued â¢ Studies analyzing the feasibility of instituting fare-free public transit â¢ Reports on transit agencies retaining totally fare-free policies â¢ Other pertinent research addressing fare-free public transit policies. RepoRts on FaRe-FRee DemonstRations that weRe DiscontinueD The literature search revealed that a number of public transit agen- cies other than those identified in this report as currently providing fare-free service have considered instituting a similar operational strategy, but discontinued after experimenting with, or analyzing the feasibility of, implementing such a fare policy. Provided here are separate summaries of reports of these various experiments since they also attempted to provide information on the implemen- tation and outcomes of fare-free transit in their communities. mercer county, new Jersey âThe Fare-Free Transit Experiments,â written in 1982 by A. H. Studenmund and D. Connor, chronicles the results of experiments that began in March 1978 (19). The Urban Mass Transportation Adminis- tration (UMTA) funded partial fare-free demonstrations for the cit- ies of Denver, Colorado, and Mercer County (Trenton), New Jersey, to determine the effectiveness of removing fares for one year on a restricted basis. In the Mercer County demonstration, no fares were charged during the off-peak time period for one year (the off-peak fare was $0.15). Peak period fares of $0.30 remained unchanged. The off-peak time was selected due to unused capacity and low marginal costs of off-peak service. While the demonstration was conducted in both Denver and Mercer County, the Denver demonstration was con- founded by several problems (e.g., lack of pre-demonstration data, a change in fare-free hours, and major route restructuring) and thus the results were not conclusive. However, it was noted that the results found were similar to the results in Mercer County. The fare-free demonstration in Mercer County, with a service area population of approximately 300,000, led to a significant increase in ridership during the off-peak periods, with a 25% to 30% increase attributed to the removal of the fare. The demonstration attracted approximately 2,000 net new riders per day to public transit. A sig- nificant shift to public transit was experienced as 69% of the new trips were previously made by other modes. Of these trips, about 50% were previously made by automobile, and one-third previously walked. It was estimated that the fare-free off-peak public transit ser- vice demonstration reduced private vehicle-miles traveled (VMT) by 30,000 miles per week. The report noted that given that the typi- cal VMT in Mercer County was 21 million miles per week, this reduction was not regarded as statistically significant in terms of reducing congestion. The number of additional trips made by young people (less than 25 years of age) increased disproportionally to the demographic makeup of the ridership before the fare was removed. Other demo- graphic groups were identified, but no other group had a dispro- portionate increase in ridership during the fare-free demonstration. The total cost of the fare-free demonstration was $339,000, including direct loss of fare box revenue, some of which was caused by ridership shifts from the shoulders of the peak to off-peak hours of service. Another cost associated with the removal of fares dur- ing the off-peak period was the need for more capacity, costing $10,000. The amount of money saved from not needing to collect fares during the off-peak hours was estimated to be $10,000. Other potential sources to partially replace the lost funding were identi- fied, such as increased sales tax revenue owing to increased retail spending, but these sources were not quantified. Aside from the loss of revenue, several issues were identified that resulted from removing fares during off-peak hours. First, between 5% and 15% of buses entering the downtown were found to be over- crowded during the off-peak hours. Second, because of the increased demand, the bus needed to stop more frequently and dwell longer at individual stops. This led to a decrease in on-time performance, with the number of late buses increasing from 25% to 45%. Third, the increased number of riders, particularly the younger riders, led to an increase in the number of situations where rowdy passengers were bothering other passengers. These issues increased the frus- tration level of the bus operators, 92% of whom reported that their job was less enjoyable. Fourth, the increase in young riders also led to increased complaints from downtown merchants about loitering and shoplifting. The report did not provide an estimate of what it might have cost to correct the problems with on-time performance, overcrowding, or controlling passenger behavior. The authors concluded that continuing fare-free public transit in Mercer County (and Denver) would not be advisable. In spite of the dramatic increase in ridership, the authors believed that the level of fares did not seem to be the impeding factor for increased mobility. One recommendation of the report was to use fare-free public transit as a temporary promotional technique for increasing long-term public transit ridership. It was found that even after fares were increased back to normal levels, ridership remained somewhat higher than expected. The report suggested that by removing fares for a short duration, it is possible that new riders may be attracted who will continue to use the system. Denver, colorado The report, Evaluation of the Denver RTD Off-Peak Fare Free Demonstration was produced by De Leuw Cather and Company in 1980 (46). The fare-free demonstration began on February 1, 1978, appenDix c annotated Bibliography
57 and continued for 12 months, ending January 31, 1979. Denver, with a population of 1.5 million in 1980, was the largest city to have experimented with a fare-free policy. The off-peak fare of $0.25 was eliminated, while the $0.50 cent fare was retained during just two hours in the morning and two hours in the afternoon repre- senting the peak commute hours. Ridership increased 49% system- wide and 52% during the off-peak, although additional service was added and many routes were restructured at the same time. The report found that the efficiency of the system, measured by cost per passenger, was substantially improved. The negative results included overcrowded buses, decreased schedule reliability, and obnoxious behavior from some passengers. The morale of drivers also declined. The principal conclusion of this evaluation is that free-fare public transit may be a more effective short-term market- ing instrument than a desirable permanent element of transportation policy for major metropolitan areas. Reduced or low fare off-peak public transit might achieve many of the same beneficial objectives of no fares, but complete removal of the fare barriers in a major met- ropolitan area appeared to generate enough undesirable side effects to undermine its overall effectiveness. topeka, Kansas A report entitled No Pay May: Project Description, Analysis of Ridership Data, and Survey Results was written in 1988 to provide information on a one-month fare-free experiment in Topeka, Kan- sas (18). This report describes the planning, implementation, and impacts of a marketing project undertaken by the Topeka Metropol- itan Transit Authority, a small urban system. Oil overcharge funds were used to pay for a month of fare-free public transit service dur- ing May 1988. Ridership increased by 83% during this short-term promotion, and a permanent ridership increase of approximately 6% held when fares were reinstituted. austin, texas This project could not locate a definitive document that reported the results of the fare-free experiment conducted by the Capital Metro Transportation Authority between October 1989 and December 1990. The report Final Report from the Free-Fare Telephone Survey written by the NSI Research group was reviewed which attempted to document peopleâs awareness and attitude regarding the fare- free program (34). As opposed to the demonstrations in Denver and Mercer County, the fare-free program in Austin, a rapidly growing city of 500,000 at the time, was available all hours of every day, not just during off-peak hours. There are conflicting interpretations of just how much ridership increased because of the fare policy. In a phone interview on June 28, 2011, with the then general manager of Capitol Metro, it was indicated that the policy was put in place because it was a newly created agency that was looking to pro- mote ridership. It was also in the favorable position of having more funds than the agency needed to operate based on revenues from a one-cent general sales tax, and the agency wanted to provide a ben- efit to the community for what they were paying. Not long before the fare-free program was instituted, Capital Metro began providing service to the University of Texas, which in turn began its universal pass program for students. Hence, it is difficult to know how much of the 70% increase in ridership was a result of the elimination of fares versus natural growth supplemented by university student use. Ironically, in the telephone interviews conducted by NSI, it was discovered that the level of fares was not among the top five pri- orities of passengers. On-board safety, on-time performance, con- venience of routes, cleanliness inside the buses, and frequency of service were the most important factors for riding the bus for both the general public and fare-free riders. Interestingly, the cost of ser- vice (the fare) was the second to the last most important factor. The telephone survey found that 12% of riders had discontinued using the bus since the fare-free policy was established, while 29% indi- cated they were using it less. Very few of those survey respondents reported that they stopped using it because of the fare-free policy. Most said they were able to buy a car or had moved. However, this particular experiment received much negative press about rowdy behavior on the buses, and the belief that the program was con- tributing to school truancy. Bus drivers were not supportive of the fare-free program; 215 drivers petitioned for an end to the fare-free service. The former general manager stated that ridership could have been increased through many different means, and offering fare-free transit was an artificial way to increase ridership given that many of the new riders were high school students who pre- ferred Capitol Metroâs buses to school buses and ended up joyrid- ing. However, he believed that the fare-free experiment might have contributed to a ridership gain that was sustained after the policy was ended. asheville, north carolina The Transit Staff Report, Fare-Free Transit, produced in February 2007, provides an overview of the benefits gained, costs, and lessons learned from a fare-free promotion in Asheville, North Carolina, a city of approximately 70,000 people (25). For 90 days from August to November in 2006, no fares were charged on any of their 19 buses. The goal of this promotion was to increase ridership on the exist- ing bus service, particularly from those who would otherwise drive a private vehicle. Data on ridership, wear on the vehicles, crime and rowdy customers, and service reliability were collected, in addition to rider surveys undertaken before and during the promotion. A $12,000 budget was provided for advertising the promotion to the public, and another $12,000 was spent on increased security services. The report indicated that ridership increased during the promo- tion by 58.5% over the same time frame from the year before. The efficiency of the bus system improved dramatically as the cost per trip declined by 14%. Ridership increased by 137,000 during the three months of the promotion over and above the total of 245,000 passengers that had been transported during the same three months of the previous year, with the vast majority of the increase resulting from the fare-free promotion. After the promotion was terminated, ridership continued to be about 17% higher than the same period of the previous year. Approximately 9% was attributed to the pro- motion. The increase appeared to be the result of lower-income individuals who were more sensitive to cost. The number of riders who owned a car remained the same both before and during the promotion. The number of riders making less than $10,000 per year increased by 7.5%, while the number of riders making more than $10,000 per year decreased by 7.5%. In particular, the promotion was deemed very beneficial to eve- ning service, where utilization increased after the promotion by 82%. Evening service carried nearly half as many trips per operating hour as daytime service, which was well beyond expectations. The loss of fare box revenue during the 90-day promotion period was approximately $97,000 for regular bus service and $13,125 for paratransit service. Non-financial costs included increased travel times and a decrease in on-time performance. The major complaint of riders during the start of the fare-free promotion was poor reli- ability due to the increased passenger loads and required stops. Travel time was estimated to have increased by several minutes per hour owing to the increased number of stops and longer dwell times
58 associated with increased ridership. Situations involving rowdy cus- tomers also increased during this fare-free promotion. Complaints about such customers accounted for 21% of all complaints made during the first half of the fare-free promotion. These complaints decreased towards the end of the promotion, which the report attri- butes to passengers adjusting to the new riders. Among public transit agency personnel, bus operators reported higher rates of verbal abuse, greater pressure to maintain schedule, difficulty with managing overcrowded buses, and safety concerns for disabled passengers. Maintenance personnel believed they were working beyond capacity just to keep the fleet running, and the buses were dirtier than normal. This well-documented report concluded with several âlessons learned.â First, fare-free service affected on-time performance. During the demonstration, on-time performance fell to 89%, but rebounded to 97.7% within three months of the conclusion of the demonstration. Second, fare-free service added passengers who tended to be younger and noisier than previous passengers. It is likely that the more disruptive passengers discouraged those with other options to ride the bus. To discourage abuse of the system by dis- ruptive individuals, it was recommended that a small fare (between $0.15 and $0.25) be charged in future promotions, rather than pro- viding completely fare-free service. Third, they found the fare-free promotion most likely resulted in greater long-term ridership. It was noted that Topeka, Kansas, had similar results of long-term rider- ship increases of almost 6% after ridership increased 83% during a one-month fare-free experiment. Fourth, they concluded that there is a pent-up demand for mobility, particularly among low-income and younger people, especially students for whom mobility costs are a financial burden. Finally, despite the lack of any fare, many potential âchoiceâ riders will still not ride the bus. For these people, service quality and reliability is a greater factor than the cost of a fare. The report provided no theory on why, but noted that demand for parking in the downtown area actually increased by 9.1% dur- ing the fare-free demonstration, higher than previous annual rate increases of 3.3%. After the fare-free demonstration, cash fares were increased from the pre-demonstration fare of $0.75 to $1.00; however, monthly passes were reduced from $30 to $15. An Annual Unlimited pass was introduced for $120, and made available for $60 to seniors and people with disabilities. milton, canada The city of Milton, Canada, near Toronto, became the first munici- pality in Canada to provide fare-free service for an extended period of time. In 2007, public transit during the midday off-peak time (9:00 a.m.to 3:00 p.m.) was made free for all users for seven months. To compensate for the foregone fare box revenue, two corporate sponsors were found to fund the project (Mattamy Homes Ltd. and Fieldgate Developments). In 2008, city staff prepared a white paper, Fare-Free Transit Pilot ProjectâFinal Report (26). The focus of the empirical analy- sis was on ridership. In each month, ridership increased more than would have been expected year-to-year had the fares remained. Rider- ship increased an average of 63% over the seven months of the experiment. One interesting topic noted was the increase in rider- ship during the summer months, when it typically declines. This was attributed to the novelty of the project (which began in June) and students on summer break suddenly having a free mobility option. A further increase seen in the fall season was attributed to the change in secondary school schedules. The new end time for school was 2:30 p.m., which falls at the tail end of the free-fare pub- lic transit service, although no further study to identify the impact of changing the school end time is provided in the report. Staff reported that a potential benefit of the fare-free program was increased ridership after fares were reintroduced. During the fare-free promotion, on-board surveys were conducted to assess rider behavior. This report provides some insight into how riders were using the system during the demonstration, and how they planned to use the system after the reintroduction of fares. No distinction was made between new riders and riders who previously used public transit. Survey results indicated that of the 80% of riders who used the bus at least two times per week during the fare-free program, 86% would continue after fares were reintroduced. On the other hand, they found that while 67% of senior respondents were fre- quent users during the fare-free promotion, only 33% indicated they would continue to be so after fares were reintroduced. The report suggested that this implied that senior citizens are more sensitive to cost increases. No information on lost fare box revenue or project costs were provided in the report. The city of Milton did not experience any loss in revenue from the project, due to corporate sponsorship. There was some concern with potential disruptive behavior aboard the public transit vehicles during the fare-free program, par- ticularly from secondary school students who had the opportunity to ride for free at the end of the midday off-peak period. To address the issue, new policies were created to allow bus drivers to restrict passengers for safety reasons. Customer satisfaction was rated very high during the fare-free program, based on on-board surveys dis- tributed during the program showing that 99% of all respondents were either âsatisfiedâ or âvery satisfiedâ with the program. Several âlessons learnedâ are provided in the reportâs conclu- sion. First, eliminating fares should only be one aspect of any pro- gram to increase ridership. Other factors such as convenience, travel times, comfort, and other service elements should also be consid- ered. Second, ridership growth and crowding should be anticipated from the beginning, particularly due to the costs of keeping up with demand. A proactive approach can help to avoid overcapacity buses and nuisance riders, although no specific methods for dealing with overcrowding and nuisance riders were provided in the report. Third, although the off-peak period of 9:00 a.m. to 3:00 p.m. was selected to promote ridership during under-utilized times, problems can arise when targeting specific segments of ridership. Some transit riders, specifically those who used transit only during peak periods, felt that it was unfair to not be included in the fare-free program. Finally, there was some negative perception of the corporate sponsorship of the fare-free program, regarding the motives of the corporations funding municipal services. However, without their sponsorship there would not have been sufficient funding to conduct the demonstration. stuDies analyzing the FeasiBility oF instituting FaRe-FRee puBlic tRansit A number of transit agencies have explored the concept of institut- ing fare-free transit, but after careful reviews decided against going forward. lane transit District, eugene, oregon A white paper entitled Fare-Free Service at Lane Transit District: An Overview of Financial and Operational Impacts, prepared by the staff of Lane Transit District (LTD) in January 2008, developed
59 an estimation of the impacts that the District would experience if a fare-free policy was implemented (7). Lane Transit carries approxi- mately 10 million passengers a year. It is located in Eugene, Ore- gon, which has a population of approximately 350,000 in its service area. Eugene is also the home of the University of Oregon. The report briefly notes the potential positive results of going fare-free, including increased ridership, decreased traffic congestion, filling âempty buses,â eliminating the costs associated with fare collec- tion, cultivating a culture of transit use among young people, and the encouragement of urban development and redevelopment. The report cites previous research that identified the factors that con- tribute to positive results, such as the size of the community, the degree of commitment to a fare-free policy by the community and the transit agency personnel, and the need to prepare for likely con- sequences of overcrowding and rowdiness. The staff report estimated that farebox cash, prepaid tokens and passes, and group pass contracts provided the agency with more than $5 million in revenue per year. A $5 million loss in revenue would be minimally offset by eliminating the cost of fare collection, estimated at $100,000 to $500,000 per year. LTDâs fairly simple fare collec- tion system further reduced potential savings from going fare-free. No employees focus entirely on fare collection, rather they have several duties. Consequently, eliminating fares would not necessarily allow for the elimination of jobs. Generously assuming $1 million might be saved in total administrative and marketing costs, the $4 million net loss in revenue would require a reduction of 20% of bus service hours. It was noted by staff that removing fares for the whole system could significantly increase the cost of required paratransit service owing to increased demand. This was not a consideration in the major fare-free experiments in Denver, Mercer County, or Austin, which were conducted prior to Americans with Disabilities Act (ADA) requirements. The maximum allowable price for a paratran- sit trip, per the ADA, is double the base cash fare. Removing fares could potentially increase the number of expensive paratransit trips per year that typically cost LTD $23.50 for a one-way trip. This report estimates a cost of $700,000 per year per 100 new riders. The potential increase in paratransit costs was not factored into the previous $4 million per year cost for going fare-free. Several potential consequences are identified when systems pro- vide fare-free service based on larger-scale public transit systems that tried a fare-free approach. Increased vandalism, vagrancy, dis- ruptive passengers, and overcrowding can adversely affect choice passenger ridership. Smaller systems will not necessarily run into these issues when going fare-free as Lane Transit discovered when discussing the fare-free policy with staff from Island Transit in Coupeville, Washington, who have not experienced these conse- quences. The report hypothesizes that this may be the result of less crowded buses, fewer homeless people in the community, and a stronger community culture that values public transit more highly. LTD staff decided not to recommend removing fares from its buses, citing an already high use of their public transit system with current fares. LTDâs report stated that the overcrowding that it was experiencing and the difficulties it was having with making connec- tions made implementing a fare-free service inappropriate, given that such a policy would increase demand (which it did not quan- tify). They concluded that removing fares would not attract enough additional riders to warrant the loss of revenue. portland, oregon The report, Fareless Transit in the Portland Metropolitan Region was completed in 1999 by the Fareless Transit System Research Work Group to research the option of providing fare-free public transit in Portland (9). Portlandâs Mayor Katz tasked the group, made up of eight local business and non-profit representatives, with studying the feasibility, benefits, costs, and challenges associated with fare- free transit. The reportâs production was funded by Tri-Met, which also made its consultants available to conduct research. Several factors that affect public transit ridership were identified and explored. These factors are ranked in order of impact, with fare costs ranked last after other factors such as reliability and safety. The typical elasticity for fare cost and ridership is provided (-0.3), with a few notes to keep in mind. First, surveys of riders generally rank fare cost lower than empirical studies of fare elasticities would suggest. Differences in off-peak and peak ridership were also briefly explored. Previous studies noted in TCRP Report 95 indicated that off-peak public transit ridership is more sensitive to changes in fare than peak ridership (6% versus a 2% increase given a 10% decrease in fare). However, the report noted that a fixed percent shift in peak ridership will be larger in magnitude than the same percent shift in off-peak ridership. Several case studies were identified in the report including those most frequently cited: Austin, Texas; Mercer County, New Jersey; and Denver, Colorado. Several take-away points were provided based on those demonstrations. First, all of the programs resulted in substantial ridership increases, with increases as high as 75% in Austin. Second, most of the ridership increases were experienced during off-peak hours. (Authorâs note: It is to be expected that the increases occurred in the off-peak hours in Trenton and Denver because that was when fares were not required.) Third, the percent- age of new trips that had been made from people changing from private vehicles was notable (30% or less), but not as large as agencies might have hoped for. Third, there were some passenger complaints of overcrowding, slower travel times, and reduced reli- ability. Lastly, bus operators did not respond favorably to the fare- free programs, citing concerns over rowdy customers and passenger complaints. The authors concluded that because of Portlandâs well- developed public transit service, ridership gains and other effects would not be as great as those from previous case studies. However, the authors noted that many of the negative side effects would occur for any major increase in ridership and should be addressed before implementation. Fare-free public transit should not be considered as an isolated strategy, but as a part of a more comprehensive and balanced plan and set of actions to increase ridership. Several implementation issues were identified and addressed in this report. From a hypothetical situation proposed by Tri-Met and reviewed by ParsonsâBrinkerhoff, a fare-free system would increase ridership by 25% during peak hours and 60% to 65% during off-peak hours. The proposed plan would cost Tri-Met $54 mil- lion, not including planned/required service improvements. One proposed plan to fund fare-free public transit, as suggested by Mayor Katz, was to implement a regional parking tax. The legal, institutional, and economic challenges are discussed in the report. Other possible options, rather than fare-free public transit, were also considered, such as a simplified fare structure that would do away with zone fares. Another option would be to expand Tri-Metâs pass programs to serve not just the private employers currently par- ticipating, but cities, counties, and other groups banded together in transportation management associations. Such programs would increase ridership while reducing costs to passengers and reducing dwell time at stops. The authors of the report concluded that before eliminating fares, service should be expanded to improve capacity and service quality. Simply making public transit fare-free is not enough by itself to entice a significant number of people who use their private
60 vehicles. One method that the report concluded should be consid- ered is a regional parking tax not only to fund public transit, but also to encourage drivers to understand the true cost of driving. The report noted that this strategy has some significant hurdles both legally and institutionally. san Francisco, california In January 2008, Sharon Greene and other subcontractors produced a detailed report entitled Fare-Free Muni System Feasibility Analy- sis in response to Mayor Newsomâs request to analyze the costs and benefits to San Francisco if fares were eliminated on the San Fran- cisco Municipal Railway (the cityâs public transit system known as Muni), which is the eighth largest transit system in the United States (8). The mayor charged the study team with determining how much ridership would increase, identifying key risk areas, what additional operational and maintenance costs would be required, what opera- tional and maintenance savings would be realized, and what policy issues would need to be addressed. Three different scenarios were considered to develop a range of potential costs, based on ridership increases of 18%, 48%, and 78%. These scenarios were based on the results taken from the literature that was reviewed. A San Fran- cisco Municipal Transportation Agency travel demand model was completed for the scenario in which all fares were reduced from $1.50 to $0.00. A ridership increase of 35% to 40% was projected from this modeling effort. A review of the literature was included to identify the results and lessons learned from other U.S. public transit systems that have attempted fare-free service. From the literature, the authors reported that smaller agencies realized the best results from fare-free public transit. These smaller agencies tended to have more under-utilized capacity, minimizing the need for additional equipment and service hours to meet increased demand. Additionally, smaller systems that recover less than 10% of their operating costs from fare box rev- enue did not experience significant budget shortfalls in their oper- ating budgets from the elimination of fares. The report noted that the cities that implemented fare-free public transit demonstrations showed ridership increases of from 13% to 75%. The operational and capital impacts from the different estimates of projected increases in ridership attributable to fare-free transit were developed (see Table C1). In addition, more facility capacity would be required to store and maintain these vehicles as all existing facilities were at or above capacity. If no fares were charged on Muni services, there would be a savings of $8.4 million in annual operations and maintenance (O&M) costs and a reduction in staff of 91 full-time employees currently needed to collect and account for fares. However, the loss in fare revenue would be approximately $111.9 million. In each of the three scenarios, the additional operations and maintenance costs resulting from additional operators, mechanics, and security to serve additional passengers would be far greater than the saved costs from the discontinuance of fare collection. Paratransit costs would increase by $1 million to $4.6 million owing to increased demand for free service. In all, the net O&M cost for operating fare-free public transit (using the middle and most likely scenario of a 48% increase in ridership) would be $184 million after accounting for the loss in fare revenue of $111.9 million. This figure includes additional needed capacity, a new central control facility, new maintenance facilities, and the cost savings from not implementing what would be unnecessary projects (e.g., upgrading ticket vending machines). In addition, the net capital costs to implement fare-free service while meeting a 48% increase in ridership demand was estimated to be $519 million. Several policy issues were identified. First, while the San Francisco Municipal Transportation Agency (SFMTA) Board has the authority to eliminate fares, it was recommended that support from local and regional policy groups be confirmed. It was also noted that a public vote would be necessary because new sources of funding would be required. This study did not identify or evaluate potential supplemental revenue streams as this was a primary objec- tive of the Mayorâs Blue Ribbon Revenue Panel. Second, SFMTA is a lead agency in universal regional smart cards (TransLinkÂ®). If fares were not required on Muni, it would be more difficult to justify promoting smart cards in the region, although the report suggested that the cards could be used to count passengers. Third, at the rate it was taking to acquire vehicles, procurement of the needed vehicles and facilities to meet projected ridership increases from fare-free public transit would take 5 to 10 years. (Authorâs note: Given the structural deficits of the SFMTA and the weakened California econ- omy, it would appear highly unlikely that the agency would be able to generate the type of revenues called for in this report.) hamilton, canada A report entitled Free Transit, Deep Discount Fare Policy and Other Strategies Employed to Create a High Ridership Transit System was prepared by Scott Steward, General Manager of the Hamilton Public Works Department in June 2008 (13). Hamilton is Canadaâs ninth largest city with a population of approximately 500,000. This report details the results of a study to investigate the feasibility and impacts of providing fare-free public transit to all citizens of Hamilton. The report reviews both fare-free public tran- sit and deep discount fare alternatives. Percentage Ridership Increase Required Additional Buses Required Additional Street Cars Required Additional Rail Vehicles Required Additional Operators Additional Annual Operating Expenses 18% 41 11 37 59 $23 million 48% 157 20 90 234 $69 million 78% 283 30 138 420 $139 million TABLE C1 PROJECTED COSTS ASSOCIATED WITH VARyING LEVELS OF INCREASES IN RIDERSHIP AT MUNI
61 It was estimated that fare-free public transit in Hamilton would increase ridership by approximately 20% to 50%, based on the review of the results from experiments in the United States, as well as fare-free programs that have been retained such as in Chapel Hill, North Carolina. This would translate to an increase from 48 rides per capita to 55 to 70 rides per capita per year in Hamilton. Initially, most of the new ridership would be a result of existing riders trav- eling more, rather than a mode shift from private vehicles. Traffic congestion is not severe, nor is there a shortage of parking in the city. It was concluded that fare-free public transit provides the best opportunity to meet the cityâs Vision 2020 goal of 100 rides per cap- ita. It was noted that free transit will likely attract riders from other modes (e.g. walking, biking, or taxi); however, with no Canadian system-wide experience, it was difficult to determine how much. The public transit system in Hamilton provides 21.2 million annual trips, with revenue of $28.2 million and an average fare of $1.33. The costs to the city for public transit resulting from an increase in ridership of 20% and the loss of all farebox revenue would be approximately $30.9 million, requiring an additional tax of $161 per year per household. This report noted that ridership alone cannot be the only metric for success. Other public policy goals include: â¢ Air quality improvement â¢ Energy conservation â¢ Congestion reduction â¢ Provision of mobility to the transportation-disadvantaged â¢ Access to jobs â¢ Promotion of economic development â¢ Promotion of livable communities. The report also emphasizes that goals of higher ridership and higher revenues are somewhat contradictory, causing concern for public transit agencies expected to do both. The report also noted the large number of external factors out of public transitâs control that make attracting riders from private vehicles difficult (e.g., low densities, high sprawl, low traffic congestion, high incomes, low parking costs, low population growth, etc.). Public transit agencies do not have unlimited funding available, which requires a balance between costs and improvements (i.e., ridership increases). No empirical results are available from this report. Hamilton had not yet instituted fare-free public transit. One very brief case study from Chapel Hill, North Carolina, is provided in the report. Fare- free transit began in Chapel Hill in January 2002. Between January and September, there was an increase in yearly ridership of 43.12% between 2001 (before fare-free) and 2002 (during fare-free). RepoRts on puBlic tRansit agencies Retaining totally FaRe-FRee policies east chicago, indiana In 1976, the report Small City Transit: East Chicago, Indiana: Free- Fare Transit in a High Density, Industrialized Area, was prepared by J. Misner for the Urban Mass Transportation Administration (47). This report was not made available; however, TRBâs elec- tronic database provides the following description: âEast Chicago, Indiana, is an illustration of a free-fare transit service operating in a high density area. The transit service was devised with a minimum of help from professional consultants, and without sophisticated routing, scheduling, or marketing plans. The background of the community is discussed along with a description of the implemen- tation process and operational characteristics of the public transit service. The process through which the community responds to the specific needs for public transit service within the local content is stressed.â When employees of the agency were called for a copy of the report, they were unaware of its existence. They stated that in the early 1970s the mayor of East Chicago simply felt that it was important for people to have a fundamental means of mobility and he persuaded others to support a fare-free system. When interviewed for this TCRP project, they believed that they were the only public transit system in the country offering fare-free service. amherst, massachusetts A report entitled Amherst, Massachusetts Fare-Free Bus Research and Demonstration Project, produced by the University of Massa- chusetts in 1977, reviews the project background and scope, details its conduct and extensive data collection and analysis, presents find- ings and conclusions, and discusses the transferability of these find- ings and conclusions to other urban areas (48). The major objectives of the project were to determine to what extent at first providing a fare-free bus service, and later, increasing restrictions on intra- campus automobile use would have in a shift away from commuting by automobile in favor of commuting by bus. There was also con- cern as to how changes in transportation services would affect com- munity attitudes toward public transportation. Significant findings were that: (1) introducing high-frequency, fare-free public transit services attracts high levels of ridership of low-income groups, while only slightly reducing automobile use and traffic congestion; (2) increased parking fees are not as effective a deterrent to automo- bile use as are reduced parking availability and strict parking regula- tion enforcement; (3) increases in parking fees that are perceived as relatively large will be met with strong opposition from lower-income workers for whom the automobile is the only available mode; and (4) fare-free public transit will have a significant positive impact on the demand for multi-family housing and sales volumes of retail estab- lishments, depending on their relative proximity to transit bus stops. state of washington The Washington State Transportation Center produced a report in 1994 entitled Fare-Free Policy: Costs, Impacts on Transit Service, and Attainment of Transit System Goals (5). This study sought to understand the potential and problems associated with fare-free pub- lic transit policy. At the time the report was written, Washington State had a number of such systems that were fully fare-free (there are now only two providing such service owing to a dramatic decrease in operating support previously provided by the state). This paper reported on the potential benefits and costs of fare- free service based on research of the public transit agencies that had implemented a fare-free policy in the United States, and partic- ularly the state of Washington. Twenty different agencies are iden- tified, although more than half of those listed provided fare-free service only in restricted areas such as downtowns or university campuses or were short-term experiments. The report attempted to answer three questions: 1. What is the net cost of fare-free transit? 2. What are the ridership and quality of service impacts of fare- free public transit? 3. How will fare-free public transit affect the agencyâs goals (i.e., efficiency of the system, mobility, environmental qual- ity, land use, public perception of public transit)? The report explores the net cost or income of fare-free public transit, noting that by eliminating fares the revenues collected are reduced
62 to zero, but that the costs related to fare collection (i.e., equipment and personnel) can also be eliminated, potentially cancelling out the loss of revenue. The Seattle bus tunnel and Island County Transit are provided as examples. In both cases the costs of fare collection were greater than or equal to the revenues collected, meaning there was no net income from collecting fares. The proportion of total operating costs made up of fare collection varies based on the size of the public transit agency. The cost of collecting fares is gener- ally between 1% and 3% of a public transit agencyâs total operating expenses, although an informal survey the authors conducted indi- cated that for smaller systems those costs were between 5% and 7%. Based on 1990 operating statistics for Washington state systems, the gross fare box recovery ratio of most public transit systems was less than 10%, with only three having a recovery ratio higher than 20%. The major point the authors make is that in the case of small public transit systems, the costs of collecting fares might be very close to the revenue those fares produce, producing net annual rev- enues of less than $30,000. Eliminating fares would allow agencies to focus on other aspects of their service and result in benefits to their communities. For instance, the LINK system in Chelan and Douglas counties was lauded for its substantial ridership growth and its importance to the elderly and others who lacked transporta- tion options. The system was recognized by the downtown busi- ness community for increasing business and was voted the best new âbusinessâ in the area for 1991â1992. The two types of impacts studied in this paper are increases in ridership resulting from a reduction of fares (to zero) and the change in quality of service due to the same reduction of fares. Several short-term fare-free experiments had a range of ridership increases from 13% in Salt Lake City in 1979, to 83% in Topeka, Kansas, in 1988. However, the most successful fare-free systems began as fare-free so that a before-and-after comparison is not possible. This paper concludes that ridership can be expected to increase by at least 25% and likely closer to 50%, with new systems having the largest increase compared with otherwise expected ridership. Four different types of ridership increases are identified: (1) choice public transit riders switching from auto, (2) public transit riders who otherwise could not make the trip, (3) public transit riders switching from alternative modes (i.e., walk, bike, carpool), and (4) joy-riders. The goal of any public transit agency is to increase the first two groups. The report noted that in Topeka, 36% of riders during the fare-free month were choice riders. One segment of the population that causes debate among those discussing fare-free service is young riders. In some cases (e.g., Austin, Texas) such riders were viewed as a negative result of fare- free policies because of joyriding, rowdiness, and overcrowding. In other cases (e.g., Logan, Utah; Island Transit, Washington; and LINK Transit serving Douglas and Chelan counties, Washington) serving youth riders was seen as a priority since it relieved parents of the need to transport their children and increased access to com- munity resources for young residents. The report also noted that other segments of the population, such as drunks and transients, can be more likely to use fare-free public transit. Both Seattle and Austin reported problems with these groups and with increased amounts of vandalism; however, other fare-free public transit agen- cies (e.g., Cache Valley Transit District in Logan, Utah, and Island Transit in Washington) had few of these problem riders and did not regard them as major obstacles to providing fare-free service. The agencies that did not have serious issues with problem riders were smaller communities with more aggressive policies and practices including education and bus suspensions. This paper found that fare-free policies can either improve or detract from the quality of service provided, based on several fac- tors such as the size of the community or the degree of commitment from management and the agency. As noted previously, problem riders can negatively impact the image of public transit and the perceived quality of the service in the eyes of other passengers. Crowding and possible rowdiness can be an issue for drivers to deal with; however, operators find that this aggravation can be offset by the reduction of conflicts between passengers and drivers at the farebox. Average boarding times per passenger should decrease by as much as 18%; however, with an increase in the number of board- ings and stops these time savings might be cancelled. Experiences with fare-free policies in the state of Washington were reported to be overwhelmingly positive, a result the authors found consistent with other completely fare-free systems in the United States as identified in their research. The paper recommends that all small- and medium-sized transit agencies in Washington State consider a fare-free public transit policy. Additionally, all new systems should consider a fare-free policy from the start. The authors believed that their positive review of fare-free pol- icy conflicts with common thinking of the policy within the public transit industry. They concluded that much of the negative interpre- tation of the policy was based on a very limited set of experiments with the policy at larger systems, such as in Denver, Colorado, and Austin, Texas. Their research points out why these earlier experi- ments should not be used to dismiss the policy and why the policyâs potential success is largely dependent on community values and agency management and how well they prepare for predictable strains on operations and maintenance that will result from signifi- cantly increased ridership. Furthermore, they present a conceptual overview of why the removal of the fare box results in substantial ridership increases above the levels predicted using standard fare elasticity relation- ships. They note that a fare-free policy not only reduces the cost of using public transit, but it also completely removes the psychologi- cal barrier of the fare box, which usually requires exact change and often confuses people who do not know what the fare is. upper Valley of new hampshire and Vermont In 2008, CTAA produced a report entitled An Analysis of the Impacts of Introducing a Fare for Riders of Advance Transit to assist that agency in determining whether it should charge a fare after oper- ating fare free since 2002 (10). The report identified the various sources of revenue that support Advance Transit, including federal grants (FTA Section 5311 Program), state funds (from both New Hampshire and Vermont), municipal funds, and local sponsorship including Dartmouth College and Dartmouth Hitchcock Medical Center. The analysis showed that if fares were reintroduced on Advance Transit buses, there would be significant costs associated with pur- chasing fareboxes and operational costs such as daily tallying of receipts and depositing money. The median cost to outfit all of the 33 buses with fareboxes would be $407,550, with a life cycle of between 15 and 25 years. The estimated costs for fare collection per year were $53,354. Estimated one-time costs associated with policy creation and public hearings would be $3,900. Marketing and edu- cation of users was estimated to cost $30,000. Total first year costs associated with implementing fares were estimated to be $441,450, with a yearly cost of $53,354 thereafter. These costs would be offset by the new fare revenue generated. The amount of money generated would depend on the fare estab- lished and the number of retained riders (also a function of fare
63 cost). A $0.50 fare was estimated to generate annual revenue of $90,688. A $1.00 fare was estimated to return annual revenue of $145,600, whereas a $2.00 fare would generate $175,550. For the eight years prior to the study, including the six when all fares were removed, ridership steadily increased for Advance Tran- sit. Between 2000 and 2002, when fares were removed, ridership increased by 32%. It was expected that reintroducing the fare would decrease ridership. The proposed fare increase would not apply to all riders. It was estimated that of the 400,000 annual riders, 208,000 would pay a fare. This report assumed an average ridership reduc- tion rate of 30%. Three fare increase scenarios are assumed. For a $0.50 fare, ridership would decrease by 26,625. With a $1.00 fare, ridership would decrease by 62,400. With a $2.00 fare, ridership would decrease by 120,225. The report identifies several other potential impacts that reintro- ducing fares to Advance Transit might have. First, given a cost per mile for auto travel ($0.585), the diversion of 62,400 trips (based on a $1.00 fare) at 5.4 miles per trip would cost previous riders $197,122. Additionally, these estimated new auto trips would gen- erate 7.8 tons of emissions and consume 13,478 gallons of gas. Second, Dartmouth College and the Dartmouth Hitchcock Medical Center (two of the main sponsors of fare-free public transit) would likely need to spend millions on new parking facilities. Third, rein- troduction of fares after so many years without them would lead to passenger confusion and slower boarding times, affecting bus schedules. Finally, even with decreased ridership associated with fares, it was not expected that Advance Transit would be able to reduce service hours or frequency. otheR peRtinent ReseaRch aDDRessing FaRe-FRee puBlic tRansit policies Impact of âFreeâ Public Transport on Travel Behaviour: A Case Study, produced in 2006, provides an analysis of how free public transport impacts mobility, including extra trips made, temporal shifts, route choice, and mode split (49). The paper does not detail the benefits of fare-free public transit at a system level, but instead gives more detail on behaviors of individual riders. In Brussels, Belgium, free public transit was provided to Dutch- speaking students but not French-speaking students. This allowed for a comparison between two groups, with the major difference being the public transit subsidy. A survey was conducted of stu- dents from both Dutch-speaking universities and French-speaking universities. The only direct cost for the fare-free program was a government subsidy of 1,446,293 Euros to refund the public transit pass costs of 8,077 students. The Brussels Public Transport Network Managing Company, which provides the transit service, did not have any addi- tional costs due to the free-fare program for students. The capacity of the service remained the same. Several benefits are identified such as the increased consumer surplus (savings from transport costs). Additionally, it is suspected that these free passes would be habit forming and the students would be more likely to ride public transit later in life. The con- sumer surplus was estimated to be approximately 706,737 Euros. Providing fare-free public transit also caused some mode shift from private auto to transit. It is estimated that students who had cars available drove 47.64 fewer kilometers during the peak period and 28.62 fewer kilometers during the off-peak period per week. This translates to 3,196.82 total kilometers per year removed from the roadways. After accounting for the monetized value of pollution, accident, noise, and congestion reduction, the total cost savings were estimated at 1,927,939 Euros per year. When considered along with the consumer surplus of +706,737 Euros and the subsidy cost of -1,446,293 Euros, there is a net benefit of 1,188,383 Euros. The authors do note that in a dense urban environment such as Brussels, the reduced demand from students on the roadway will be filled by increased demand from other segments of the population, thereby reducing some of the benefits. During the first year in which free passes were made available, 47% of students had used the free pass. The report notes, however, that French-speaking students who did not have free passes rode public transit more often than Dutch-speaking students. This is likely owing to other factors that influence public transit ridership, such as housing locations and perceptions of the city. Just 36% of Dutch-speaking students live in the city, whereas 81% of French- speaking students live in the city. These Dutch-speaking students are more likely to commute to class and leave the city afterwards. Fare, Free, or something in Between? This paper, produced by the National Center for Transit Research in 2003, is a synthesis of several fare-free public transit agenciesâ experi- ences and reviews the costs and benefits of these programs (20). The paper identifies several potential disadvantages of fare-free public transit, including costs, vandalism, problem-riders, and overcrowding. Cost disadvantages include the loss of farebox revenue and the expenses of required additional capacity in terms of equipment, per- sonnel, and repairs. The loss in fare revenue may not be great for smaller agencies where fare box revenues typically account for less than 10% of the operating cost of the agency. However, for a large system such as MiamiâDade Transit, which had discussed the pos- sibility of providing fare-free service, a significant amount of rev- enue to operate the system is gained through fares ($70 million was collected annually to help pay for the total operating expense of $210 million in 2001). Replacing that amount of money with another source would be difficult without significant community support. This paper reviewed a considerable number of articles that described the fare-free experiment in Austin, Texas, conducted by Capitol Metro between October 1989 and December 1990. Rider- ship was reported to have increased by 75%, although expanded service, the institution of the University of Texas universal access program, and adjustments for normal growth complicated any- oneâs capability to determine just how much of the increase was due to fare-free policies alone. Nonetheless, the experiment was regarded as successful in attracting ridership, but problematic in that it attracted undesirable riders who drove away quality rider- ship. Seventy-five percent of all bus drivers petitioned the authority policy board to end the fare-free program due to these problem rid- ers and the stress they were causing. A study conducted during the spring of 1990 for Capital Metro of riders and the general public during the fare-free demonstration found that the five most important factors in determining whether or not to ride the bus were: 1. On-board safety 2. On-time performance 3. Convenience of routes 4. Cleanliness of the bus 5. Frequency of service Almost ironically, the three least important factors were cost (fares), outside appearance of the bus, and driver courtesy.
64 The authors opined that when there is no cost associated with using public transit, riders are likely to not have the same respect for the service that negatively impacts the image of the bus system, and this causes problems for drivers. In Austin, public transit officials noted a substantial increase in truants, vagrants, and other âdubi- ous categoriesâ of riders. These problem riders required additional security, and maintenance personnel to handle the repairs necessary on the buses. Lastly, the paper notes that there are two types of riders who can overwhelm the system and drive away âqualityâ riders. These include riders who would have used other modes for short trips (walk or bike) and those riders who use the system for negative and criminal purposes. The increased ridership from these types of rid- ers will lead to higher aggregate boarding times and more frequent stops. None of the experiments referenced in the report found that fare-free public transit led to a significant shift from private auto- mobile to public transit. The paper concludes that fare-free service may work better in smaller transit systems where the cost of fare collection may cancel out revenues and where âproblem ridersâ may be easier to deal with because of the size of the community. For large cities, pre-paid fares may be more reasonable in that the revenue stream does not end, but the farebox is still removed from the front of the bus, possibly increasing efficiency. Externalities by Automobiles and Fare-Free Transit in GermanyâA Paradigm Shift? The case study used in this paper is of Templin, Germany (22). The city has a population of 14,000 and serves as a health resort town 60 miles northeast of Berlin. The bus system consisted of two main lines and two auxiliary lines. The service was made fare-free in December 1997. The purpose of the fare-free policy was to reduce automobile usage, noise, pollution, and accidents. The article discusses whether or not fare-free public transit is able to induce mode shift from private auto to transit. Several issues associated with fare-free service are noted. First, public transit is not just a substitute mode for cars, but also for walking and biking. Second, free public transit will likely induce more travel from cur- rent users. Third, previous empirical studies found that the potential for attracting automobile users to public transit is small, with most new ridership coming from induced travel, pedestrians, and shifts between peak and off-peak times. In the first year of the fare-free program, ridership increased quite spectacularly from 41,460 to 350,000 passengers per year. In two more years, ridership exceeded 512,000 passengers per year. A previous study found that the majority of new riders were ado- lescents. It is noted that other fare-free programs ran into similar issues of youth making up a large portion of new riders, leading to increased cases of vandalism. Most of the passengers indicated that they previously walked (35% to 50%) and bicycled (30% to 40%). Approximately 10% to 20% would have shifted from auto use. The potential for mode shift was greatest for work and school trips. Cost savings from fare collection was considered negligible for such a small system, although removing the need for ticket check- ing saved approximately 5,000 to 10,000 Euros. While the marginal cost per passenger during off-peak times can be considered zero, it is significant during peak periods. An above-average increase in peak riders will lead to substantial costs, estimated at 20,000 Euros. Perhaps this paperâs greatest contribution to thinking about fare- free public transit in a new way was that it applied cost values to several car-related externalities. The reduction in pollution from a reduction in auto travel was valued at 5,000 Euros. From a road safety perspective, fare-free public transit attracted a substantial number of pedestrians and bicyclists, which in turn reduced injuries and fatalities from accidents. A cost reduction of between 43,000 Euros and 120,000 Euros is approximated. The overall benefits of fare-free public transit are esti- mated to be between 33,000 Euros and 115,000 Euros depending on how environmental and safety costs are priced. The lost fare revenue was estimated to total 90,000 Euros. Therefore, there is a positive or negative net effect depending on the monetary values placed on envi- ronmental and safety factors. However, the article questions the propri- ety of accomplishing most ridership increases by people changing from non-motorized modes to a motorized mode (the bus). TCRP Report 95: Traveler Response to Transportation Systems Changes: Chapter 12âTransit Pricing and Fares The goal of this report is to provide insight into how public transit ridership responds to changes in fares, including changes to fare-free service (14). Changes in fare are categorized as increased fare to increase revenue to account for increased operating costs, decreased fare to stimulate ridership, or changed fares to increase equity among users. This report provides empirical data to identify fare elasticities for various fare change situations (both increases and decreases). In addition to changes in costs, different fare structures are analyzed (i.e., discounted prepaid fares, peak and off-peak fares, and fare dis- counts for certain demographics). Finally, fare elasticities are com- pared across travel demographics and trip characteristics, such as trip purpose, income, and age of the traveler. The report notes that there were several demonstrations of fare-free public transit funded by the federal government in the 1970s. The fare elasticities for several fare-free demonstrations are provided, based on hours restrictions (off-peak or all hours) and service restriction (central business district [CBD] only, senior citizens, students, and no restrictions). The average fare elasticity for demonstrations with âno restrictionsâ is -0.28 for off-peak and -0.36 for all hours. The highest fare elasticities were found in CBD areas, where walking is the primary mode. The average fare elas- ticities for CBD areas were -0.61 during off-peak hours and -0.52 for all hours. A summary of 20 fare-free public transit programs is provided. In general, it was concluded that fare-free public transit programs significantly increased transit ridership, even more so than would be expected by the SimpsonâCurtin rule, which would indicate a 30% increase in ridership with a 100% decrease in fare. A case study is provided for the fare-free zones within the CBD areas of Seattle, Washington, and Portland, Oregon. In both cities, a substantial portion of trips within the CBD were carried using pub- lic transit. Fare-free zones were designated in the downtown areas, which were both later expanded. The reasoning behind instituting these fare-free zones was to improve passenger boarding times and increase ridership. Surveys were conducted in Seattle in July 1973, May, 1974, and 1977. Surveys were conducted in Portland in May 1975 and November 1977. In Seattle, ridership increased from 4,100 trips per day to 12,250 trips per day within the CBD, mostly during the midday lunch period (11:00 a.m. to 2:00 p.m.). Of these trips, 25% would not have otherwise been taken, 31% would have been from walking, 19% would have been by the replaced Dime Shuttle, 15% by other buses, and 10% from other modes. Similar success was seen in Portland, with ridership increasing from 900 to 8,200 trips per day 34 months later. Most of these trips were made
65 at midday (65%), between 9:00 a.m. and 4:00 p.m., and 22% dur- ing the evening peak between 4:00 p.m. and 7:00 p.m. It is noted in this report that several major changes were made in Portland during the evaluation of fare-free public transit. In the 1980s, there was some consideration of removing the fare-free areas in Seattle due to a lack of support from the business community. However, studies indicated that Seattle saved more money in operational costs from not collecting fares than they lost in revenue, although the specifics for these operational costs are not provided. Similarly, in Portland, there was some talk of removing the fare-less square, but this did not happen owing to public outcry. âFree Public Transportâ Written in 1973, this paper reviews the benefits and costs of fare-free public transit, especially for German agencies (12). The price elas- ticities for several public transit agencies were estimated based on fare changes in cities including Hanover, Germany; The Hague and Utrecht, Netherlands; and Boston, Massachusetts. The elasticities experienced indicate that the SimpsonâCurtin rule of thumb of -0.3 elasticity is reasonable, although it will vary based on trip purpose. Fare-free policies will have negligible impact on business, journey to work trips, and social trips (i.e., recreation and entertainment), but may have significant impacts on lunch trips or shopping trips (e.g., more trips to the city center rather than the suburbs). The paper adds that joyriding trips should be expected from young riders. When analyzing the effects of fare-free public transit, the reduc- tion in fare cost should be compared with four other factors: (1) the influence of travel time (in and out of vehicle travel time, fre- quency, reliability); (2) quantity-related (convenience and safety); (3) route-related (length of lines and transfers); and (4) status/image. The most important factor from several opinion studies that were synthesized was speed, followed by fares, reliability, frequency, comfort, punctuality, seating, no transfers, and accessibility. Thus, convincing private vehicle users to switch to public transit should not be done with just a reduction in fares, but by improving all aspects of public transit service. The projected costs of fare-free public transit for several German towns are provided, ranging from 22 million DM in Kassel to 350 million DM in Hamburg. These estimates take into account lost fare box revenue, remaining advertising revenue, increased capacity required during peak periods, savings from the elimination of fare collection, savings from greater productivity of buses as travel times decrease owing to reduced congestion, and savings from overhead costs derived from eliminating money collection. These costs are seen as a substantial burden to municipalities, and the authors are doubtful that the German government would be willing to completely finance public transit. Finally, the authors provided some insight into the relationship between fare-free public transit and redistribution of income. One noted argument for free public transit is to improve conditions for the poor, elderly, very young, and disabled, as well as to equalize the distribution of incomes. The authors noted that in 1958, the amount spent on commuting to work averaged 3% to 5% of a householdâs income. A study in Hamburg found that over time, private vehicles were becoming more affordable to lower-income households. Addi- tionally, higher-income travelers may choose to ride public tran- sit to avoid congestion and parking, or for other reasons such as health and age. As to the question of the redistribution of income, the authors concluded that the increased tax required to subsidize free public transit would not be sufficiently effective and that other methods are better suited.