xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam 2200253Ia 4500
controlfield tag 001 022589663
008 941117s1989 flu s000 0 eng d
datafield ind1 8 ind2 024
subfield code a C01-00263
Transit service costs and the role of transit in serving larger and more dense urban areas /
prepared for the Governor's Task Force on Urban Growth Patterns by the Center for Urban Transportation Research, University of South Florida.
Tampa, Fla. :
b Center for Urban Transportation Research, College of Engineering, University of South Florida,
iv, 36, 17 leaves ;
Also available online.
x Cost of operation.
Urban transportation policy
Governor's Task Force on Urban Growth Patterns.
University of South Florida.
Center for Urban Transportation Research.
USF Faculty and University Publications.
t Center for Urban Transportation Research Publications [USF].
y USF ONLINE ACCESS
9 7 098
Transit Service Costs and The Role of Transit in Serving Larger and More Dense Urban Areas Prepared for Tbe Governor's Task Force on Urban Growth Patte r ns By Center for Urban Transpor tation Research University of Soutb Florida May 1989
Table of Contents Sectio n Page Executive Summary 1 I. Introduction 4 D. The Unit Cost of Transit Service Delivery 6 Data Availa b ility 6 Cost Variations 7 Unit Service Deli v ery Costs as a Fu n ction of Area Size/Density 7 ill. Transit Serv i ce Consumption as a Factor i n Transit Service Costs 10 Factors Influencing Total T rans it Service Operating Costs 11 Transit Demand I mp act of Various Density Neighbor h oods 12 IV. Empirical Data on Transit Costs for Various Size/Density Areas 15 Florida Data 15 Observations on Florida Data 15 N ational Data 20 Observat i ons on N ational Data 20 Additional National Data 25 V. Transit Service Concepts for Various Urban Configurations 27 CBD Radial Services 27 The Grid Netwo rk 28 Express Services 28 Transit Center Service Concept 30 Other Service Concepts 32 VI. Using Transit to Support and Encourage Denser Development 32 Travel Be havio r Density Relationships 32 Regional Versus Cor ridor Density and Density Versus Land Use Configuration 33 Expanding Transit Use 34 Using Transit to Encourage Higher Density 35 Appendix A Graphic Summary of U.S. Transit Industry Performance A-1 11
List of Tables Table Table 1 Typical Operating Characteristics of Various Size T r a n sit Operations Table 2 Hypo thetical Scenario Showing Transit Demand Impact of Vario us Density Neighborhoods Table 3 Summary of Percent Changes in Scenario Variables Table 4 F lorida Transit Service Performance Data Proper tie s Ranked by Density Table 5 Flor id a Transit Service Cost Characteristics Prope rties Ranked by Density Table 6 F lorida Transit Service Cost Characteristics Prop erties Ranked by Population Table 7 Nationa l Transit Service Performance Data Properties Ranked by Density Tabl e 8 Nationa l Transit Service Cost Characte r istics Propert ies Ran ked by Density Table 9 National Transit Service Cost Characteristics Properties Ranked by Population JU Pag e 8, 9 12 13 14 16 17 18 22 23 24
List of Figures Figure Page Figure 1 General Transit Performance Trends as a Function of Urban Size/Density 5 Figure 2 Interrelationships Between Factors Influencing Total Transit Service Operating Costs 11 Figure 3 -Labor Allocation for Florida Transit Properties 19 Figure 4 Flori da Transi t Performanc e versus Population 21 Figure 5 -Florida Transit Performance versus Density 2 1 Figure 6 National Transit Performance versus Population 26 Figure 1. -National Transit Performance versus Density 26 Figure 8 -Typical Transit Network Configurations Rad ial and Grid Concepts 29 Figure 9 -Transit Center Service Advantage for Dispersed Activities 30 Figure 10 Typical Transit Network Configur a tions Express and Transit Center Concepts 31 Figure A 1 Operating Cost Per Revenue Vehic l e Hour A-2 Figure A2 -Operating E xpense Per Veh i cle Revenue Hour A -3 Figur e A3 -Annual Compe n satio n Per FTE E m p l oyee A-4 Figure A 4 -Annual Vehicle Revenue Hours Per FTE Em p l oyee A-5 Figure A5 -Ope r ating Cost Per Passenger Mile A-6 Figure A6 Route Miles Per Square Mile A-7 Figure A7 Annual Vehicle Revenue Miles Per Route Mile A-8 Figure AS Passenger Miles Per Revenue Vehicle Hour A-9 Figure A9 -Passenger Miles Per Vehicle Revenue Hour A-10 A10Passenger lVllies Per Capita A-11 Figure All Operating Expenses Per Passenger Mile A-12 Figure A12 Subsidy Per Capita A-13 Figure A13 Average Annual Vehicle Miles Per Employee A-14 Figure A14 Public T ran sport Share of Perso n Trips A -15 Figure A15 Transit Share of Ur ban Passenger Miles A-16 Figure Al6 Transit Use for Urban Work Trips A-17 iv
Transit Service Costs and The Role of Transit in Serving Larger and More Dense Urban Areas Executive Summary Public transportation can play an inc reasingly important role in u rban mobility fo r F lor ida cities that are growing and b eco ming more densely populated. Public trans i t can support higher densities as well as encourage growth and hi gher density. Adequate reso urces, careful planning and a full understanding of the appropriate public transit fac ility and service inve stme nts are required to maximize the contribution of pub li c transit to u rban mobility and quality of life goals. Transit ridership tends to be greater in larger, more dense urban areas. Similarily, t he more dense par ts of a given urban area tend to have greater transi t use. This g re ater r ide r demand allows im proved service (if service is expanded proportional to the greater demands) or allows improved productivity (if the increased demands are accommodated on existing services). However, the u nit cost of prov idi ng t rans it service tends to increa se in la rger, more de nse urban areas. This partially offsets the efficiencies possible with greater ridership. In general, as an area grows, the total subsidy for transi t operations increases as does the cos t pe r cap ita for the u rban area. Thus, transit will require additional public dollars if it is to play an increasing role in urban mobility. This financial need will be increased farther in urban areas where substantial capital facility investments a r e made. Transit can allow higher densities to be accommodated and may he lp to encourage higher den s ities, but a var iety o f other factors including market conditions and oth er incentives and disincentives will be required. While fixed guideway transit has been successful at encouraging higher density at employment and commercial s tati o ns, public acceptance of higher residential dens i t ies and t!Jeir attractiveness to the market may be more difficult to obtain. Density s hould not be the oaly land use consideration The distribution .of activities and urban design will also affect transit's success a n d the overall co ngestion/travel efficiency in a region High speed radial transit lines can e ncourage sprawl as well as can similar expressway f acilities. Persons considering the tradeoffs in terms of ql,!ality of life, air 1
quality, and energy consumption need io recognize that the comparison may be between a 20 mile express bus or rail ride to a CBD versus a 5 mile auto trip to a suburban concentration of employment, rather t ha n between comparable length trips by different modes. While higher density will enable better and more attractive transit services, a significant share of the additional travel demand associated with higher density will continue to have to be accommodated by the roadway system. Unless adequate reserve highway c apacity exists, higher density will require continued expansion of the roadway network capacity. Part of the difference in travel behaviors between denser urban areas and suburban areas is explained by the density; however, part is also explained by the life styles and socio economic conditions. Newly developing high density areas are not likely to have as great a degree of t ransit dependency as are o lder low in come and highly transit dependent areas that characterize the high density areas of some major American cities. The differences in Florida t ransit market conditions relating to the rapid growth, the age distribution of the population, the nature of the employment, the weather, and other conditions require a detailed understanding of the t ravel market when evaluating the potentia l of transit in Florida cities. If transit service is used to e ncourage transit market growth where the capacity expansion precedes growth in demand, the cost efficiency of transit will decline in the near term. The required financial and policy/political support necessary to enable t ransit to increase its role in urban transponation will be significant. The response of land use and public travel bebavior to improved transit services and in vestment is not completely understood or predictable. However, the improved accessibility that transit can provide will have an influence on development patterns. It should be recogniz ed that the land use influence o f transit may not be felt until there is sigrtificant progress on the implementation of a transit system/service plan. Much of the growth expected during the next 25 years will occur before the presence of extensive service expansion or the implementation of fixed guideway facilities; h ence it is important to recogrtize that this increment of growth will not fully reflect a long range plan for transit. The ability of transit to influence land use is also affected by a number of facto r s including the other incentives and disincentives that influence l ocation decisions. A variety of factors s uch as good urba n design, complementary parking policies, adequate feeder services, cost effective design of facilities and services, cooperative efforts w ith the private sector for j o int design and funding, and assorted o ther good planning practices can make the difference between successful and unsuccessful transit service and facility investments. The timin g of investment in transit services and facilities must be sensitive to the travel market conditions. Urban areas should implement services that support mobility goals and evidence public commitment to a high density tran sit corridor, but that do not incur huge expenditures so far in advance of t he transi t r idership market development that the projects are very cost ineffective and risk the loss of public support for transit. Transit p lanrting in rapidly growing urban areas will require considerable expenditures for high quality data collection, planning, a n alysis, evaluation, market studies, and public education. 2
Understanding att itude s towards tr ans it use, the full costs o f providing t r a n sit, a nd monitoring system p e rformance will be e ssen t i al in developing efficient and e f f ec t ive pr ogra m s In addition the extent to which tra n sit i s us ed as a proactive too l to encourage hi gher dens ity or in an effo rt to capture the o t her advantages of t rans i t s ervice the more likely it is t ha t the trans it s e rvice will b e a l ess eff i cien t ope ration and require grea ter subsidy. A variety o f condit i ons inc l uding low g aso line costs greater l ab or f orce p art ici patio n by women, and subu r b anizati o n of r e sid e n t ial and more recently retail and employment locati ons have contributed to the current si t u ation w he re transit is p la y ing a smaller ro le i n ur ban mobility and suffers unde r increa s ing pressures to reverse declines in productivity and cost effectivene ss. Thus, for trans it to p lay an e ver greater r ole in addressing the increasingly seve re mobili t y problems and quality o f life goals trans i t is being called upon t o help re ach, presents a f o r mida bl e challenge. Transit doe s not offer a qu ick, low cos t or single-han de d solution to mobility pro bl ems bu t it can play an impo rta nt role given adequate resource s strong policy support r easo n a b le expectations and informed, careful planning. 3
L Introduction This technical memorandum was produced by the Center for Urban Transportation Research, University of South Florida, to assist the Governor's Task Force on Urban Growth Patterns. The information and ideas are meant for consideration and discussion. They do not yet reflect any feedback from the committee and are not intended to be recommendations to or from the committee. This eff or t is intended to assis t in determining the role that transit may play in satisfying the transportation needs of urban areas as they grow and become denser and more congested. The findings may help in determining what role transit might play in just ifying and enc ou raging concentration of growth into higher density patterns and in determining what types of transit services, facilities, and policies may be most effective in serving mobility needs. This technical memorandum focuses on two major subject areas. The first s ubject (Sectio n s II, III, and IV ) discuss e s the cost of providing transit service in urban areas of various sizes and densities. This includes a review of available comparative cost data from exis,ting transit operations. The characteristics of the ur ban areas, service concepts and administrative factors that influence operating costs for t ransi t are discussed. The u nit costs for providing service in areas of vario us densities and sizes is discussed. The overall effectiveness and efficiency of transit services in various size and density configurations is also discussed. This includes the influence that density and urban size have on the ability of transit to attract riders. Figure 1 presents a graph of the general trends in tr ans it performance relationships as t h e density of the trans it market is increased The discussions and data in subsequent sections of the paper explain and quantify these trends. The second subject of the paper (Sections V and VI) discusses the types of transit service and issues relating to expa nding the role of transit in urban areas o f various types and densities. The different types of service configurations are shown. A general discussion of the role that tra ns i t can play in encouraging and supporting higher density development is provided. Expanding the role of transit is discussed and some possible actions are proposed. A final appendix provides graphical data s howing transit efficiency and effecti veness for various t ransi t ma rkets. 4
Figure 1 General Transit Performance Trends as a Function of Urban Size/Density Unit Cost of providing service (cost per vehicle mile) Service Uti I ization (pass. mi/ veh mi) Cost Effectiveness (cost/ pass mi) Subsidy requirement per trip Operating Expenditures per capita low density/ high density/ small ----large 5 I better better f l better l better I better
D. The Unit Cost of Transit Service Delivery A variety of percepti ons exist about the r elat ionship between the costs of delivering public' transportation service and the density and land use pattern of urban area s. Transit serv ice has historically been viewed as an efficient mode of transportation for "masses" of people where the econo mies of scale of mass transportation can be utili:zed Traditionally, the greater the density of the area the more efficiently it could be served by public transporta tion. New York and Chicago can transport people by rail transi t for less than $0.20 per passenger mile while bus services in lower density areas could cost upwards of $0.50 per passenger mile. However, data from the major transit properties in the top several metropolitan areas (the areas that dominate aggregate transit stati stics) cannot be extrapolated to l ower density moderate to small urban areas o r to suburbs of major urban areas. Additionally, the changes in the transit industry that are taking place today are expected to produce different design and cost characteristics than those of the older systems that have been the basis for many judgm ents and perceptions about transit cost and urban area density and size relationships. While some aggregate statistical comparisons can and will be made, there are several cautionary conditions that should be understood by anyone using the data. It is particularly importan t to understa nd exactly what is being reported in the statistics being used. There are differenc e s between the cost of delivering service and the cost effectiveness of the service when the levels of demand are taken into account. The cost for transit service in a given area is driven by two pri mary factors. The first is the unit cost of providing service and the second is the quantity of service required based on the transit deman ds and servi ce standard s of the area. Each of these determin ants i s discussed below. Data Availability Limited data is available on operations that are exclusive to small or less dense areas. Operations in these areas are occasionally administered in such a manner that disaggregate data canno t be obta ined either because it is not collected and reported or because operations are integrated with othe r governmental unit functions and difficult to break out. Often a regional agency operates s ervice in both the denser inner core and in th e less de nse suburban areas and disaggregate data are n ot available on operating costs for the various service areas. Additionally much of the t ransit market is central bus iness district (CBD) bound work trip commuting where part of the bus trip occurs in the lower density suburban area but the route also travels through more dense urban areas into the CBD. Thus, pure relationships between tbe cost of providing transit service and the urban area density are not easy to develop and have to be used carefully in light of the quality of the data. The data on the cost of providing service as a function of urban size is mor e readily available as, typically, a single transit authority serves the metropolitan area. In the discussions that follow there are reference s to both urban area size and density In 6
general, only the l arge r a r eas have dense ope rating env i ronments, how e v er, l arge ness alone is n o t sufficient t o result in hig her density. The comparative analysis reviews transit system p e rforman ce for various u rban areas ordered by both size and d e nsity. Cost Variati ons Operating cost s vary so significantly b etwee n ag encies that it is d iffic ult to interpre t the causal fac tors explaining the differences. Labor costs, work rules, a dministrative efficiencie s extent of capitalization used t o off-set operating costs extent of in tern alization of administration marketing, planning and related overhead costs, the servic e standards driving t he levels of service, and other cost consi de rations might com e i n to p l ay in explaining cost differences. The list in T able 1 itemizes som e of the factors that influen ce the cost of delivering set"Vice f or vario u s op e rations. The variations in cos t s caused by all these factors can be more s ignificant tha n the cost diff erences t ha t might b e reas onably attrib ute d to density. Unit Service Delivery Costs as a Function of Area Size/Density Typically, l arge r and more den se u rb an areas have a hi gher co s t o f living and o ften h i ghe r transit wage rates. Land costs and other costs te nd to be hi gher and the l eve ls of serv i ce are o f te n muc h greater in larg e r urban areas. For examp le, secu rity is a bigger c oncern in some larg e ci t ies a n d the transit authorities then internalize this cost by s tarting thei r own police for ce and/ or i ncu r hig her costs for m ai ntenan c e of va nda lized vehi cles and facil iti es. More congestion (and therefor e slowe r running speeds for buses) and hig her accident cla i ms rates might also be expected in a larger and more dense area. The net i mpact of the s e fa ctor s wil l vary across propert i es. The characterizations in Table 1 may typify or reflect the stereotype of w hat a l arge regional or major cen tral city operat ion is like in contr ast to an operation for a sma ller city or a suburban tr ansit property How ever there is a great deal o f va r i ati o n among properties and these conditions a r e by n o means absolu te s for any given property or o pe ration size. Som e of the cond iti ons clearly affect costs; for example, a s l ower operating spe ed c o mmon in a large city with congested inne r city and CBD operations would impact t he costs of delivering service. Som e of the spec i alizat i o n and m o re sophisticated equipment that a larger property may have might i m p rove effic iency o r service qual ity. In ad dit io n t o thes e considerations a variety of o the r facto rs may influenc e costs For example the p rivatization of s erv ices (cont rac ting out) is frequent l y note d as red u c ing costs by as much as 20 to 40%. Qther fac to rs particulady l a bo r costs and work rules, ca n a l so s ignifica ntly im p act costs It is important to recognize that some of these characteristics t hat affect co sts can have a gr e ater im p a ct o n overa ll co sts than does the urba n d ens ity or s iz e of an are a 7
. Table 1 Typical Operating Characteristics of Various Size Transit Operations Moderate Size City /Suburb Operation Admini stration Part of city or county government. Limited decis ion making and top administrative costs, possibly shared administrative facility with general purpose government or with operations facility. Support Services Often rely on general purpose government for facility maintenance, purchasing com pute r, security, and rela ted support. Work Rules Less likely to have organized labor, more likely to have flexibility with res pect t o use of p art t im e help, spread time and other labor costs. Infrastructure More likely to rely on exist ing infrastruc ture. Limited sp ecial facili ties and amenities. Modest signage, shelter, park ing facilities and garage facilities. 8 Large, Dense Urban Operation Separate entity with its own board or commission, possibly paid, and full administrative staff. Often separate he ad q uarters facility. Full range of administrative support for huma n resources finance, MBE/DBE, operations, marketiog, maint e nance, procurement, etc More inclined to have in-house maintenance (facilities), marketing, and independent police/security services. More incline d to have organized labor and rigid work r ules more likely to have extensive personnel procedures a nd job description sc h emes Likely to have more extensive, more elaborate shelters, transit centers and amenities including landscaping, art programs, etc. More likely to have ded i cated and secured parking facili ties. More likely to have specialized facil ities for such functions as money counting, transit police, paratransit operations, etc.
Table 1 (cotit.) Typical Operating Characteristics of Various Size Transit Operations Moderate Size City/Suburb Operat ion Vehicles and :Equipment More likely to have smaller vehicles with less sophisticated equipment. More likely to have a standardized fleet. Services and Operating Environment Likely to have less off-peak service. Likely to have a benign operating environ ment. Likely to have less congestion and higher operating speeds. 9 Large, Dense Urban Operation More likely to have larger variety of equipment with more sophistication such as electronic destination signs, electronic fare boxes etc. More likely to own special ized towing or r oadside rep air equipment. Likely to have more extensive off-peak service and weekend and owl service. Likely to operate on some streets in deter iorated condition. Likely to have vandalism problems in some a reas. More likely to bave detailed service standards and service driven by service allocation/ equity concerns. Likely to have some very congested a r e a s and have lower operating speeds. May have greater deadhead t r ave l times (travel to and fr o m garage).
m. Trnnsit Service Consum p tion a Factor in Transit Service Costs While the prev ious discussion described factors that may influence unit operating costs (cost per mile or hour of serv i ce), the o ther key e l ement in determining the overall cost of providing transit to an area is the amount of service that the area con sum es The level of transi t service provided is affected by the public's consumption of transit service and the policy demands for service. The policy de mands for service are influenced by the financial situation and by the goals of the policy makers; these in turn are related to the environmental, economic development, mobility, land use and other goals of the region and to the perceived ability of transit to help achieve these goals in contrast to o ther prog r a m s and needs. The ride r demand for transit service is affected by the tradition ally noted facto rs such as soc i o-economic condi t ions, the land use/activity distribution which influ ences travel pat t e rns, the comparative performance of t he t ra nsit service with t he auto a lternative, and the compa rative costs of transit and auto alternatives. If service is near capacity, hi gher levels of demand result i n the need for additional service in t he form of more frequen t service (lower headways), higher capaci t y vehicles, a more extensive network o f rou tes, longer hours of operation, o r some combination of these factors. If the service has adequate excess capacity, the additional demands will increase the productivity of t he service a nd result in a more cost effec tive operat i on with additi onal fare reve n u es to offset constant operating costs. Policy influence on levels of service can be exerc i sed in several forms. Often budget constraints govern the levels o f service provided, and man y urban areas have service standards that control these l evels. These s tand ards can be expressed i n a variety o f ways. Some relate to the levels of u tilizati on of service (performance based, i.e. load le ve ls s hall not exceed 1.25 times the number of seats, or the service must h ave more that 2 boardings per vehicle mile), and o t hers are based on equity objectives or are designed to increase t ransit use. They migh t include such things as specification of service frequency (i.e. no more than twenty minutes between buses in the rush hour and 60 minutes b etwee n buses in the off-peak) specification of geogra p h ic coverage (i.e no one shou ld h ave t o wa lk more than one-half mile to a bus route), or specificat i on of hours o f operation. Figure 2 graphically portrays the interrelationships between the fa ctors influe ncing total transit service operating costs for a given area. Table 2 shows hypot he tical scenarios of how the density charac t eristics might impact the leve ls of transit se rvice p rovided to a neighborhood. As the simple calculations show, the additional demand d ue to high density could result in improved transit service and a higher t ransit mode sha re, or could r esult in a more efficient operation where the new dem and fills available capacity and results in m ore efficient services. While transit does better in the higher density scenario with some combination of efficient operations or improved service and highe r mode share, the volume o n th e r oadways from the higher density also increases. 10
Figur e 2 Interrelationships Betwee n Factors Influencing Total Transit Service Operating Costs -Comparative Cost, Auto vs. Transit -Resources -comparative Performance,.Auto vs. Transit -Land Use/Activity Pattern, Density -Goals and Objectives -socio-Economic Conditions Rider Demand <==============> Policy Demand for Service for service ----------------------Amount of Service Unit Cost Provided <======> for Service 11 $$$$$$$$$$$$$$$$$$$$ $ $ $ Total Transit $ $ Service Cost $ $ $ $$$$$$$$$$$$$$$$$$$$
Table 2 Hypothetical Scenario ShOwing Transit Demand Impact of Various Density Neighborhoods Lower Density Scenario Area: Density: Labor Force Part i cipat i o n : Work Trips: CBD Sha r e of employme nt: Commuters to CBD: Share on Transit: Transit trips to CBD: Average load per bus: Number of bus trips required Buses/hour for 2 hr rush Headway (mins. between buses) CBD commute trips on road 5 sq. mi. neighborhood 3000 persons per sq mi. .5 5 X 3000 X .5 = 7500 10% 750 12% 90 IS 6 3 20 min. 660 Higher Density Scenario With &Nice Increase Are a : Densi ty: Labor Force Participation: Work Trips: CBD Share of employment: Commute r s to CBD: Share on Transit: Transit trips to CBD : Average load per bus: Number of b u s trips required B u s es/hour for 2 hr rush Headway (mins. between buses) CBD commute trips on r oad 5 sq. mi. neighborhood 4500 persons per sq. mi. s 5 X 4500 X .5 = 11250 10% 1125 15% 169 15 12 6 10 mi n 956 12
Table 2 Continued Hypothe tical Scenario Showing Transit Demand Impact of Various Density Neighborhoods HiMer Density Scenario With Improved Service Utilization Area: Density: Labor Force Participation: Work Trips: CBD Share of employment: Commuters to CBD: Share on Transit: Transit trips to CBD: Average load per bus: Number of b us tri ps required Buses/hour for 2 hr rush Headway (mins. b etween buses) CBD commute trip s on road 5 sq. mi. neighborhood 4500 persons per sq. mi. .5 5 X 4500 X .5 = 11250 10% 1125 12% 135 24 6 3 20 min. 990 High Density Scenario with No New Roadway Capacity Area: Density : Labor Force Partic ipation: Work Trips: CBD Share of employment: Commuters to CBD: Share on Transit: T ransit trips to CBD: Average load per bus : Number of bus trips required Buses/hour for 2 br rush Headway (mins. between buses) CBD commute trips on road 5 sq. mi. neighb o rhood 4500 persons per sq. mi. .5 5 X 4500 X .5 = 11250 10% 1125 47% = mode share required for 465 constant roadway volume 31 15 7.5 8 min. 660 13
The fourth scenario fixe s the roadway voiunie at the sa m e level as in the base low de n sity scenario and derives t he transit mode share required to keep the volume of traffic on the roadway a constant. This results in the requirement for a dramatic increase in mode share. The changes b r ought on ca n be sig n ificant Table 3 summa r izes t he changes. The higb density scenario po i nts out that t he roadway capaci t y need s of the highe r density area do not increase in proportion to the increase in density. Simi l arly the transit capacity needs increase in greater than proportional percentage terms to meet the needs of higber dens ity development. Table 3 Summary of Percent Changes in Scenario V ariables Low High High Higb Dens i ty D e ns ity Density D e nsity I m proved I mproved No Ne w Serv i ce Efficie ncy Roads % increase i n density +50% +50% +50% % inc r ease i n transit service + 1 00% + 0% + 150% % i nc r ease in t r ans i t ri ders +87% + 50% +417% % increase in mode share +25% + 0% + 292% % inc r ease in roadway volume +45% +50% 0% change in subsidy r equ i rement Increased Reduced Increased The scenarios suggests that the cost per capita for transit services for the lower density area would be lower based on the f act t hat the leve l of cons u mptio n is lowe r res u lting i n fewer units of t ransit s e rvice bei n g consumed This, however, could be affected by the unit costs for transit in each environment. None of the findings speak to the to t al or public sector cost of providing all surface transpor tati o n services If one did not t ailor the levels of service provided to demand but rather p r ov i ded a comparable level of t ransit service (measured in terms of frequency of service and density of network) then the cost of providing t h is service to the same number of persons in a low densi ty development pat tern (o b v i o u s l y covering m o r e area t han a high dens i ty development with the same number of persons) wou ld be higher as a result of the greater ne ed for miles of service to cover the larger area However, when the levels of service are tailored t o demand then the higb density area may be a higher cost area to s erve since the densities create a demand for transit serv i ces. I n the extreme, the per cap i ta spe n ding f o r trans i t in Wyoming is far lower than t he per capita spending for transi t in New Jersey simp l y because the Wyoming market doe s not support any significant transit use. The per capita expenditures f or other means of transportation may be very diffe rent. 14
IV. Empirical Data on Transit Costs for Various Areas Recognizing the constraints on the relationships between transit costs and urban size and density the empir i cal data for selected trans it operations in Florida and elsewhere in t he U.S were ev aluated in relationship to density and size of the t ransit service area. Florida Data Tables 4 consist of both raw da ta and var ious eff i ciency and effectiveness cost and service level ratios for most Florida transit properties. The demographic data was obtained from the American Public Transit Associatioh (APT A) data base as provided by the transit author ities. Boundary definition problems and keeping data estima tes updated in a rapidly growing area make the demographic data particularly difficult to gather Service area boundaries should be some approximation of t he geographic area that has re asonabl e access to t ransit. I n practice j urisdicti onal boundaries where data is available o r boundaries that define the area whe r e financial resources are collected, are more commonly used. The farther we get from the last census, the greater the possibility that extrapolations or other J ess precise estimating methods are used The demographic da ta uses was data reported t o APTA for 1 98 7. The rema ining data fo r the Florida t rans it properties is data from the "Florida Public Transit Pro file. This 1986 data is compiled by FDOT. The Florida properties selected for inc lus ion were all thos e properties for which APTA had demographic data on file. Table 4 shows t he raw data for the Florida properties. The properties are ranked in descending order based on density. Table 5 shows several types of da ta rel ating to cost effec tiveness, performance, service supp lied and service consumed data for the va ri ou s p ropert ies. Table 6 shows the same data this time wit h the properties ranke d in descendin g order by the size of the urban area. Figure 3 is a graph from the Florida Transit Proftle which shows administrat ive l abor as a share of total l abo r for all Florida Transit properties catego r ized int o small medium and large properties. Observations on Florida Data A number of observations can be made based on a review of t he data in tables 4-6. In many instances the Miami data is significantly d ifferent than data from the other properties and care must be taken in drawing conclusions given the significant difference in size and characteristics (for example Mia mi includes rail and a vehicle mile of rail service has m ore capacity than a vehicle mile of bus service). However, there are clearly some identifiable trends. Table 5, showing the performance measures for the p roper ti es ranked by density, show a modest relationship between the indicators and density. Revenue mil es per capita, annual hoardings per capita, annual passenger miles per capita, annual expenditures per capita, hoardings per revenue mile cost per revenue mile, revenue miles per square mile, and driver wages all tend to be higher for t he m ore dense service areas. Cost per passenger boarding and cos t per passenger mile appear to be slightly lower for high density cities, however the trend is very weak. Revenue miles per revenue hour appear to be 15
1A9L 4 F l orida Transit Service Performance D ata PROPERTIES RANKED 8Y OENSJTY system Lex::.\ too Oper. Al'l"'Ual T ri p Annual Operating Revmue Revenue Name APlA APTA APTA Passen g ers P.assengcr expense-s of Hours of Revenue s ser. Area ArC3 oensi ty itrdlngs> Miles Oper-ation Operation 87' Pop. ser-ved (000's> (000'8) (0001S) (000's) (OOO's) (()()0'$) Btolofard COlXtty 911() 1,200,000 300 4,000.0 17,757 4.87 86,477 $27,590 8,375 609 SS,095 P inel las co. PSTA 8109,00 0 276 3,076.1 9,613 6 .32 60,754 $13,119 5,527 389 S3 ,lZ9 Dade Ccx.anty I
TAILE 5 florida f l"iimit Service Colt PROPERTIES RANKED BY DENSITY syateCD LocatiOn Oper. Aevenua Amual operating 8oardi09s Cost Cost cost .......... Revenue HU. hrt/ w .... Mites 8o&rdlngs Pass. Hi. Expenditures per per ptr per Hiles Driver Operating per per por por Revenue Revenue Pan. Peu. per por b.p Cnpl to Capita Capita Hilo Milo 8oardlnst Hlle Sq. Ki. Rev. Hr. Broward County BRO 7 1 5 n $23 2.12 $3.29 ".55 S0 .32 27,917 13.8 $ 11.1 7 18X Pinellas Co. PSTA 7 11 n $15 I. 74 $2,37 S1.36 $0.22 zo.ozs 14.2 $9.10 25X dado c .... ty or 13 192 St10 3.32 $8.21 $2.47 S0.57 34,597 12.8 S11.'S 20X r a ll ahassee TAl n 22 56 $27 1.69 $2.11 $1 .25 S0.49 29,000 13.8 S8.53 32X Jac.ksorwi lle JTA 7 11 63 $19 1.45 $2.54 $1.75 S0.30 11,072 12.9 S9.80 3SX East YOlU$10 Co 'lOT 7 11 42 $17 1.65 $2.48 1.1.SO S0.41 9,000 12.7 $8.05 20X .... .... Orw.p lr-iCo. 0$0 6 11 42 $14 1.70 $2.24 S1.32 $0.33 7,882 14.4 $8.96 3TX Pal Beac:h co. P8CT 3 3 11 S9 1.07 $2.84 $2.66 $0.78 3,498 16.0 $10.70 22X A lachua Co. RIS 11 18 53 $24 1.62 $2.19 S1.35 $0.46 9,580 15.6 $8.66 23X Hi tlsborough Co. HART 9 16 n $18 1.85 $2.04 S1.1 0 $0.25 6,448 14.7 $8.50 28ll sarasota County SCAT 4 19 >10 I .09 $2.50 $2.30 $0.53 2,863 13.6 $7.98 19X 8revard County SCTA 3 15 $8 0.44 $2.67 S6. tO so.sa 890 13. 7 18.50 It X sums a 18 82 $36 2.26 S4.52 $2.00 s o .lt4 10,235 13.6 S9.28 22X
6 Flo r i da Transi t Scr v1ce Cost Characccr tstlcs PROPE.RTIES lt.ANKEI> 8'C POPULAlJON Systecn locatiO n O pe r Revenue Amual A nnuat Operating Boardi n g s Cost cost Cost Revenue Revonuo """ Fare/ Namo Hiles 8oardin g s Pass Kl. Expendi tures per per per per Mites "I les Dri ver Operating per per per per Revenue Revenu e Pass. Pass. per per l.'age Cost Capit a c a p i t a C apita Capita Mile Mile Boar d i n g Mile S'q. Ki. Rev. H r Dade County HOT 13 192 S liD 3.32 $8. 2 1 $2.47 S0.57 34,597 12.8 $11.45 2 0 % Browar d Count y BRO 7 1 5 72 S23 2 1 2 $3.29 $1.55 $ 0.32 27.919 13.7 Sll.17 18X PlnliD$ Co. PSTA 7 11 72 $15 1. 74 S2.37 Sl 36 $0.22 20,029 14.2 S9.10 2SX Jacksonville JTA 7 I I 63 $19 1.45 $2.54 $1.75 $0.30 11.073 12.9 $9.80 35% Hltlsbor009h Co. HART 9 16 72 $18 1.85 $2.04 SI.IO $0.25 6,448 14.7 S8.SO 28% Palm B ea ch co. PBCT 3 3 11 $9 1.07 $ 2.81! $2.66 $0.78 3 500 1 5.9 $10.70 221: -"' orange t r i Co. oso 6 11 42 $14 1.70 $ 2.24 $1.32 $0.33 7,884 14.4 $8.96 31% Brev ard COU'lty SCYA 3 I 1 5 sa 0.4lt $2.67 $6.10 $0.58 891 13.7 S8.50 T1% East Votus i a Co. VOl 7 11 42 S 1 7 1.65 $2.48 SI.SO $0.41 9,000 12. 7 S8.05 ZO% sarasota Count'( SCAT 4 4 19 1.08 $2.49 S2.30 $0. 53 2,866 13.5 S7.98 19% Alac hua co. RIS II 18 53 S24 1.62 $2.19 $1.35 $0.46 9 581 15.6 $8.66 23% T<1ilt()h3S:see TAL 1 3 22 56 $27 1.69 S2.10 Sl.25 $0.49 29, 015 13.8 SS.53 321: sums 8 18 82 $36 2 .26 S4.52 S2.00 S0.44 t0,236 13 6 $9.28 22%
LARGE SYSTEMS FIGURE 3 Labor Allocation for Florida Transit Properties MEDIUM SYSTEMS 10% 'IlWISPORT.a.nON D ADlllli)S"I'RA7oll :.3 liA!ImlWICZ SMALL SYSTEMS 14% 8% Source: Florida Public Transit Profile, 1 986 . Prepared by FOOT. 60% 78%
slightly lower for the high density areas. More comparably defined service boundary definitions may affect these findings somewhat; however, such an effect is not l ikely to reverse the relationships. The trends in Table 6, which compares the indicators with service a rea size are noticeably stronger. Revenue mil es per capita, annual boardings per capita, annual passenger miles per capita, annual expenditures per capita, boardings per revenue mile cost per revenue mile, revenue miles per squar e mile, and driver wages all tend to be higher for the large service areas. The cost per passenger boarding indicator does not appear to be systemati cally rela ted to service area size nor doe s operating speed appear to b e strongly related to service area size. Looking at the key indicators as discussed above, the unit cost of service (cost per reve n ue mile) does not show any economy of sca l es to either density or service area size and in fac t, shows diseconomies of scale for the Florida properties. The consum ptio n of service (annual b oardings per capita) is as expected, higher for t h e larger and dense r areas. Not surprisingly, the per capita expenditures for transit are then highe r for the dense r and larger cities. Interestingly, Alachua Co, Gainesvi ll e and Tallahass ee both performed noticeably different than their immediate peer cities. This may be a result of the presence of major universities in both areas Figur es 4 and 5 show graphic representations of t hese results. National Data Tables 7-9 show the same data as tables 4-6, for a selected sa mple of cities outside of Florida. The source of operating statistic data is the Section 15 report compiled by U.S.DOT based on data submitted by the various properties. The cities were ch osen from size categories (based on number of vehicles operated) that are used to group properties that report Section 15 data This is the same data source used. by FDOT to compile the ir Florida operating data. The cities chosen were among those that have submitted service area population and area data to APTA Beyond that no particular basis was used in the selection of cities. In general cities without rail systems were chosen, however, a few cities with rail systems were included. Observations on Nationa l Data A number of observations can be made based on a review of the data in tables 7-9. Care mus t be taken in drawing conclusions given the significant difference in size and characteristics (for example several cities in clude s rail and a vehicle mile of rail service has mor e capacity than a vehicle mile of bus servic e). Howeve r, there are clearly some ide ntif iable trends. Table 8, showing the performance measures for the properties ordered by density show a modest rel ations hip between the indicators and density. Revenue miles per capita, annual hoardings per capita, annual passenger mil es per cap ita annual expenditures per capita, hoardings per revenue mile, cost per revenue mile, revenue miles 20
Rev Mile$! SQ. Mile Rev. Miles/ Figure 4 Flor ida Transit P e rforma nce vs. Population 35 30 25 T housands 40 20 30 Board lngs/ C a pita 25 20 15 10 5 15 10 5 0 0 320 640 960 1 ,260 Populat ion (OOO's) __._Rev. Miles/Sq. Mile Figure 5 Florida Transit Performance vs. Density T 40r 35 30. 25 46 40 35 30 25 Bo3tdlng$/ SQ. Mile 20 20 Capita 15 1 0 5 15 10 5 0 0 1 2 3 4 Persons per Square Mile (OOO's) Miles/Sq. Mile -0Boardi n gs / Caplta 2 1
TABI. 7 Natio n a l Transit Serv ice Perfor m ance D a ta PROPERTIES RANKED BY OENSJTY Systea location Operating Trip .,.,.I Operoting Rcvewe Revenue F&re ..... APTA APTA APTA Passe ngers length P assenger Expenses Hiles o f Hours of Reve nues Serv. Area A rea Density (boardtngS) Kilos Operation 87 Pop. Served (000'S) (000'S) cooos> (()()() $ ) (000'$) (000J$) Chicago ll CIA 3,900,530 212.8 18,329.6 642,163 3.35 2,152,251 $599,323 120,827 9,179 S273,791 lOS Angel cs, CA SCRTO 8,086,382 1,446.0 5 592.2 497,158 3.92 1,946,524 91,960 7 ,042 $126,318 N o.rfolk, VA TTDC 698,963 155. 0 4 ,509.4 16,034 5.01 80,295 $18,227 8 492 609 $1,301 Omaha, NE TA o f Omah a 538,650 145.4 3,704 6 9,045 3 .51 31,792 S 12,868 5,142 395 $3,934 OH Ctevelcsnd RTA 1, 752,424 515.1 3,402.1 90,553 3 97 359,370 $114,76 1 24,146 1,528 $37 726 Roc-kford, IL Rockfor d HTO 1 3 7,702 41.4 3,326.1 2,05 9 4.23 8,702 $3,293 1,097 96 $676 Boston, HA KSIA 3,070,366 1,03-6 0 2,958.0 276,636 2.67 738,718 S36 t ,862 43,008 3,083 $104,542 Knoxville, TN Knoxvil l e TA 284,706 99.6 2,858.5 3,80 1 1'. 243 $4, 863 1,655 133 $ 1 ,309 "' "' Cinclmot t, OH SOliTA 863,989 327.0 2,642.2 28, 901 4.36 125,933 S40,662 10,067 757 $11,578 Atlanta, Gl\ HAll! A 2,063,500 875 0 2,358.3 155, 736 3.45 53 7 ,744 $112,703 34,930 2 345 $35,500 Phoenix, AZ PIS 1,701,170 756 6 2,248.4 1 8,847 4 .39 S24, 725 7,432 495 S6, 703. Uoshlngton, D C 'MAT A 2 900, 0 00 1,500 0 1,93 3.3 308,736 3 .84 1' 186,827 S3S0,178 63,040 3,744 S 1 82,6tl1 Port l and. OR T ri HTD 1 ,100,000 1,000.0 1,100.0 55,217 3 27 180,301 sn,241 19,562 1,203 $18,103 Mimcapol is, HN Him. H f C 1,675,0 0 0 1, 700 0 985.3 74,297 4 60 341 n1 $97,879 2t.,635 1,744 >31 ,951 Spc>Jtanc, lolA STA 360.0 969.2 5.94 45,2n 4,576 $3 ,136 las V e g as, NV Las Vegas Transit 450,21 7 570.0 709.9 5,351 3.57 1 9 085 S3,559 984 93 $4,120 An.hOI"49e, A K Anch. Public Tran 200,000 350 0 571.4 3,684 5.20 19,157 Stt ,034 2,147 134 S 1 ,331 F lint Ml Flint HfA 331,931 648.0 512.2 4,243 3.54 15,0 15 1' 921 140 $1,241 A l ba ny, NY Ois t TA 5 1 0,000 1,575.0 323.8 16,11' 3.46 55,n5 $16,558 5,682 443 $5,976 suns 30 614,451 13,315 2,299.3 2,216 ,197 3 58 7,941 544 12,347 555 471,303 33 ,485 857,917
TABlE 8 Nationa l T ra nsit Service Cost C h a ractcrlsties PROPERJIES RANKEO BV DENSITY I System Location O pera t ing Revcrue Annual Annual Operating BoaNf1ngs Cost Cost Cost Rcvcn...c For e / .... Miles Boardlngs P ass. H i Expenditures pr per per per Hiles Kites oper a t ing ,., per per per R e venue Pass Pass. per per Ex.p. capita capita capi u Capita Mile Mile Boar ding HHe Sq. HI. Rev. Hr. Chi C-390, I L CTA 3 1 165 552 $ 154 5 .31 $0.93 $0.28 567,796 13.2 46% l.<>S Angetes, CA SCRTO I I 6 1 241 $56 5.41 $4.92 $0.91 $0.23 63,596 13. I 28% Norfolk VA TJOC 1 2 23 115 S 26 1.89 S2.15 S1. 1( $0.23 54,787 1} 9 40% Oha, NE T A Of Om<)h3 1 0 17 59 $'24 1.76 $2.50 St-42 ,0.40 35,365 13.0 31% C l eveland, OH Cleveland RTA 14 52 205 S65 3.75 $4.75 St .27 S0.32 46 .876 15. 8 33% Rockf ord, ll Rockford MTO 8 15 63 $24 1 .&8 $3.00 $1.60 $0.38 26.498 11.4 Boston, HA NSTA 14 90 241 $118 6.43 $8.41 $1.31 $0.49 1.1,434 14 0 29% N k noxville, TN TA 6 1 3 50 $17 2 .30 $2.94 St.28 S0.34 16,616 12.4 29% w Cincinnati, OM SORT A 12 33 146 $47 2.87 $4.04 $1.41 S0.32 30,786 13.3 28% A tlanta, GA MARTA 17 75 26 1 S55 4.46 $3 .23 $0.72 S 0.21 39,920 14.9 31X P hoenix, Al: PTS 4 II 49 $15 2 .54 $3.33 $\.31 $0.30 9,823 15.0 27X Llash i ngton, D.C. 'IDtATA 22 106 409 S131 4.90 $6.04 $ 1.23 SO.lZ 42 027 16. 8 48% P ortland, OR Tri-Cty M T D 1& so 164 S66 2.&2 $3.69 S1. 3 1 10. 40 19,562 16.3 25X M i nneapolis, fiM Minn. HfC 15 204 $58 3 .oz $3.97 $1.32 $0.29 11\,,91 ,, 1 33X Spokane, I.'A STA 13 22 13 0 S41 1.67 $3.13 $1.88 "$0.32 12.711 14.2 Las Vegas, NV Las V egas Transit 2 12 42 sa 5.44 $3.62 S0.67 $0 19 1,726 10.6 116% Anchorage, AK Anch. Publi c Tran 1 1 18 96 $55 1.72 $5.1 4 $3.00 $0.58 6,134 16 0 Flint, HI Flin t HTA 6 13 45 $16 2.21 $2 77 $1.25 $0 3 5 2,965 13.7 Albany, NY C apital Ois:t. 1A 1 1 32 109 $32 2.84 $2.91 $ 1.03 S 0.30 3,608 12.8 36% s.,. 15 72 S77 4.70 S1.06 .30 3S.l97 1 37X
TABL E 9 Nationa l Transi t service cost chatacteristics PROP E R TJES RANtED BY POPULATJON System fon Oper-ating Revenue Aonul Amuat O,.rating 8o
. per square mile, and driver wages all tend to be higher for the more dense servic e areas. Cost per passenger boarding and cost per passenger mile appear to be slightly lower for high density cities, however this trend is very weak. Revenue miles per revenue hour (speed) appeared to be slightly lower for the high density areas. More comparably defined service boundary definitions may affect t he s e findings somewhat, how ever, such an effect is no t likely to reverse the relatiO.l\Ships. The trends in Table 9, which compares the indicators with service area size are noticeably stronger. Revenue miles per capita, annual hoardings per capita, annual passenger miles per capita, annual expenditures per capita, hoardings per revenue mile, cost per revenue mile, revenue miles per square mile, and driver wages all tend to be higher for the large service areas. The cost per passenger boarding in d ica to r does not appear to be systemati cally related to service area size nor does operating speed appear to be strongly related t o service area size. Looking at the key indicators as discussed above, the unit cost of service (cost per reve nue mile) does not show any economy of scales to either density or service area size a nd in fact shows diseconomies of scale. The consumption of service (annual hoar dings per capita) is, as expected, higher for the larger and dense r areas. Not surprisingly, the per capita expenditures for transit are then bigber for the denser and large r cities. Figures 6 and 7 show graphic representations o f these results. Additional National Data Appendix A contains add itional national data. This data is for the years 1980-1985 for all transit properties reporting section 15 data. The c ategorizat ion by urba n size is based on the census defined urbani ze d area definitions. Tbis data supports the general conclusions of tbe Florida and natio nal data detailed above and provides additional graphic information on the trends in the transit industry 25
Figure 6 National Transit Performance vs. Population 600 200 500 160 Rev. t.meSt Sq. Mi le 300 Boardlngs/ 100 Capita 1 0 1.84 3.28 4.92 9.56 Population (Millions) Miles/Sq. Mile -a-Boardings/Caplta Figure 7 National Transit Performance vs. Density T 600r 200 500 . 400 Rev. Miles/ 300 Sq. Mile 200 150 100 Boardlngs/ Capi t a 5 0 3.7 7.4 11.1 14.8 18.5 Persons per SQuare Mile (OOO's) -4-R ev. Mllos/Sq. Mile Boardings/Ca p lta 26
V. Transit Service Conee.tm for Various Urban This section of the paper will discuss the various transit markets tha t are characteristic of urban areas and briefly discuss transit service concepts that can be used in various operating environments. The focus will be on determining which types of service might be helpful in promoting greater transit use in a given area and which types of service might be successful in serving less dense suburban areas. A transit market can be defmed in a number of ways. Among the common ones are t r ip purpose based (i.e. commuter sei'Vice, shopping shuttle, etc.), geography b ased (i.e. downtown circulator, suburban commuter, crosstown), and level or type of service b ased (i.e express, loca l). No s ingle or systematic set of generic types of service can be prescribed for a given area, bu t rather a full understanding of the market characteristics is required in the actual design of transi t service. An experienced professional service planner with a knowledge of the local market conditions and the operating cost and performance characteristics of the s ervice options is required to specify service for a given area. The types of service are n o t mutually exclusive, as a given route or route system can have varying degrees of several of the characteristics referenced Review of some traditional service patterns is helpful in under standing general market types. The service concepts discussed below are those most common in the trans i t industry. In general the order of the discussion follows what would be a typical evolution of service as an urb an area grows. Paratr ansit and specialized transi t se rvices are n ot discussed. While these services are very important to their u sers and s erve an important social service they are no t "mass" transportati o n of the type that can have a significant im pact on congestion, land use development, energy savings, or air quality i mproveme nts. CBD Radial Services The his toric' bread and butter of the transit indu s try ha s been the provision of services to the Centr a l Business District (CBD). These services concentrate on the CBD which typically is t he densest and single largest conce ntration of employment in a region. CBDs r epre sent attractive transit markets because this concentration offers a la rge enough market to j ustify frequ ent transit se rvice and good coverage with the concentration e nabling a very limit ed circulation for dropping-off or picking-up p ersons in the CBD. Additionally, CBD parking is typically the most expensive in the regio n p r oviding additional t ransit ince nt ive In the largest urban areas CBD conges tion and the market size often lead to exclusive guideway transit facilities, giving t ransit additional advantages. A l oc ation within t he CBD often serves as th e major transfer po int for i ransit services. Persons traveling b eyond the CBD can take advantage of the concentration of service to this locatio n and transfe r to routes serving other locations in the urban area. The actual physical layout of routes is constrained by t he orientation of t he street system, bu t most 27
typic ally will foll ow a rela tively direct path to the CBD. Figure 8 shows a typical rad i al bus system route plan. Advantages: Focuses on best trans it marke t and the traditional lo ca tion of the mo s t con ge sted roadway facilities thus providing the greatest congestion relief advantage. Disadvantages: CBD focus makes crosstow n tra vel difficu lt. The Gri d Network As urban areas grow the share of employment o utsi de of the downtown tends to grow Additionally, the residential market begins to get very large and the distance to the CBD becomes l arger making service to non-CBD l o cations mor e difficul t when one relies on the downtown for a major tran sfer point for r a dia l bus r outes. T hese fa ctor s, a s well as t he grid stree t syste m so typ i ca l o f urban areas, resu l t in bus service increas i ngly evo l ving to a grid bas ed system with routes on major arteri als A major s hare of the ro ute s still serve the CBD; h owever, the addition o f cros s town type r ou te s al lows be tter tran sfer opportunities and b e t ter serv ice to non -CBD employment and activities. F igure 8 shows a grid system Advantages : Provi des impr oved crosstown service and impr oved service to non CBD locati ons. Par t icularly effect ive f or dense mixed use area s May encourage non CBD developmen t. Disadvantages: CrosstOwn Segments are traditionally less productive serv ices. May require numerous transfers to co mpl ete non CBD bound trips. Express Services As the urban areas cont inues to grow, the radi a l rou t e s that exist typically get extended and reac h a point where the l e ngth of the radial t ri p through ne ighborhoods b ecome s excessive O ften express s e rvic es are ini tiated t o s erv e t hese fringe markets This s ervice tr i es t o fill up t he vehicle at a park and ride loca ti on or thro u gh t he use o f a modest se cti o n o f co ll ecto r j circulator r ou te segment, then takes advantage o f higher speed rou tes to the C BD. Express services are an addition to t he basic gr id s y s tem Figure 10 shows an expres s serv ice system. Advantages: Provides a p remiu m service for more distant travelers. Can resu l t in "full buses" f o r th e li ne hau l inbound trip. Disad vantages : Often res u lts in the desire for a more com p lica te d fare system of two or more tiers Seldom provides premium service to non-CBD l oca tions. Often requires additional ca p ital investment in parking and amenities at ou t e r tr i p end. Prod u c e s low cost effectiveness for capi tal fa cili ties which typically are n e eded only on weekdays with n o t urnover for parked vehicles. The vehicles are primarily 28
utilized in the peak periods, and for more distant lots i t is not uncommon to get only one or two peak direction trips per peak hour. Transit Center Service Concept As an urban area becomes still larger and nodes of activity develop outSide of in the CBD, additional service focal pointS are som etimes developed. This can result in bus services radiating out from multiple nodes in the region. These nodes are o ften connected by high volume/ capacity transit s ervices. While t hese nodes may be major employment centers or regional activity centers like a regional mall, occasionally the t ransit service plan can create a major transit activity center by virtue of concentrating transit service i n certain loc ations. This service concept has been utilized quite extensi vely within the past few years as reg ional tran si t authorities have more aggressive l y tried to capture a la rge r sha re of the suburban market. The design of the service pattern specifically tries to compensate for the lower density and dispersed travel patterns of the suburban areas by crea ting transfer hubs. To further improve the service for the patron, all t he routes serving the node are s cheduled to meet at the transfer point at a given poim in time cre ating a "timed transfer" thereby not requiring the passengers to wait for infrequently scheduled buses. Often these tim e d transfer centers are further complemented with park and ride lots and express bus service to the CBD. Figure 9 show the advantage of transit center routing for dispersed activities Fewer, shorter routes are needed to serve a given number of origins and destinat ions. Figure 10 show a transit center system. Figure 9 Transit Center Service Advantage for Dispersed Activities Tradi tional Routes Origins A B c Pulse Point Transit Center Concept Origins A T ransit Centers Destinations 9 routes Destinations /D 7 shorter routes, transfers required c F 30
. w ..... -I I I I I I I I I I FIGURE 10 Typical Transit Network Configurations Express and Transit Cent I I -r--1 _l I v L I i /.L _/ cso ......... r" I I I I I I I I I CBO Oriented Grid with Express Services !r Concepts I I I I I I I I I I -w---', .... I ..._ I '-' / I I jf I v I v / I ; / L cso <.......1 ----1 I I I I I I I Grid/Express with Transit Centers
' Advantages: Provides a stronger presence in new fringe markets. Provides better service to non-CBD bound trips. Complements t h e development of additional activity nodes, Provides distribution service for reverse commute transit trave lers. Disadvantages: Can be expensive with inefficient scheduling of vehicles. Can be expensive and difficult to find lo c ations to schedule as ma jor transfer points. May be difficult to serve suburban m arkets if service area boundaries inte r fere with natural travel patterns. Other Service Concepts Several other service concepts are being implemented experimentally by various propert i es. These include subscription services that require regular commitments to the service, and route deviation service where in an effort to improve the attractiveness of the service, the vehicle deviates from a regular route to serve the specific destination of t he passe ngers. To improve the cost effectiveness of operations, these services are inc rea singly relying o n privatization and on th e use of small vehicles. Several tra n sit properties are getting mor e aggressive in doing service redesign studies and quantitative service standards are increasingly being u sed to i n sure prod uct ive and equitably allocated service. VI. llsing to Support and Encourage Denser Development It is clear that traditionally more dense areas do make greater use of transit. This greater use is explained by the fact that the density (and size) justicy a greater density of transit services, more frequent services. In the case of some of the larger metropo li tan areas the density justification for fixed guideway services provide the performance advanta ge that makes transit more attractive. However, the information presented suggests that transit is not necessarily cheaper t o provide on a unit cost basis for larger or more dense urban areas In fact, the costs have tended to i n crease. Travel Behavior Density Relationships Caution is justified in extrapolating the full impact of density on transit use. The traditiona l characteristics of higher density areas, may no t be characteristic of high density areas that are yet to develop in new growth areas. For example, in some high density neighbo r hoods it is virtually impossible t o get a par king space and roadway capacity is v irt ually u navailable. In certain older cities the high density environment i s often lower income, which also contributes to the greater transit use. In growing areas the new high density neighborhoods where the density is provided by new development, are not likely to be low income and the residents may not have as great a propensity to use transit as in l ow income, high density areas. An additional caution relates to the fact that when there is less choice in residential location patterns, it is more likely that the characteristics of high density inhabitants will not be the same as those who occupy high density environments by choice. As an example one person may choose the high density lifestyle because it compleme n ts life style traits such 32
as going t o dinner downtown after work, u sing the health club in the apartment bu ilding, and taking advantage of urban social activities such as clubs and boutiques On the othe r hand, the stereotypical suburban dweller with two cars (and maybe a truck) a boat, at l east one pet, and three kids (wit h agendas that include boy scouts, band, camping, soccer practice, swimming lessons, etc.) ma y not, eve n if relocated to a high density a r ea, adopt the travel behavio r and mode c h oice habits of today's typical higher density dweller. Regional Versus Corridor Density and Density Versus Land Use Configuration While regional density figures do provide useful indicators of overall density, more relevant to fixed guideway transit market analys is is the specific corridor density or the density in the immediate vicinity of the stat i o n si t es. An area that is fully built out with no rema inin g vacant space may have a relatively high overall density figure but may not have the pockets of high density that suppor t a walk t o transi t market. This micro level de nsity conside r ation is important in evaluating transit market potential. An ad d itional caution when conside ring density is t he need to recognize abso lut e densit y and not think on l y in relative terms. F or examp le, the residential density may doub l e when moving from acre lots t o half acre l ots, yet in absolute terms the density r e m ains very low in terms of transit potential. Providing p r omising transit markets by increasing resident i al densities may require very dramatic changes in densities of target locations in order to move toward a strong walk access market for transit While all the data suggests t ha t, indeed, dens ity provides a transit service advantage, it should be realized that the layout of land use activi t ies may also have a significant impact on transit us e and on the overall efficiency of the transportatio n syst em. This i s particularly relevant with regard to work trip travel to a Central Business Dis trict. While the CBD affords the dens it y and concentration of employment to justifY, good transit service and e n courage higher transit use, as an area grows very l arge the extent to which there is a transportation advantage to continuing to support a single central employment concentration becomes questionable. Some of the largest U.S. cities are good examples of cases where the strength of the CBD in attracting employment has res ulted in commute trips to the CBD from very d i stant locations Commu ter transit and rapid transit systems serve the very same function as do high-level-of-service radial urban freeways. That is, they make it possible for someone to li ve a great distance from the CBD and still commute by subsidized trans it services to a CBD employment center. A logical quest i o n arises: is it better from a n air q u ality, ene rgy use, or infrastructure cost perspective, to support a 30 mile comm u ter rail trip to downtown t ha n a 5 mile auto trip to a s ubu rban employment center? Various studies have indicated that polynuclea te d (multi-nodal) ci t ie s are more optimal from a uansportation perspective for very large cities. 33
Expanding T ransit Use The spread of conges tion from its tradit ional home in central cities t o suburbs, continuing concerns about air q uality and energy use and growing concerns abou t p rovi d ing acces s i bility to the growing employment centers in s ubu.r b an areas have continued to challen ge the trans it i nd ustry and the p arent gover nment bodies to p rov i de trans it solutions Reso u rce co nst r aints and low gas prices have c o mplic a te d t his job, and the trans it indu stry has cer tai nl y n o t given up, but it is unde niabl y a difficult challenge with success s tori es being the exception rather than the ru le. While the motivations remain as s t r ong as ever, a number of conditions c ont i n u e to make the job difficult. Auto owne rship has con t i n ued to grow and au t o op e rating costs are a s l ow i n real t e rm s as t hey have been in years Both e mployment growth and resi dentia l g rowt h c ont i nue to b e on the fringes of urban areas. E ven fast growing southern cit i es h ave had sta b l e, and in some cases declining populat ion in the central city areas, while growth has soared in suburban and frin ge areas I n many cases the new empl o yment growth i s i n campus-style suburban complexes with substant i al auto p arking available. E ven new CBD e m p l oymen t is accompanied by s ub stantial amounts of garage parking c o ntigu o u s with the of fice building. The eve r mor e dominan t t rend toward working p are nts results in limi te d tim e availab l e fo r tra vel, a co nditi o n that in all but a few loca tions favors au to use. Tri p linking for everything fro m day care s to ps to banking and grocery stops makes t he auto alternative convenient. E ven l oca tion trend s for employment are challenging the tra nsit market. CBDs r e main attractive e mployment lo c ations for high profile business firms s uch as l ega l and banking headquarter operations; h o wev er, so me massive e m ployers of middl e level white colla r employees (suc h as insurance firms) a r e moving to suburban locations to b e closer t o th e labor fo rce. Everything fro m free parking to t he need for an auto for b u sin ess u s e d urin g the day, t o t he availability o f mobile cellular p h o nes ma kes the g rowing n u m ber of CBD-located executive s a difficult market to attract t o tran si t. Asstuning that transit resources are p resently committed t o the b e s t transit markets i m p lie s t ha t s ervice expansion will be in weaker mar kets. This results in the overall pe rfor man c e of t ransit systems deteri o ratin g unless t he market conditions for trans i t can b e improve d by changing t he bas ic characteristics such as market size (dens ity) and the competitive cost and t rave l t ime tr a de offs be twe en trans i t and auto Resource constraint s make major imp r ove ments i n the l e vel of serv ice d ifficult. While there are no easy answers to the problem o f improving transit u tilizati on, there are a variety of activities u nderway at different t ran si t auth orities across the country a imed at improving cos t efficien c y and effectiveness of public t ranspo rtation. T he type s of actio ns range the full spectrum of activities affecti ng the quality a nd cost of service. New service configurations are being test e d and e valua ted in variou s mark ets Thes e include more e xperimentation with park and ride and transi t center s ervice c on cepts and other flexible serv ices to meet the different nee d s of non -traditional tr a nsi t mark ets. Mark eti ng is receiving i n creased att e ntion both to increase customer a w areness of the transit optio ns and t o p ro vid e ha r d infor mati on (s uch as schedules) to encourage transit use. P rov idin g a more convenient way to buy fa r e medi a an d encouraging e mployer participation in providing informa t ion and distri bution or subsidization tr ans it passes are a mong the act iviti es being pursued. Priva t izati o n and new types of equ ip ment and facilities are 34
focusing on controlling costs. The constant changes in methods requires a sophisticated transit management staff and policy makers willing to experiment with new idea s to try and meet the challenge of increasing the roll of tr ansit in provid ing urban mobility. Using Transit to Encourage Higher Density Acknowledging that transit service can be better and/ or more efficient in denser environments, the complementary challenge is to determine what role transit can play in helping to increase density of development. Traditionally the response has been to promote fixed guideway facilities as a means of encouraging development and increasing density. While the comprehensive set of planning activities and supporting expenditures and policies that accompany major fixed guideway facilities has, to varying degrees, resulted in increased densities around transit s tations, the obvious financial implications of this solution limit it to application in a relat ive ly small number of corridors. Additionally, the fixed guideway option is only appropriate for application in areas that already have a substantial density in place, as the constmction a nd operating costs of fixed guideway facilities proh ibit buil ding such facilities decades before t he marke.t reaches a size and density appropriate to support the facility. Thus, the challenge is to find additional transit service or facility investments tha t can encourage the deve l opment of densities to make t he trans i t operations efficient. Obviously, transit alone cannot be relied upon to create t he ma r ket appeal that resu lts in higher density however policy and service commitme nts to transit can be used tO indicate a commit m ent to q ua lity transit service that will encourage dev elopment. Within the transit and economic developmem communities there is a great deal of debate and discussion as to the role of transit in encouraging land use impacts and the magnitude of the land use influe nce associa ted with the various typ es of trans it. The common pe rce ption is that the greater the magnitude of the investment dec i s ion s t he greater the influence will be on la nd use. Thus heavy rail systems would have the most Influence and smaller investmen ts would have correspondingly less influence. The greater the i nvestm e nt, t he greater the assurance that the facility \vill h ave a strong physical presence and that it will continue to operate. In all probability the magnitude of investme nt should be proportional to the expected vol ume of passengers and the enhanced access ibilitY asso cia ted with the system. While the accessibility of transit clearly contributes to the attractiveness of locating i n the proximity, a variety of other attractions also exist. Most frequently the nature of the commitment to a fixed guideway f acility is part of a major urban planning effort. The attractions to the par ties potentially committing to development near transit incl ude improvements in other infrastructure, in cluding streets and sewer and water. Other attractions include ex pedited approvals of plans and permits for development, exemp t ions from various requirements and other incent ive s s uch as higher floor area ratios, easier rezoning, changes in setback requirements, height, parking o r other restrictions, and often good 'viii 'vith city officials. The presence of an incentive to locat e near transit is not t he only factor contributing to the extent of development that ha s followed some transit investments. In very tightly controlled 35
building environments like some Canadian cities limitations on development elsewhere strongly influence the development patterns around transit 36
Appendix A Graphic Summary of U.S. Transit Industry Performance Source: ''The Status of the Nations's Local Mass Transportation: Performance and Conditions," Report of the Secretary of Transportation to the United States Congress, June 1988. L ist of F igures Figur e Page Figure Al Operating Cost per Revenue Vehi cle Hour A-2 Figure A2 -Operating Expense pe r Vehicle Revenue Hour A-3 Figure A3 -Annual Compensation Per FTE Employ ee A-4 Figure A4 -Annual Vehicle Revenue Hours per FTE Employee A-5 Figure AS -Operating Cost Per Passenger Mile A-6 Figure A6 Route Miles Per Square Mile A-7 Figure A7 -Annual Vehicle Revenue Miles Per Route Mile A-8 Figure A8 Passenger Miles Per Revenue Vehicle Hour A-9 Figure A9 -Passenger Mil es Per Vehicle Revenue Ho ur A-10 Figure A10 Passenger Miles Per Capita A-ll F igure All -Operating Expenses Per Passe n ger Mile A-12 Figure A12 Subsidy Per Capita A-13 Figure A13Average Annual Vehicle Miles Per Employee A-14 Figure A14Public Transport Share of Person Trips A-15 Figure Al5 Transit Sbare of Urban Passenger Miles A-16 Figure A16-T ransit Use for Urban Work Trips A-17 A-1
. 80 60 20 0 FIGURE Al OPERATING EFFICIENCY BY AREA SIZE AND REGION GROUP OPERATING COST PER REV VEH HOUR 1985 DOLLARS PER REV VEH HR NORTH MIDWEST SOUTH WEST EAST UZA POPULAT ION PACIFIC COAST '"55m' 50 200K B 200-500K 6J 500K -1M BUS, > 1M B MULTIMODAL, > 1M SOURCE: UMTA STAFF ANALYSIS OF 1980 -1985 SECTION 15 DATA A-2
. A2 OPERATING EXPENSE/VEHICLE REVENUE HOUR 1980-1985 Constant 1986 Dollars 1 Mil 1 Mil soo,ooo-2oo,oooso,ooo-All multi-modal bus only 1 Mil 500,000 200,000 URBANIZED AREA SIZE AND MODE Constant 1985 Dollars $801------, Northeast Midwest South West Pac. Coast All REGION 1980 B 198 1 ID 1982 fi1llll 1983 e;osf 1984 0 1985 SOURCe: UMTA STAFF ANALYSIS OF 1980 -1985 SECTION 15 DATA A-3
FIGURE A3 ANNUAL COMPENSAtiON PER FTE EMPLOYEE 1980, 1984, 1985 Thousands of 1985 Dollars 1 Mil 1 Mil 500,000-200,000-50,000-All multlmodal bue only 1 Mil 500,000 200,000 URBANIZED AREA SIZE AND MODE Thousands ot 1985 Dollars $40 $0 No r t heast Midwest South West Pae. Coast REGION 198o m 1984 liS'iil1985 SOURCE: UMTA STAFF ANALYSIS OF 1980 1965 SECTION 15 DATA A-4 All
FIGURE A4 AN N UAL VEHICLE REVENUE HRS PER FULL TIME EQUIVALENT EMPLOYEE 1980-1985 1,200.------------------------, 1,0001-------aool----1 Mil 1 M il 600,000 200,000 60,000-All multlmod.a l bu only 1 Mil SOO,OOO 200,000 URBANIZED AREA SIZE AND MOD E 1,200.------------------------, 400 Northeast M idwest South West Pac. Coast All REGION 1980 B1e81 8ii!l1;e2 11111111983 ''"""'1984 CJ 11l85 SOURCE: UMTA STAFF ANALYSIS OF 1980 -1985 SECTION 15 DATA A 5
0.3 0.2 0.1 0 FIGURE A5 OPERATING COST EFFECTIVENESS BY AREA SIZE AND REGION GROUP OPERATING COST PER PASS Ml 1985 DOLLARS PER PASS M l NORTH MIDWEST SOUTH WEST EAST UZA POP U LATION PACIFIC COAST llill3 50 200K B 200 500K R 500K -1M B BUS, > 1M MULTIMODAL, > 1M SOURCE: UMTA STAFF ANALYSIS OF 1980 1985 SECTION 15 DATA A-6
FIGURE A6 ROUT E M ILES PER SQUARE MILE 1980-1985 4r------------------------------------------, 1 Mil 1 Mil soo,ooo-2oo,ooo .. mulllmodal bus only 1 Mil 500,000 60,000 -200,000 URBANIZED AREA SIZE AND MODE All Sr---------------------------------------, Northeast Midwest South West Pac Coast All REGION liilil1e8o a 1981 !WI 1 982 11ZI'ihe83 m':'a 1964 CJ 1985 SQ. miles In u r banized area as of 1980 SOURCE: UMTA STAFF ANALYSIS OF 1980 1985 SECTION 15 DATA A-7
FIGURE A7 ANNUAL VEHICLE REVENUE MILES/ROUTE MILE 198iH985 Thousands of Veh Rev Miles 40.----------------------------------------, 1 Mil 1 soo ooo20o,oooso,oooAll multi-modal bua only 1 Mil 600,000 200,000 URBANIZED AREA SIZE AND MODE Thousands of Veh Rev Miles 25.-------------------------------------, Northeast Midwest South West Pac Coast All REGION 1111111980 19 81 6!!1!982 !'!ii!ll!983 ''''"" 1984 CJ 1985 SOURCE: UMTA STAFF A NALYSI S OF 1980 1985 SECTIO N 15 OATA A-8
350 250 200 150 100 50 0 rzGURE AS SERVICE EFFECTIVENESS BY UZA SIZE AND REGION GROUP PASS MILES PER REV VEH HR NORTH MIDWES T SOUTH WEST PACIFIC COAST EAST UZA POPULATION '""""" 50 200K -200-500K B 500K1M. 8 BUS,> 1M MULTIMODAL, > 1M SOURCE: UMTA STAFF ANALYSIS OF 1980 1985 SECTION 15 DATA A-9
FIGURE A9 PASSENGER MILES/VEHI CLE REVENUE HOUR 1980-198 5 350.----------------------------------1 Mil 1 Mit soo,ooo-2oo,ooo-so,oooAll multi-modal bua only 1 Mit 500,000 2 00,000 U RBANIZED AREA SIZE AND MODE 250 Northeast M idwest South West Pac. Coast All REGION 196 0 1981 am 198 2 II'Ei!!i 1963 l;l!ml 1964 0 1985 Source : UMTA STAFF A NALYSIS O F 19 8 0 1 9 8.5 SEC TI ON 15 OATA A -10
FIGURE AlO PASSENGER MILES PER CAPITA 1980 -1985 100 1 Mil 1 Mil 600,000-200,000-60,000-All multt-modal bua only 1 Mil 500,000 200,000 URBANIZED AREA SIZE AND MODE 600.-----------------------------------------, Northeast Midwest South West Pac Coast All REGION 1980 -1981 1982 1983 gg,il1984 CJ 1985 Using t980 urbanized area popul ation SOURCE: UMTA STAF F ANALYSIS OF 1980 1985 SEC T ION 15 DATA A-ll
FIGURE All OPERAT ING EXPENSE PER PASSENGER MILE 1980-1985 C $0.50r $0.00 1 Mil 1 Mil 600,000 200,000 50,000 All muiU-modaJ bus onl y 1 MU soo ,ooo 200,000 URBANIZED AREA SIZE AND MODE C $0.40r $0.361------=------=-.--------1 $0.30 1----.-,-$0.26 $0.20 $0.16 $0.10 $0.06 $0.00 Northeast Midwest South West Pac. Coast All REGION 1eso lml 1981 1!81982 lr2'ZI 1 983 t==i! 1984 D 1985 SOURCE: UMTA STAFF AN AI.YSlS OF 11180 1985 SECTION 1 5 DATA A-12
FIGURE A12 SUBSiDY PER CAPITA 1980-1985 Constant 1985 Dollars $801-1 Mil 1 Mil 500,000-200,00-0-50,000-All multi-modal bua only 1 Mil soo,ooo 200 .000 URBANIZED AREA SIZE AND MODE C $80r Northeast M idwest South West Pac. Coast All REGION Elil1980 -1981 lml1962 -1983 !!! 1984 0 1985 Using 1980 urba nize d area po p ulation SOURCE : UMTA STAF F ANALYSIS OF 1980 1985 SECT ION 16 OATA A-13
FIGURE Al3 TRANSIT LABOR PRODUCTIVITY: AVERAGE ANNUAL VEHICLE MILES PER EMPLOYEE 1965 TO 1983 THOUSANDS OF VEHICLE M I LES PER EMPLOYEE 15 14 13 12 1 1 1965 1970 1975 1980 SOURCE: APTA, TRANSIT FACT BOOKS A-14
FIGURE Al4 PUBLIC TRANSPORT SHARE OF PERSON TRIPS (Excludes walking and bicycle trips) YEAR 1969 1977 ,.... 1983 0 P.T. Share = 3.6 % P.T. Share = 3.0 % P.T. Share = 2.6 % I I I I 50 100 150 200 Billions Of Person T r ips -PUBLIC TRANSPOR T CJ ALL OTHER SOURCE: FHWA, SUMMARY OF TRAVEL TRENDS, 1985 (Corrected) A-15 250
,. .... "' FIGURE Al5 THE ROLE OF URBAN TRANSIT, 1983 TRANSIT SHARE OF URBAN PASSENGER MILES [3s B illion Transit Passenger M i les l Transit 1.95% O ther MOdGS Nonurban Over 50 Miles All Travel 1 ,947 Bi l l ion Passenger Miles Transit 2.77% Other Modes Nonurban A ll Local Trave l 1,367 Bill ion Passenger Miles SOURCE: NATIONWIDE PERSONAL T RANSPORTATION STUDY, 1983 ----Transit 4 .75% Olher Modes -.J v Loca l Urba n T rav el 799 Billi on P ass enger Miles
FIGURE A16 TRANSIT USE FOR URBAN WORK TRIPS BY MARKET SEGMENT, 1970 AND 1980 (DAILY COMMUTERS WITHIN URBANIZED AREAS) A. Central City Homes to Central City Jobs 1970 1980 B. Suburban Homes to Central City Jobs 1970 l + 11.6% 6 7 Mill i o n Tot a l 1980 10.4 M i Ilion To tal C. Central City Homes to Suburban Jobs 1970 [ + 10 %0 4.4 Million Total 1980 [ +5 % 4.9 M illion Total D. Suburban Homes to SuburbanJobs 20.5 Million Tota l 1970 !II 4 % Q 12.8 M i II ion Total 1980 .. 2 "' 0 "' :_ 19.5 Mi II ion T ransit Users D Others SOURCE: UMTA, DEMOGRAPHIC CHANGE AND R E CENT WORK TRIP TRAVEL TRENDS, 1985 A-17 Tota l