USF Libraries
USF Digital Collections

Metro-Dade Transit Agency fare cross-elasticity and allocation

MISSING IMAGE

Material Information

Title:
Metro-Dade Transit Agency fare cross-elasticity and allocation
Physical Description:
x, 79 leaves : ill., forms ; 28 cm.
Language:
English
Creator:
University of South Florida -- Center for Urban Transportation Research
Publisher:
University of South Florida, Center for Urban Transportation Research
Place of Publication:
Tampa, Fla
Publication Date:

Subjects

Subjects / Keywords:
Local transit -- Fares -- Florida -- Miami-Dade County   ( lcsh )
Local transit -- Ridership -- Florida -- Miami-Dade County   ( lcsh )
Genre:
local government publication   ( marcgt )
non-fiction   ( marcgt )

Notes

Bibliography:
Contains bibliographic references (p. 79).
Additional Physical Form:
Also issued online.
Statement of Responsibility:
prepared by Center for Urban Transportation Research.
General Note:
"April 1994."
General Note:
"Prepared for: Metro-Dade Transit Agency, Miami, Florida."

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 025524152
oclc - 668425533
usfldc doi - C01-00101
usfldc handle - c1.101
System ID:
SFS0032214:00001


This item is only available as the following downloads:


Full Text

PAGE 1

METRO-DADE TRANSIT AGENCY FARE CROSS-ELASTICITY AND ALLOCATION Prepared for: Metro-Dade Transit Agency Miami, Florida M E T R 0 D A D E I R \ 'i S I r Prepared by: Center for Urban Transportation Research University of South Florida Tampa, Florida CUTR APRIL 1994

PAGE 2

Metro Dade Transit Agency c/o Metro-Dade MPO Secretariat Office of County Manager Suite 910 Ill N.W. First Street Miami, Florida 33128-1904 (305) 375-4507 Executive Director: Project Manager: Chester E Colby Alex Rey Panama Center for Urban Transportation University of South Florida 4202 E. Fowler Avenue, ENB 118 Tampa, Florida 33620-5350 (813) 974-3120 Director: Project Manager: Project Staff: Gary L. Brosch Daniel K. Boyle Dennis P. Hinebaugh Joel R. Rey Victoria A. Perk Tony Rodriguez Suzanne Dieringer Rebecca Rahirni

PAGE 3

TABLE OF CONTENTS TABLE OF CONTENTS LIST OF FIGURES ........ ... ...... ........................... iii LIST OF TABLES . . . . . . . . . . . . . . . . . . . . . . v PREP ACE . . . . . . . . . . . . . . . . . . . . . vii EXECUTIVE SUMMARY . . . . . . . . . . . . . . . . . . . ix CHAPTER I: Existing Fare Policies/Pass Usage Characteristics . . . . . . . I CURRENT MDT A FARE POLICY . . . . . . . . . . . . . . . I Discounted Cash Fares . . . . . . . . . . . . . . . . . . . 2 Transfers . . . . . . . . . . . . . . . . . . . . . . 2 T ime of Day ....... .................................... Passes . . . . . . . . . . . . . . . . . . . . . . . . Parking . . . . . . . . . . . . . . . . . . . . . . . . Tokens ........... ............ ............ ..... ........ Free Fares ............................... ...... ....... 4 4 4 4 5 Other Service . . . . . . . . . . . . . . . . . . . . . 5 CURRENT MDTA PASS USE . . . . . . . . . . . . . . . . 5 COMPARISONS WITH OTHER SYSTEMS . . . . . . . . . . . . 9 MOTA NEEDS FOR PASS USE INFORMATION . . . . . . . . . . 10 CHAPTER 2: Survey Design and Administration . . . . . . . . . . . . 13 SURVEY DESIGN . . . . . . . . . . . . . . . . . . . . . 13 ADMINISTRATION . . . . . . . . . . . . . . . . . . . . . I 4 CHAPTER 3 : Who Uses Monthly Passes: Results of the Metropass Survey . . 15 INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . 15 CURRENT MOTA PASS USE . . . . . . . . . . . . . . . . . 16 METHODOLOGY . . . . . . . . . . . . . . . . . . . . . 17 Survey Design . . . . . . . . . . . . . . . . . . . . 17 Survey Administrat i on . . . . . . . . . . . . . . . . . 1 8 Survey Response . . . . . . . . . . . . . . . . . . . . 19 S URVEY RESULTS . . . . . . . . . . . . . . . . . . . . 20 Usage Characteristics of Metropass Users . . . . . . . . . . . . 20 Demographic Characteristics of Metropass Users . . . . . . . . . . 22 Trip-Making Characteristics of Metropass Users . . . . . . . . . . 25 CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . 30

PAGE 4

FARE CROSS-ELASTICITY AND ALLOCATION CHAPTER 4: Characteristics of Cash/Token Users . . . . . . . . . . . 33 INTRODUCTION . . . . . . . . . . . . . . . . . . . . . 33 CURRENT MDTA CASHffOKEN USE . . . . . . . . . . . . . 33 METHODOLOGY . . . . . . . . . . . . . . . . . . . . . 34 Survey Design . . . . . . . . . . . . . . . . . . . . . 34 Survey Administration . . . . . . . . . . . . . . . . . . . 35 Survey Response . . . . . . . . . . . . . . . . . . . . 3 5 SURVEY RESULTS . . . . . . . . . . . . . . . . . . . . . 36 Trip-Making Characteristics of Cash/Token Users . . . . . . . . . 37 Demographic Characteristics of Cash/Token Users . . . . . . . . . 40 Awareness of and Willingness to Use Metropass . . . . . . . . . 41 CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . 48 CHAPTER 5: Predicting Changes in Fare Payment Method . . . . . . . . . 51 INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . 51 TfiE DATA ........................ .......... ....... . .... 51 THE MODEL . . . . . . . . . . . . . . . . . . . . . . 53 FINA L MODELS . . . . . . . . . . . . . . . . . . . . . 56 Pass-versus-Cash Model . . . . . . . . . . . . . . . . . . 57 Pass-versusToken Model . . . . . . . . . . . . . . . . . . 58 RES UL TS . . . . . . . . . . . . . . . . . . . . . . . . 58 Validation . . . . . . . . . . . . . . . . . . . . . . . 58 Predictions . . . . . . . . . . . . . . . . . . . . . . . 60 LIMITATIONS OF THE MODELS . . . . . . . . . . . . . . . . 63 CONCLUSIO N S . . . . . . . . . . . . . . . . . . . . . . 64 APPENDIX : INSTRUMENTS . . . . . . . . . . . . . . . . . . . 65 MDT A Pass User Survey . . . . . . . . . . . . . . . . . . . 67 MDTA Cash/Token Survey . . . . . . . . . . . . . . . . . . 71 Sample Letter . . . . . . . . . . . . . . . . . . . . . . 15 NOTES . . . . . . . . . . . . . . . . . . . . . . . . . 79 II

PAGE 5

LIST OF FIGURES LIST OF FIGURES Figure I Travel Diary Portion of the Metro Dade Transit Pass User Survey . . . . . . . . . . . . . . . . . . . . . . . 18 Figure 2 Reasons for Using Mctropass . . . . . . . . . . . . . . 21 Figure 3 Comparative Demogtaphics: Metropass Survey v. Dade Cowtty . . . 24 Figure 4 Trips per Month per Pass User by Mode . . . . . . . . . . 25 Figure 5 Average Trips per Month by Pass Type and by Mode . . . . . 26 Figure 6 Weekday T rips per Month per Pass User by T i me of Day . . . . 27 Figure 7 Percent Distribution of Pass User Trips by Mode . . . . . . . . 28 Figure 8 Percent Distribution of Pass User Weekday Trips by Time of Day . . . . . . . . . . . . . . . . . . . . . . . 29 Figure 9 Weekly Pass User Trip Distribution . . . . . . . . . . . . 29 Figure 10 Survey Respondents by Mode and Time of Day . . . . . . . . 36 Figure I I Fare Payment Method . . . . . . . . . . . . . . . . 37 Figure 12 Frequency of Transit Use . . . . . . . . . . . . . . . 38 Figure 13 Number of Transit Trips the Prev i ous Day . . . . . . . . . . 39 Figure 14 Number of Transit Trips the Previous Weekend . . . . . . . . . 39 Fig ure 15 Comparative Demographics: Cash Token Users v Dade Cowtty . . . 42 Figure 16 Comparative Demograph ics: Cash Token Users v. Metropass Users . . . . . . . . . . . . . . . . . . . . . . . 43 Figure 17 Awareness of Metropass . . . . . . . . . . . . . . . . . 44 Figure 18 Previous Purchase of Metropass . . . . . . . . . . . . . . 44 Figure 19 Reasons for Not Using Metropass . . . . . . . . . . . . . 45 lll

PAGE 6

FARE CROSS-ELASTICITY AND AUOCAT!ON Figure 20 Willingness to Purchase a Metropass at $50 per Month . . . . . . 46 Figure 21 Willingness to Purchase a Metropass at $40 per Month . . . . . . 47 Figure 22 Willingness to Purchase a Metropass at $60 per Month if Cash Fares Rise to $1.50 . . . . . . . . . . . . . . . . . . 47 Figure 23 Factors that Would Encourage Metropass Purchase . . . . . . . 48 IV

PAGE 7

LIST OF TABUS LIST OF TABLES Table 1 MDTA Transfer s . . . . . . . . . . . . . . . . . . 3 Table 2 Cash, Pass, and Token Boardings by Mode . . . . . . . . . . 7 Table 3 Total Pass Sale s by Pass T ype (June 1993) . . . . . . . . . . 8 Table 4 Estimated Metropass Ridership (June 1 993) . . . . . . . . . . 8 Table 5 Mo n thly Pass Fare Compariso n . . . . . . . . . . . . . . 11 Table 6 MOTA Basic Fare Structure (July 1 1993) . . . . . . . . . . 16 Table 7 MOTA Total Pass Sales by Pass Type (June 1993) . ............ 17 Table 8 MDT A Ridership (June 1993) . ........... ...... ......... 34 Table 9 MOTA Cash Boardings (June 1993) ................ ....... 34 Table 10 Estimated C h oice Probabilities of Buying a Metropass . . . . . . 54 Table II Distribution among Metropass, Cash. and Token User s (June 1993) . . . . . . . . . . . . . . . . . . . . . 59 Table 12 Distribution among Metropass, Cash, and Token Users ; Compiled from the Pass-v.Cash and Pass-v -Token Models ................ 59 T able 13 Distribution among Cash and Token U sers . . . . . . . . . . 60 Table 14 Distribution among Metropass, Cash, and Token Users: Price of the Metropass = $40 ; Compiled from the Pass-v.-Cash and Pass-v .To ken Models . . . . . . . . . . . . . . . . . . . . 61 Table 15 Distribution among Metropass, Cash, and Token Users: Price of the Metropass = $50 ; Compiled from the Pass-v.-Cash and Pass-v.Token Models . . . . . . . . . . . . . . . . . . . . 61 Table 16 Distribu tion Among Metropass Cash and Token Users: Price of the Metropass = $60.00; Full Cash Fare = $1.50; Token Fare = $1.20; Compiled from the Pass-v.-Cash and Pass-v. Token Models . . . . . . . . . . . . . . . . . . . . . . 61 v

PAGE 8

FARE CROSS-ELASTICITY AND ALLOCATION Table 17 Distribution among Metropass, Cash, and Token Users: Price of the Metropass = $40; Compiled from the Pass-v. -Cas h and Pass-v Token Models; Adjusted for Stated Preference Responses . . . . . 62 Table 18 Distribution among Metropass, Cash, and Token Users: Price of the Metropass = $50; Compiled from the Pass-v.-Cash and Pass-v.Token Mode l s ; Adjusted for Stated Preference Responses . . . . . 62 Table 19 Distribution among Metropass, Cash, and Token Users: Price of the Metropass = $60.00; Full Cash Fare= $1.50; To ke n Fare= $1.20; Compiled from the Pass-v -Cash and Pass-v.-Token Models; Adjusted for Stated Preference Responses . . . . . . . . . . . 63 VI

PAGE 9

PREFACE PREFACE The Center for Urb an Transportation Research (CUTR) has prepared this report under the tenns of the inter local agreement between the University of South Florida and Metropolitan Dade County This report is in accordance with Work Order Number 7 of the interlocal agreement. Our thanks go to reviewers at CUTR, including Perry Maull and Steven E Polzin Special t hanks to Alex Rey Panama, Pamela Levin, and Elaine Ramirez at Metro-Dade Transit Agency for their review and comments throughout the course of this project. VII

PAGE 10

this page IS blank

PAGE 11

EXECUTIVE SUM.IIARY EXECUTIVE SUMMARY The purposes of this study are to obtain information on the usage patterns, behavior and motivation of Metro-Dade Transit Agency (MOTA) riders who purchase the monthly Metropass and to develop a technique to predict the proportion of riders who choose each of three fare payment modes (pass, token and cash) under various pricing scenarios. Conduct of this study involved a review of existing MOTA fare policy and current pass usage. Two surveys were developed: a Metropass survey of current pass users, and a cash/token users survey of riders paying by other modes. Survey results were used to develop models to predict changes in fare payment methods . The Metropass allows unlimited rides for one month on Metrobus (express or local), Metrorail and Metromover. The monthly Metropass costs $60. A discount pass for the elderly, students and persons with disabilities can be purchased for $30. Corporate passes are also available through area businesses at group discount rates of $54 (for between five and 99 passes) and $52 (for 100 or more passes). County employees may purchase passes for $52. Finally, college and university students are eligible for a $45 monthly pass. The executive summary reports some of the more important and interesting findings of the study. These fmdings include: The majority of passes sold (58 percent) are $30 discount passes purchased by the elderly students and persons with disabilities. Only 18 percent of all Metropass users pay $60 for the regular monthly pass. The single biggest reason for purchasing the Metropass is to save money. Eighty percent of respondents on the pass use survey indic ated this as one reason to purchase a pass. This suggests that the pass is being bought and used by those riding transit most frequently. An average Metropass user makes 83 trips per month. In this context, a trip refers to a boarding of a transit ve hi cle; if a passenger transfers from bus to rail, then this counts as two separate trips. IX

PAGE 12

FARE CROSS-ELAST ICITY AND ALLOCATION Transit usage is highest among pass users who purchase either the $60 Metropass are the $52 pass for County employees. Corporate pass users are much more likely to ride Metrorail than Metrobus. If Metrorai l and Metromover are considered together as a rail mode then the majority of Metropass users (65 percent) ride only on a single mode during the course of a typical month. Cost of the monthly pass and insufficient use of transit to justify the expense are the major reasons for cash and token users not to purchase Metropass. Cash and token users show considerable interest i n a weekly pass. Fifty-three percent of respondents report that availability of a weekly pass would encourage them to purchase a pass Respondents also suggest the need for more convenient sales outlets. Models to predict changes in fare payment method have been developed. Key variables affecting payment method decisions are pass price and the difference in monthly cost between pass and cash or tokens. The models take the legit form. Chapter I of this report presents an overview of existing MDTA fare policies and current pass usage. In Chapter 2 there is a brief description of the des i gn and administration of the pass user survey and cash/token user survey. Results of the pass user survey are presented in Chapter 3 and Chapter 4 contains t he results of the cash/token user survey Chapter 5 describes the development of the models to predict changes in fare payment method. X

PAGE 13

CHAPTER 1: EXISTING MDTA FARES AND PASS USAGE CHAPTER! Existing Fare Policies/Pass Usage Characteristics T he purpose of this study is to obtain information on the usage patterns, behavior and motivation of Metro-Dade Transit Agency (MOTA) riders who purchase the monthly Metropass and to develop a technique to predict the proportion of riders who choose each of three fare payment modes (pass token and cash) under various pricing scenarios. Conduct of this study involved a review of existing MDT A fare policy and current pass usage. Two surveys were developed: a Metropass survey of current pass users, and a cash/token users survey of riders paying by other modes. Survey resu l ts were used t o develop models to predict changes in fare payment methods. Metro-Dade Transit Agency (MDT A) is a division of Dade County in southern Florida M i ami is the major city in Dade County. MDT A's service area is 310 square miles with a population of approximately 1.8 million. The public transportation network operated by MDT A consists of three primary components: Metro bus with 70 local and express routes; Metrorail, a 21.1 heavy rail system with 21 stations; and Metromover, a 1.9-mile, fully automated, double track loop serv ing the Miami Central Business District (CBD). Unlinked passenger trip s for F iscal Year 1992 totaled 71.7 million, with 54.8 million on Metrobus, 13. 9 million on Metrorail and 3.0 million on Metromover. MOT A also operates a demand responsive system for the transportation disadvantaged. Chapter 1 summarizes the existing Metro-Dade Transit Agency (MOTA) fare policies and current pass usage characteri s tics and offers a comparison to other similar systems. The chapter also discusses the types of information need e d by M D T A regarding pass usage CURRENT MOTA FARE POLICY MOT A currently provides live modes of transit service. Local Metrobus service has a $1.25 full cash fare discounted to $0 60 for the e l derly, disabled and students. Express Metrobus service has a $1.50 full cash fare discounted to $0 75 for the elderly, disabled and students. The Metrorail fares are identical to local Metrobus. The Metromover has a $0.25 full cash fare discounted to $0.10 for elderly, disabled and students The tina! mode of transit service

PAGE 14

FARE CROSS-ELASTICITY ,fND ALLOCA T ION deliv e red is the Special Transportation Service (STS). This is a door-to-door service provided for the physically disabled The cost for thi s service is $1.50 (0-10 miles), $2.50 (lJ-20 miles), $3.50 (21-30 miles) and $4.50 (over 30 miles). The fare for STS service is scheduled to chang e in October 1 993 to a flat fare of $2 for trips within Dade County. Discounted Cash Fares To ride the Metrobus at a discounted fare, students in grades 7-12 must obtain a student permit thro ugh the Dade County School system Students in grades 1-6 do not need a permit to ride Metrobus at the reduced fare. To ride the Metrorai l or Metromover at a reduced fare, all students (grades 1-12) must purchase a reduced fare rail permit (school ID required) for $0.50 available at the transit infonnation center at the Government Center Station. Preschoolers less than 42 inches tall can ride the Metrobus, Metrorail, and Metromover free at all times when accompanied by an adult. Senior citizens and disabled riders can obtain a free reduced fare bus permit (proof of age/disability required) at the Government Center Station. This permit enables the patron to pay the discounted fare on Metrobus. A Medicaid card is also valid as identification for reduced fares on.Metrobus. To ride at the discounted rate on the Metrorail and Metromover, se niors and disabled p a tron s must purchase a reduced fare rail perm i t for $0.50 at the Government Center Station. Transfers Table 1 provides the current transfer rates between Metrobus Metrorail and Metromover. N o transfers are available for the STS, as it is operated independently as a door-to-door service Transfers from bus to bus are valid for two hours after the time they are punched, and are limited to use at locations where routes meet or cross each other or where bus routes link with the Metrorail system. Patrons cannot transfer to the same r oute in either an ongoing or backriding direction, though transfers can be made at transfer points to a new route going in the same direction as the one the patron transferred from. A morning rail transfer (bus to rail) is valid until 12:00 noon, whereas an afternoon rail transfer is valid from noon until 1 2 :00 midn i ght. Rail to bus transfers must be purchased at the station where the passenger boards the train. Bus operators will not accept transfers from stations where the passenger 2

PAGE 15

CHAPTER 1: EXISTING MDTA FARES AND PASS USAGE exits the train. Transfers are good for only one trip, although a transfer can be purchased on a transfer Rail to mover transfers occur at the Goverrunent Center Station via a direct connection within the fare control area. Table 1 MOTA Transfers : ; '. . ..: Transfer. .:ra :. , ; ... Loca l Bus Local Bus $0 .25 Local Bus Metrorail or Mover $0.25 Local Bus E>
PAGE 16

FARE CROSS ELASTICITY AND ALLOCATION Metro bus pay a fare upgrade of $0.40 for l ocal bus service and $0.65 for express bus service. The BCT transfers are accepted only at specified inter-county transfer points. Time of Day Like most transit systems in the U.S. the MDTA does not currently have a peak period surcharge for any of their modes of transit service. MDT A fares remain the same regardless of time of day or day of week. Passes The cost of a MDTA monthly Metropass is $60. A discount pass for the elderly disabled, and students is available for $30 The Metropass allows unlimited rides for one month on the Metrobus, Metrorail, and Metromover. The pass is also valid on express bus service at no additional cost MDT A sells Metropasses at a group discount rate of $54 for between 5 and 99 passes, and $52 for I 00 or more passes Passes can be directly purchased at the Goverrunent Center Station or at all MDT A sales outlets. Group passes are sold through the mail, with orders received by the I Oth of the preceding month. College students may purchase a monthly Metropass for $45 at participating colleges and universities. Students must be registered full time at the time of sale and must be regis t ered at the institution wh ere the pass is purchased. Parking Most Metrorail stations have parking facilities available for rail riders. The daily parking rate is $1. Metropass ho l ders can purchase a $2 parking permit with their pass that allows unlimited use of the rail parking facilities for that month. Tokens MOTA sells tokens at I 0 for S I 0 These tokens can be used as full fare on both Metro bus (local) or Metrorail. To use tokens on express service, patrons must also pay a $0.25 upgrade. Tokens are available at all rail stations as well as at all MOTA pass sales outlets. 4

PAGE 17

CHAPTER 1: EXISTING AfDTA FARES AND PASS USAGE Free Fares Currently, only MDT A employees, retired MDT A employees (having had 30 years or longer of service) and their spouses, and various uniformed officers (police, correction officers, court bailiffs, and firemen) with proper identification can ride for free on Metrobus. No such policy currently exists for Metrorail or Metromover. Other Service MDT A a lso offers a Brickell shuttle service for a full cash fare of $0.25, and a discounted fare of$0.10. Park-and-ride service for special events (foo tball, tennis, etc .) is also offered. Fares for special events differ by each event and its location Table 2 summar izes the distribution of hoardings by fare payment method and by mode from October 1990 to June 1993.1 As can be seen from the table, token use on both the bus and rail has increased since its inception in December 1990. Token use on the Metrorail has more than tripled during this period. No significant seasonal or overall trends have been noticed on pass usage. The pass category includes special day pas ses signature passes, and T ri-Rail passes along w i th the Metropass, and so is not a pure measure of Metro pass usage. From rail and mover data (bus data are not broken down by type of pass), it appears that Metropass accounts for approximately 85 percent of total pass use reported. CURRENT MOTA PASS USE The following observations are made of current MOTA Metropass usage based on information supplied by MDTA and derived from the Ridership Report and Ridership Technical Report Table 3 summarizes the pass sales by pass type for June 1993-' Using June 1993 as an example month: 83% of corporate Metropass sales include a rail parking permit; therefore it can be assumed that at l east 83% of corporate pass buyers ride Metrorail. 14% of outlet pass sales include a rail parking permit. 5

PAGE 18

FARE CROSS-ELASTICITY A N D ALLOCATION 82% of outle t sales are discounted passes. In total 58% of passes sold are $30 discount passes. The average purchase price of a pass was $40.36. Assuming the 21-,28, and 24 percentage pass boardings by mode stated abo ve and 17,907 Metropasses sold, on average 83 trips per pass were taken on the three modes. The average pass rider would have made 62 bus, 19 rail, and 3 mover trips in the month of June 1993, as shown in Table 4 These nwnbers include all pass h oard ings, not only Metropass, and s h ould be considered rough estimates. 6

PAGE 19

CHAPTER 1: EXISTING MDTA FARES AND PASS USAGE Table 2 Casb, Pass, and Token Boardings by Mode ; 2' :: ; ; i Me!JOraiiA 'MeiFOmovar ' <:aal) IPan.l cash I Pass! I Token caah: 1 pasfi MOnth < >:' ; % ',,."!:;{, I ' '; :%' ,. : : . ; "''1.0: Oct 90 80 20 67 33 74 26 Nov 90 80 20 68 32 74 26 Dec 90 76 22 2 66 30 4 79 21 Jan 91 77 21 2 63 32 5 76 24 Feb 91 76 22 2 62 32 6 75 25 Mar91 76 22 2 62 32 6 75 25 Apr 91 76 21 3 62 32 6 76 24 May 91 76 21 3 61 32 7 75 25 Jun 91 76 21 3 63 31 6 76 24 Jul 91 77 20 3 63 30 7 78 22 Aug 91 79 19 2 64 29 7 78 22 Sep 91 77 20 3 61 32 7 72 28 Oct 91 76 21 3 60 33 7 75 25 Nov 91 76 21 3 62 31 7 77 23 Dec 9 1 78 1 9 3 64 29 7 80 20 Jan 92 77 20 3 62 30 8 77 23 Feb 92 76 20 4 61 31 8 78 22 Mar 92 76 20 4 58 31 11 76 24 Apr 92 75 21 4 57 31 12 72 28 May 92 75 21 4 56 31 13 72 28 Jun 92 75 20 5 56 30 1 4 70 30 Jul 92 77 19 4 56 30 14 75 25 Aug 92 73 23 4 57 29 1 4 79 21 Sep 92 73 23 4 57 29 14 79 21 Oct 92 73 23 4 58 28 14 73 27 Nov 92 74 22 4 58 28 14 66 34 Dec 92 76 20 4 60 27 1 3 76 24 Jan 93 76 20 4 58 29 1 3 72 28 Feb 93 75 20 5 57 29 14 71 29 Mar 93 74 21 5 55 30 15 67 33 Apr 93 74 21 5 55 30 15 69 3 1 May 93 74 21 5 56 29 15 71 29 Jun 93 74 2 1 5 57 28 1 5 76 24 Source: Metro-Dade Transit Agency, MDTA Transit Ridership Report Miami, FL: MOTA, Monthly Reports from October 1990 through June 1993. 7

PAGE 20

FARE CROSS-ELA S TICITY AND ALLOCATION Table 3 Total Pass Sales b y P ass T ype (June 1 9 93) ;<(:,. ;, . ) \; i / : l ; "f;otatt: t ; ; P a s mr-1?1t'% 'C;t '' )/ . Regular Pass 3,14 3 1 7.55% $60 $188,580 26.09% Discount Pass 10,374 57.93% $30 $311,220 43.06% Corporate Grou p (5-99) 1 235 6.90% $54 $66,690 9.23% Corporate G roup ( 1 00+) 1 245 6.95% $52 $64,470 8.92% College/Un i versity 1 ,067 5 96% $45 $48,015 6.64% CountyfTransrt Employees 843 4 71% $52 $43,8 3 6 6 06% T otal 17, 907 100. 00% ' $722 811 100. 00% ... ''. Ave rage Cost per Pass $40.36 ' ., Source: Metro-Dade Transit Agency, Summary o f Monthly Pass Sale s by Pass Type and Outlet: June 1993, M iami, FL: MO TA, 1993 . Mode. < Bus Rail Mover Total .. Table 4 Estimate d M e trop as s Rid e rship (June 1 993 ) Jline1 993. . Total No: % Pus Use o f Pass .,, Ridership. ., ' B 0ard l ngs 5 ,289,640 21% 1,110,824 1 ,209, 370 28% 338 ,624 208 593 24% 50,062 6 707 ,603 22.1% 1,499,510 . A verage : Trips;per Pass 62 1 9 3 8 4 Sources: Me t ro-Dade Transrt Agency MOTA Transit Ridership Report M i ami, FL: MOTA Mon th l y Reports from O c tober 1990 through June 1 993 ; a n d Met ro-Dade Transit Agency. Ridership Technical Report: Metrobus Metrorai/, Metromover, Peratransit June 1993 Miam i FL: MOTA, 1993 8

PAGE 21

CHAPTER 1: EXISTING MDTA FARES AND PASS USAGE Other items of importance include: County personnel can pay for the monthly pass through payroll deduction and receive the pass in the mail. Purchase of corporate passes at work sites is inconvenient, with orders needing to be placed a month in advance. COMPARISONS WITH OTHER SYSTEMS CUTR recently completed a survey of the 17 largest transit systems for information related to both the Fare Cross-Elasticity and the Fare Policy projects with MDTA.4 Table 5 presents base and monthly pass fare information for MOTA and similar transit agencies. As can be seen from this table, MDTA is one of only a few agencies not offering a discount on its base monthly pass for both bus and rail modes. The breakeven number of trips for both bus and rail is 48, higher than most other agencies. It must be not ed, however, that Metrorail users save $1 a day in parking costs when they buy a $2 monthly parking pass along with the purchase of a Metropass. Many MDT A patrons must transfer when travelling by bus or rail. With a transfer, the average fare per trip would be $0.75. This would create a breakeven point for number of trips at 80 unlinked trips, but only 40 /inlaid trips. For rail patrons who use the paid parking facilities, the breakeven point of a monthly Metropass and monthly parking permi t combined ($62) would be 35 trips, based on an average fare of $1.75 ($1.25 fare+ $0.50 average per trip for parking). Other systems were contacted regarding their methods of allocating fare box revenue by mode and forecasting ridership by fare payment method. All transit agencies contacted currently use exi sting ridership information to allocate fares by mode. None of the systems surveyed used any formal method to predict ridership by fare payment method 9

PAGE 22

FARE CROSS-ELASTICITY AND ALLOCATION MDTA NEEDS FOR PASS USE INFORMATION First, with regard to allocation, MDT A indicates that there is currently no good information on the reasons why people buy passes; the breakeven point for different users ; where and how the passes are used; and whether usage is most sensitive to price, convenience, o r other variables Farebox and turnstile data currently provide MDTA with the number of hoardings by pass for each mode. No information is available on the pass holder or on linked trips. Chapter 2 presents the method used to survey current pass and cash users by mode. The second issue is forecasting pass usage. Current MDT A forecasting models assume a constant relationship between the price of the monthly pass and the base cash fare. MDT A would like to integrate cash and pass prices directly into the models and, thus, be able to foreca st the ridership impacts for separate changes in cash and pass prices. The survey results from this and other studies will be used to estimate cross-elasticities by fare payment method and the likely results of cash/pass price differentials. 10

PAGE 23

Table 5 Monthly Pass Fare Comparison -' -' ---v,> -.,_. v <> , ,. ..,. 'l!;,;;j< "< '<;._. -,_,_ Moll!llly .PAA.t :t , ,, Adult Base Fare' Monthly_ < < -.;;. z->c. .. t N if 'u' ;;.: <: -l;- ;>< MotOr Bu ,, "' System "' ' .* -' -' Breakeven :; -W '' Motor Bus Rail'-:, Motor Bus Rail'' c'f i'". < Cost -' '><<' ,, '' ()$ -< T!ips' N"" .. > '' Atlanta, GA $1. 25 $1.25 22% 22% $43 34 $43 34 Baltimore, MD $1.25' $1.25z 24% 24% $42 34 $42 34 Boston MA so.eo' $0.85' 24% 28% $20 33 $27 32 Ch ica go, IL $1. 50 $1.50 8% 18 % $78 52 $78 52 Cleve l and, OH $1.25' $1.5o' 18% 18% $45 36 $54 36 Dallas, TX $0.75 30% $23 31 Los Ange l es, CA $1.10 $1.10 13% 13% $42 38 $42 38 I $1.25-' .. ' 48' -Miami, FL $1.25 -9% % $60 -New Je rsey S1.0o' $1 .0o' 7% 7% $4 1 41 $4 1 41 New Yo rk, NY $1.25 $1.25 NA NA NA NA NA NA Philadelphia PA $1.50 $1.50 12% 12% $58 39 $58 39 PA $ 1 .25z $1 .25' 27% 27% $40 32 $40 32 P ortland, OR $0.95' $0.95' 26% 26% $3 1 33 $31 33 San Anton io, TX $0.4o' 9% $16 40 San Franc i sco, CA $1.00 $1.00 27% 27% $32 32 $32 32 San Jose, CA $1.00 $1. 00 32% 32% $30 30 $30 30 Wash i ngton, DC $1.00Z $1.0o' NA NA NA NA NA NA March 1, 1993 Source: Center tor Urban Transportation Research, Fare Policy Work Order March 1 1993 Comparison to 44 base tare trips/mont h "Rail" does not include commuter rail Technical Memorandum Number 1 Tampa FL: prepared for MOTA z Additional distance o r zone cllarge by CUTR. July 1993.

PAGE 24

this page IS blank

PAGE 25

CHAPTER 2: PASS AND CASH/TOKEN SURVEYS DESIGN CHAPTER2 Survey Design and Administration Chapter 2 describes the development of the cash/token and pass user surveys, and the procedures for conducting each survey. In developing both the cash/token and pass user surveys, the Center for Urban Transportation Research (CUTR) surveyed peer transit systems as well as those systems known to have recently conducted similar analyses. As a result of this analysis, it was decided that surveys from Chicago and Toronto would be used as models in the development of the MOTA survey instruments Copies of both the cash/token and pass user surveys may be found in the appendix. SURVEY DESIGN The main objective of the pass user survey for MDT A is to obtain travel information by time of day and by mode, using a seven-day diary. Information regarding the length of pass use, reasons for purchasing a pass, pass use in conjunction with a parking permit, and demographic characteristics are also obtained. Analysis of the survey results focuses on how pass usage differs by mode or combination of modes used, by pass type, by geographic location, by demographic characteristics, and by time of day. The end results include a profile of Metropass users and a detailed description of pass usage. The main objectives of the cash/token user survey are to gain an understanding as to why patrons do not currently purchase a Metropass and what factors would influence their considering a future purchase. Frequency of system use and standard demographic information are also obtained from the survey. The survey results and information f rom other studies are combined to estimate cross-elasticities based on cash/pass price differentials. Demographic information re quested in the two surveys mirrors that included in the on-board survey, conducted during the months of May and June 1993. Both surveys were printed in English and Spanish. 13

PAGE 26

FARE CROSS -ELAST ICITY AND AUOCATION ADMINISTRATION The pass user survey was administered in May 1993, when patrons received their June Metropass. MDT A personnel delivered surveys to the pass outlets when delivering June passes The outlets were requested to hand out the surveys to each person purchasing a June pass. Also, surveys were given to pass purchasers at the Government Center outlet and were distributed to county employees along with their pass. Finally, corporate buyers were provided with surveys to distribute to pass purchasers. The cash/token user survey was administered during two weekdays in June by CUTR personnel. Random runs on four bus rou tes and the morning peak period and midday period on the rail system were surveyed. The appendix contains a sample letter distributed to pass sales outlets and corporate sales contacts requesting their assistance in distributing the pass user surveys along with the June 1993 Metropass. As an incentive to increase the response rate of the pass user survey, a tota l of I 00 monthly Metropasses were awarded to those patrons who participated in the survey. Patrons were instructed on the survey to write the pass number from the middle of their Metropass in the space provided on the survey. One hundred winners were randomly drawn from the returned surveys. Listings of the winning pass numbers were then mai led to outlets, corporate offices, and to county personnel pass holders. Outlets and corporate purchasers were reimbursed for the free passes awarded upon receipt of each winning pass No survey incentives were used on the Metrobus for the cash/token survey. Initially on Metrorail, cash/token surveys were distributed at selected stations without incentives. The response rate on Metrorail was very poor, due in part to the timing of this survey just after the on-board survey and in part to the absence of survey personnel on the train s to encourage completion of the survey. Thus, Metrorail was resurveyed by on-board personnel on selected trains. For the resurvey, a token was offered as incentive to complete and return the survey. 14

PAGE 27

CHAPTER 3: PASS USER SURVEY RESULTS CHAPTER3 Who Uses Monthly Passes: Results of tbe Metropass Survey Chapter 3 reports the conduct and results of a survey of monthly pass users on the MOTA transit system. In conjunction with a broad review of its fare policy, MOTA wished to gather detailed information on the demographic characteristics of Metropass users, reasons affecting the decision to purchase a monthly Metropass, and travel patterns of pass holders. Farebox and turnstile data provide i nformation on the number of boardings using Metropass for each mode, but MOTA needed more detailed information focused on the pass user. INTRODUCTION Unlimi ted-use monthly passes are available in over two-thirds o f the transit sys tems in the United States.' These passes are seen as a way to attract and retain riders by offering convenience and a discount over the regular fare Unfortunately the revenue impacts of unlimited use passes have not been positive, especially in c ases where large discounts have been offered. Table 6 shows the current MDTA fare structure. The base fare is $1.25, with a $0 .25 charge for Metromover. There is a slight surcharge for express bus service Transfers within or between modes are $0.25, with the exception of transfers betwee n Metrorail and Metromover The rail-to-mover transfer is free, whereas the mover-to-rail transfer requires a $1.00 payment. MOTA has been influenced by the work of Oram7 in establishing its pricing policy to address specific markets Tokens are the discounted fare media with ten tokens priced at $10, a 20 percent discount. A prepaid discount is intended to encourage additional use of the transit system by infrequent riders. As noted previously, the monthly Metropass is targeted at very frequent riders and is accordingly priced at a high le vel. The $60 charge, equivalent to 48 full-fare trips, is one of the highest for a monthly pass in the United States. 15

PAGE 28

FA .RE CROSS-ELASTICITY AND ALLOCATION Table 6 MDT A Basic Fare Structure (July 1, 1993) '<. '-if ''-lt"'t-<'S.' > A-f' >t,_ -... . .. ,;:. !<-<:< __ --<, -; .. : 0; '.; ; '' ':r ;;:: ,1;, :, A '':,: . r" "-<>. H '<> "' ., A.. .. , ''. : .. ,. .. Full : :; ,, .. ,., '.' i:: J DI8counfild.,; Metrobus $1.25 $0.60 Metrorail $1.25 $0.60 Metromover $0 25 $0.10 Express Bus $1.50 $0 .75 Transfer $0.25 $0 .10 Metropass $60.00 $30.00 Corporate Group (5-99) $54.00 NA Corporate Group ( 1 00+) $52.00 NA College/University $45.00 NA Token 10 for S10.00 NA Passes can be directly purchased at the Government Center Station or at approximately 80 sales outlets. Half of these sales outlets are check-cashing establishments; no agreement has been reached with pharmacy/supermarket chains to sell passes at all of their stores in Dade County. Group passes are sold through the mail bU1 orders must be received by the lOth of the preceding month. CURRENT MDTA PASS USE According to June 1993 ridership data, 21 percent of Metrobus hoardings are by pass whereas pass usage accounts for 28 percent of hoardings on Metrorail and 24 percent on Metromover. Trends over the past three years indicate that the level of pass usage has remained fairly constant by year as well as by month, as shown in Table 2 of Chapter I. It has already been noted in Chapter 1 t hat these numbers include passes other than Metropass. I nterestingly, tokens have not achieved significant market penetration accounting for only 5 percent of Metrobus hoardings and 1 5 percent of Metrorail hoardings in the month of June. 16

PAGE 29

CHAPTER 3: PASS USER S U RVEY RESULTS T a b l e 7 summarizes the Metropass sale s by pass type for June 1993. Fo r the month of June, 17,907 passes were sold of which 6 4 pe r cent were discounted (e l derly, disabled, and srudents). The $60 pas s acc o un ted for o n ly 1 8 percent of all sales. Average cost per pass was $40 36. . PanTpe Regula r Pass D iscou nt Pass Corpora t e G r o u p (5-99) Corporate G r oup (100+) College/U ni versity CountyfTransit Emp l oyee Total T a bl e 7 M DT A Total Pass Sales by Pass Type ( J u n e 1993) Number Sold Peicentot Tota l 3 143 17. 55% 10, 374 57.93% 1 235 6.90% 1,2 4 5 6 95% 1 067 5.96% 8 4 3 4 7 1 % 17,907 100.00% Averag e Cost P e r Pass METH ODOLOGY S u rvey D esign . -, CostPerPass $60 $30 $54 $52 $45 $52 . ,;; $40.36 To obtain demographic and usage characteristics of existing Metropass holders, CUTR deve l oped a pass user survey. In deve l oping this survey, CUTR surveyed peer transit systems as well as those known to h a v e recently condu c ted similar analyse s. A s a result of this analysis, it was decided that recently co m p l et e d surveys from Chicago and Toro n to would be u sed as m odels in developing t he MDT A pass user survey ins t rument. 9 10 A copy of the MDT A pass use r survey is included in the appendix As noted i n Chapter 2, the main objective of the pass user survey for MDT A was to obtain travel information by time of day and by mode, using the seven-day diary shown in Figure I. Information regarding the length of pass use, reasons for purchasing the pass, whether t he 17

PAGE 30

FARE CROSS-ELASTICITY AND AUOCATION pass is used in conjunction with a parking permit, and standard demographic characteristics were also obtained. Figure 1 Travel Diary Portion of the Metro-Dade Transit Pass User Survey METRO-DADE TRANSIT PASS USER DIARY Time Started 4:00 a,m, :45 a m 10:46 a m -1 :4S p m 1 :46 p.m.-6:45p.m. 6:46 p.m. 3:00 a.m. Survey Administration The pass user survey was administered in May 1993, when patrons received their June Metropass. MDT A personnel delivered surveys to the major pass outlets when delivering June passes. The outlets were requested to hand out the surveys to each person purchasing 18

PAGE 31

CHAPTER 3: PASS USER S URVE Y RESULTS a June pass. Also surveys were given to pass purchasers at the Government Center outlet Finally, corporate buyers were given surveys to distribute to pass users. The survey form requested that patrons return their completed pass user surveys to the bus operator, at survey return boxes at the Metrorail stations or by mail to the address highlighted on the survey form. Postage was not prepaid MDTA bus operators returned the surveys to the garage for transport to the downtown administrative office Security personnel at the rail stations collected the surveys on a daily basis and returned them to the downtown administrative office. As was anticipated, surveys continued to come in thro ugh the first two weeks of June. MDT A agreed to provide I 00 free Metropasses as an incentive to increase the response rate of the pass user survey. Patrons were instruc ted to write their June Metropass number in the space provided on the survey One hundred winners were randomly drawn from the returned surveys Listings of the winning pass numbers were then mailed to outlets, corporate offices, and to county personnel pass holders. Outlets and corporate purchasers were reimbursed for the free passes awarded upon receipt of each winning pass. Survey Response A to tal of approximately 15, 000 surveys were provided to major pass outlets and corporate offices along with the supply of passes for June. It is not possible to determine how many of these surveys were actually distributed, but it appears that most outlets were cooperative. By the third week of June, 450 usable surveys had been returned. Of these, 89 did not include travel diary data, resulting in a total of 361 surveys with completed travel diaries. The lev el of precision for a sample of this size is .2% at the 95 percent confidence level. It should be noted that early analysis of the survey data ind i cated that Metropass purchases were almost evenly split among regular, corporate, and discount passes (37 percent regular, 32 percent corporate, 29 percent discount). This distribution however, did not accurately reflect actual pass sales for June 1993. Because of the potential differences in demographics and travel characteristics among users of the three pass types the decis i on was made to weight the data according to the number of each type of Metropass sold during June The factored data would then reflect the actual distribution of pass users by type of pass purchased 19

PAGE 32

FARE CROSS-ELASTICITY AND ALLOCATION (22 percent regular, 14 percent corporate, and 64 percent discount). Those buying the discount pass include students, elderly, and the disabled, whereas county transit employees were included in the regular pass category. It is believed that this weighting scheme provides a more accurate representation of Metropass users. SURVEY RESULTS The instrument used to conduct the survey was structured so that information on Metropass users could be collected in three primary areas: usage characteristics, demographic characteristics, and trip-making characteristics. Metropass usage information and user demographics were obtained via 12 survey questions that required respondents either to provide a brief answer or to select an answer from a preset, multiple-choice list of responses. Determination of the trip-making characteristics of the Metropass users necessitated the use of the multi-celled travel diary illustrated previously in Figure I. Survey participants were instructed to list all the trips that they made using the Metropass on the bus, rail, and mover modes. The diary provided for trips to be listed by day of week and time of day for a seven day period. The resulting survey data was analyzed using the SAS System for Personal Computers, Release 6.04. Various SAS procedures were utilized to calculate survey variable frequencies as well as crosstabulations of multiple variables to gain a better understanding of the relationships between certain usage, demographic, and of the Metropass users. Selected results from these analyses are presented and discussed in the following three sections. Usage Characteristics of Metropass Users The second question on the survey, which asked respondents how long they had been using the Metropass, produced an interesting distribution. The largest group of survey respondents were those who have been Metropass users for approximately one to two years (24 percent). New Metropass users ( zero to six months) accounted for 10 percent of the respondents. T he most surprising result was that 20 percent of the respondents have been using the pass for longer than seven years, or since it was first issued This result shows a high degree of loyalty and stability among Metropass users. 20

PAGE 33

CHAPTER 3: PASS USER SURVEY RESULTS Figure 2 shows survey responses concerning reasons for using the Metropass. The percentages in the distribution swn to more than 100, since respondents were asked to check all answers that applied. As expected, the most popular reason for using the pass was to save money: 80 percent of survey respondents checked this choice. In the survey design a conscious decision was made not to include "convenience" as a suggested answer to this question but, instead to provide more specific choices. Taking extra rides also a high frequency of responses (71 percent), suggesting that the use of the Metropass encourages increased ridership on the transit system. Reasons related to exact change and transfer cost were mentioned less often. Respondents were able to check "Other" as a reply and a blank was provided so the other reasons could be given. These ot her factors i nc luded practicality, convenience, and saving time. Some pass users indicated that the Metropass is an employee benefit: their employers purchase the pass for them. Figure 2 Reasons for Us ing Mdropass To Save Money Avoid Transfer Charges Take Extra Rides I Might Not 011\eowise Take No Need for Exact Change Other 0% 20% 40% 60% 80% Metropass purchasers have the option of purchasing a Metro rail monthly parking permit for an additional $2 (the daily charge is $1). Nearly 40 percent of pass purchasers also bought the parking permit. Among corporate pass res pondents, 73 percent purchased the parking permit. 21

PAGE 34

FARE CROSS-ELASTICITY AND ALLOCATION Demographic Characteristics of Metropass Users Demographic data on the age, gender, ethnicity, number of household vehicles, and annual household income of the respondents were collected from the second group of questions on the survey. These results are shown in Figure 3 and can be compared with similar demographics gathered from Dade County's 1990 Census data shown in the same figure. The age distributions from the Dade County Census data and the survey are similar with the exception of the IS-and-under and 65-and-over categories. In the pass user survey, only I percent of the respondents are age 15 or younger. Youths generally account for only a very small proportion of monthly pass users. The largest percentage of survey respondents (35 percent) fall into the 65-and-over age group, whereas only 14 percent of the population of Dade County is 65 or over. The large number of elderly discount pass users (52 percent of discount pass users are age 65 or over) accounts for much of this difference. When considering ethnic origin, the major differences between the survey results and the Census data occur in the percentages of whites and Hispanics. In Dade County, 30 percent of the population is white, but 46 percent of the surve y respondents (and 60 percent of the corporate pass users) are white. Similarly, the Census data indicate that 49 percent of the population is Hispanic, whereas Hispanics comprise only 34 percent of survey respondents. Some of this d iff erence may be attributed to the mixture of race and ethnicity in the phrasing of this question. It does appear, however that whites and to a lesser extent, African Americans are overrepresented among Metropass users. The male/female ratio of Metropass users is very similar to that of Dade County as a whole. This is somewhat uneKpected, inasmuch as user profiles for most transit systems show a larger proportion of females. Females account for nearly 60 percent of regu lar pass users and over 70 percent of corporate pass users, whereas a slight majority of discount pass purchasers are male. When comparing the number of vehicles owned by households in Dade County and by Metropass user households, it is evident that the level of vehicle ownership i s lower among pass users. Forty-two percent of the survey respondents are in households owning zero vehicles, whereas in Dade County only 16 percent of the households own zero vehicles. Over 22

PAGE 35

CHAPTER 3: PASS USER S U RVEY RESULTS half of the discount pass users are in household s owning zero vehicles. At the other end of the scale, 39 percent of regular pass holders and 61 percent of corporate pass ho lders are in households owning at least two vehicles. Finally, the greatest number of respondents to the pass user survey (32 percent) have a household income level below $10 000. This is a higher figure than the percentage for the same level of household income in all of Dade County (19 percent) Again discount pass purchasers skew the overall results; 44 percent of this group report a household income below $10 000. lncome distribution of the regular pass holders is very similar to the distribution for Dade County, whereas 40 percent of corporate pass holders have household incomes of $60 000 and over. 23

PAGE 36

FARE CROSS -EUSTICITY AND ALLOCATION F igure 3 Comparat ive Demograph ics: Metropass Survey v. Dade County Metropass Survey Age Dade Coun ty Two 24% 15 &under 1610 19 201029 301039 401059 601064 85 &over Ottlor African American Olher 7% White ol6% M ale 46% 13% Three or Man> 7% Ethnic Origin Gender 2% Two 33% Household VehiCles Ono 27% To tal Annual Household Income 2 4 40% Atrk:an American 19% White 30% Male 48% Three or More 14%

PAGE 37

CHAPTER 3: PASS USER SURVEY RESULTS Trip-Making of Metropass Users Analysis of the travel diary data focused on the number of trips taken by Metropass users during a seven-day period, by mode and time of day. From the data, it was calculated that an average Metropass user made approximately 83 unlinked trips for the month of June. As shown in Figure 4, almost one-half of these trips are made using the Metrobus mode. This breakdown by mode based on survey results is different from the calculated number of trips per month by mode using MDTA farebox data (see Table 4 in Chapter 1) The overall far ebox estimate of trips per month was almost exact (84 trips), but bus usage was estimated to be much higher (59 trips) and rail/mover usage lower (19 and 3 trips respectively). Many transit systems report problems with the reliability of farebox data. Turnstile counts are generally more accurate but mover hoardings that are transfers from Metrorail are not counted because this transfer is free. Calculations using on-board survey data 11 support the results of the pass user survey. Figure 4 Trips per Month per Pass User by Mode 100.0.--------------------, 80.0-t--------------25

PAGE 38

FARE CROSS-ELASTICITY AND ALLOCATION Figure S presents trip-making behavior by mode for each pass rype Corporate and discount pass users both average 79 unlinked trips per month, but the distribution of these trips by mode is very different. Over 50 percent of corporate pass users trips are on Metrorail, with an additional 25 percent on Metromover, whereas discount pass users make 57 perce n t of their trips on Metrobus. The regular pass holders average over 90 unlinked trips per month, with a distribution very similar to tjle overall distribution among modes. FigureS Average Trips per Month by Pass Type and by Mode tD ... .... 1111 .,_ I '"'" Some consideration was also given to weekday trip-making (Monday through Friday) versus travel on the weekend. Survey results indicate that 89 percent of the monthly trips were made on weekdays, with the remainder of the trips being taken on Saturdays and Sundays. In Figure 6 the weekday trips are illustrated as well as the distribution of these trips by the time period during which they began. It should be noted that the four time periods indicated on the travel diary (shown previously in Figure I) represent Metro-Dade's morning (4:00 a.m.-10:45 a m ) midday (10:46 a m -1:45 p m ) afternoon (1:46 p m.-6:45 p.m ), and 26

PAGE 39

CHAPTER 3: PASS USER SURVEY RESULTS evening (6:46 p.m -3:00 a.m ) service periods as defined by the transit agency for the purposes of Section 15 reporting. Figure 6 Weekday Trips per Month per Pass Use r by Time of Day 60.0,---------------------, 73.6 Further analysis of weekday-versus-weekend trip-making indicated that the Metrobus mode is used more frequently by pass users than the other two modes during the weekend than for wee k day travel. Forty-six percent of all weekday pass user trips are by bus, whereas 74 percent of these patrons' weekend travel i s made us i ng this same mode. This increase is presented in Figure 7, along with the weekday and weekend trip distribution data for the rail and mover modes. Similar ly, the time-of-day distr ib ution of weekday pass user trips was also examined by mode and the result s are di s played in Figure 8. According t o the data, the majority of weekday trips occur during the a.m and p.m peak periods for all modes. This makes intuitive sense, given that a high proportion of weekday transit trips are home-based work trips and would typically occur during the peak hours of travel. In addition, it is a lso interesting that the 27

PAGE 40

FAR CROSS-LASTJCI1'Y AND ALLOCATION time-of-day results evidenced a large number of midday Metromover trips, possibly indicating a significant utilization of this mode for lunch trips by employees within the CBD. Figure 7 Percent Distribution of Pass User Trips by Mode Finally, a "mode usage" code was established to indicate how pass users utilize the system. For example, pass users who made all trips by bus were classified as "all bus." Similarly, pass users making 67 to 99 percent of their trips by rail (Metrorail plus Metromover) were classified as "predominantly rail." Figure 9 indicates that 65 percent of pass users utilize only one mode (37 percent rail only, 28 percent bus only). Mode usage by regular pas s purchasers is very similar to the overall percentages shown in Figure 9. Corporate pass users are more likely to ride only rail (62 percent), whereas discount pass users have a greater tendency to ride only bus (37 percent). 28

PAGE 41

CHAPTER 3: PASS USER SURVEY RESULTS Figure 8 Percent Distribution of Pass User Weekday Trips by Time of Day Figure 9 Weekly Pass User Trip Distribution AI 81.11 (100% of 0% 10% 20% NOTE: Trips on Sallltday ancl Sw'lday arei'\CIIIndudld. 29

PAGE 42

FARE CROSS-ELASTICITY AND ALLOCATION CONCLUSIONS The primary market for MDT A's monthly Metropass consists of those who purchase the pass at a discount, including elderly, disabled, and student purchasers. Only 18 percent of aU pass users pay $60 for the regular monthly pass. The Metropass market appears to be stable, with new pass users (zero to six months) accounting for only 10 percent of respondents. Demographic characteristics of pass users reflect the high proportion of discount passes sold. Pass users are overrepresented in the elderly, white, zero-car, and low income demographic groupings. A more detailed analysis reveals that pass users are not so transit dependent as they appear at first to be, with 23 percent reporting household incomes of over $40,000 and 31 percent owning at least 2 vehicles. Discount pass users exhibit many transit-dependent traits, but the income profile of regular pass users is remarkably similar to that of the average Dade County resident. Corporate pass users are hardly transit dependent: 61 percent are in households owning at least two cars, and 40 percent have incomes of $50,000 and over. Even with the high price of a Metropass, the single biggest reason for purchasing the pass is to save money. This indicates that the pass is being used by those riding transit most frequently. Most pass usage is on weekdays during peak periods (very broadly defined for this study). Metromover is an exception to this trend, with a heavy concentration of midday trips. The average pass holder makes 83 unlinked trips per month, with 50 percent of these trips on Metrobus, 35 percent on Metrorail, and 15 percent on Metromover. Nearly two thirds of pass holders reported using either bus only (28 percent) or rail and mover only (37 percent). The preponderance of discount passes has several implications. There may be a threshold above which the one-time charge for a monthly pass is perceived as too high. Although the break-even point is the same in terms of number of trips for the regular and the discount passes, there may be a greater willingness to pay $30 in one lump sum than to pay $60. More widespread availability of the regular pass through additional sale outlets in popular stores is needed. As noted earlier, many transit agenCies have agreements with major pharmacy/supermarket chains to sell their passes in all stores within the transit service area. Finally, marketing efforts focused on potential corporate and regular pass purchasers could result in increased market penetration. MDT A is preparing a major marketing effort to 30

PAGE 43

CHAPTER 3: PASS USER SURVEY RESULTS inform employers of recent tax code changes increasing the amount of transit pass subsidies that may be excluded from an employer's gross income. MDT A's pricing policy for its monthly Metropass is designed to appeal only to very frequent riders; thus, it is not surprising that pass users ride frequently. The three types of passes offered have resulted in three different markets, with differing demographic profiles and travel patterns. The survey results quantify the similarities and differences among pass users and so provide a reliable basis for pass pricing and marketing decisions. 31

PAGE 44

this page IS blank

PAGE 45

CHAPTER 4: CASH/TOKEN USER SURVEY RESULTS CHAPTER4 Characteristics of Cashffoken Users T he Center for Urban Transportation Research (CUTR) at the University of South Florida designed and conducted a cash and token user survey for Metro-Dade Transit Agency (MDT A) The purpose of the survey was to gather detailed information on demographic characteristics of cash and token users reasons affecting the decision to no t purchase a monthly Metropass, factors that might encourage Metropass usage, and travel patterns of cash and token users. Chapter 4 describes the conduct and results of th e survey. T he survey results quantify the similarities and differences among cash/token users and provide a reliable basis for pass pricing and marketing dec i sions. INTRODUCTION This chapter reports the conduct and results of a survey of cash and token users on the MOTA transit system. In conjunction with a broad review of its fare policy, MOTA wished to gather detailed information on demographic characteristics of cash/token users, reasons affecting the decision not to purchase a monthly Metropass, factors that might encourage cash/token users to purchase a Metropass and travel patterns of cash and token users. Farebox and turnstile data provide information on the nwnber of boardings using cash and tokens for each mode, but MDT A needed more focused info rmation concerning the cash and token user s CURRENT MDTA CASHffOKEN USE Accor ding to June 1993 ridership data tokens account for only five percent of Metrobus boardings and 15 percent of Metrorail boardings. Cash is the predominant payment method on all modes, as shown i n Table 8 Over the past three years cash usage has declined slightly while token usage has experienced a slight increase (see Table 2 in Chapter I). However, tokens have not achieved significant market pen etration. 33

PAGE 46

FARE CROSS-ELASTICITY AND ALLOCATION Tabl e 8 MDT A Ri dership ( June 1 993) '' . f:; :. ::;, ,;.;:.: '' ; ':. I b _, "> 0 '!. l c' 'i _' .... <::-'' :.t tt ./1;: '. :;. .t,, Metrobus 7 4% 5% 21% Metro ra il 57% 15% 28% Metromover 76% NA 24% Table 9 presents cash boacdings by type of face paid for cash mode for the month of June 1993. Metrobus has a disproportionately high share of discount faces (19 percent) compared to other modes . ' '' Metrobus Metrorail Metromover METHODO LOGY Survey Design Tabl e 9 MDTA Cash Hoardings (June 19 93) ;,, ' . : !,, ; F l l { f ' '' u .. ara. ,_. <: 50% 31% 61% 37% 62% 37% .... 19% 2% 1% I n order t o obtain demographic and usage characteri s tics of existing cash and token users, CUTR designed a cash/token survey This survey was developed in co n junction with the Metropass survey, described in Chapter 2 A copy of the MOTA cash/token survey is included in the app e ndix. 34

PAGE 47

CHAPTER 4: CASHiJOKN USER SURVEY RESU LTS Major objectives of the cash/token user survey are to gain an understanding of the reasons for not u s ing a Metrop as s, and the factor s infl u encing a future purchase S e veral stated preference questions were i n c l uded in the survey to gauge th e sensitivity of riders to the cost of the Metropass Standard demographic characteristics and information on frequency of transit use were also obtained. S u rvey A dministr a tion The cash/token survey was conducted during two weeks in June by CUTR personnel. Randoml y selected runs on four Metrobus routes were surveyed, along with the morning peak and midday periods on Metrorail. Patrons were requested to return the comp l eted cash/token user surveys to the surveyors on the bus e s and trains, or by mail to the MOTA address highligh ted on the survey Postage was not prepaid No survey incentives were used on the Metrobus for the cash/token survey. Initial ly on Metrorail, cash/toke n surveys were distributed at selected stations without incentives. The response rate on Metrorail was very poor due in part to the timing of this survey and in part to the abse n ce of survey personnel on the trains to encourage comp l etion of the survey. Thus, Metrorail was resurveyed by on-board personn el on selected trains. For the resurvey, a token was offered as in centive to comp l ete and return the survey. S u rvey Resp onse A total o f 374 surveys were returned and u s ed for analysis. Of these 130 surveys carne from Metrobus patrons and 244 were received from Metrorail patrons. On Metrobus 76 respondents rode during the peak period and 54 respondents during the m i dday per i od Of the Metrora i l passengers surveyed, 94 rail respondents were riding during peak period and ISO rail respondents were rid ing during the midday period The breakdown o f responses by mode and time of day are shown in Figure 10 Metrorail riders are overrepresented among respondents; however rail riders may be more likely candidates to purchase a Metropass than bus riders. T h e ov e rall level of precision for a sample of this size (374) is ;!:5. 1% at t h e 95 percent confidence level. 35

PAGE 48

FARE CROSS-ELASTICITY AND AI.WCATION Figure 10 Survey Respondents by M ode and Time of Day Off-1M1ak Bus Poalc Bus Peak Rail Olfpoalc Rail SURVEY RESULTS The instrument used to conduct the survey was structured so that the information on cash and token users could be collected in three primary areas: trip-making behavior, demographic profiles, and awareness of and willingness to use Metropass. The resulting survey data was analyzed using the SAS System for Persona l Computers, Release 6 04 Various SAS procedures were utilized to calculate survey variable frequencies as well as crosstabulations of multiple variables to gain a better understanding of interrelationships among usage, demographic, and travel characteristics of cash and token users. Se l ected results from these analyses are presented and discussed in the following three sections 36

PAGE 49

CHAPTER 4: CASH/TOKEN USER SURVEY RESULTS Trip-Makin g Characteristics of Casbffoken Users The first question of the survey asked the riders by what method they paid their fare when they boarded the vehicle. As indicated in Figure 11, 49 percent paid full-fare cash, whereas 26 percent used tokens and II percent boarded with a transfer. Discounted cash payments (student disabled, and senior citizens) accounted for 13 percent of ridership surveyed. Figure 11 Fare Payment Method Full Fare Cash Token Transfer Student CUh Senior Cash Disabled Ca.h The second and third questions of the Cash/Token Survey dealt with how often riders utilize Metro -Dade Transit When asked how often they ride Metro-Dade Tran sit, 52 percent of the respondents indicated they ride five or more days a week and 20 percent said they ride 3 or 4 days per week (Figure 12). 37

PAGE 50

FAR CROSS ELASTICITY AND ALLOCATION s or More Days per Week 3 or 4 Days per Week 1 or 2 Days per Week Less than Once per Week Figure 12 Frequency of Transit Use NoResponse % Asked specifically how many trips were made the previous day, 32 percent said they utilized transit for two trips whereas nearly 16 percent did not take a transit trip the previous day (Figure 13). As to how many transit trips were made the previous weekend seven percent made two trips and eight percent made four trips (Figure 14). Twenty-four percent made no transit trips the previous weekend, whereas more than 50 percent did not respond. 38

PAGE 51

CHAPTER 4: CASH/TOKEN USER SURVEY RESULTS Figure 13 Number of Transit Trips the Previous Day 210 0.3" ... 10 ,,,, Figure 14 Number of Transit Trips the Previous Weekend 0 1 2 3 5 10 12 20 39

PAGE 52

FARE CROSS-ELASTICITY AND ALLOCATION Demographic Characteristics of Cashffoken Users Demographic data on age, gender, etbnicity, number of household vehicles, and annual household income of the respondents were collected from the fmal group of questions on the survey. These results are shown in Figure 15, and are compared to similar demographics gathered from Dade County's 1990 Census data. Also, a comparison of cash/token user demographics and Metropass user demographics are shown in Figure 16 Sixty percent of the survey respondents are between the ages of20 and 39, compared to only 32 percent of the Dade County population and 22 percent of Metropass users. Worth noting is that only 3 percent of cash and token users are 65 and over, whereas 35 percent of Metro pass users fall in this age category ( 14 percent of Dade's population are 65 and over) Senior citizens can buy the monthly Metropass at the discounted price of $30, a more affordable one-time payment. On the subject of ethnic origin it seems that African-Americans are overrepresented among cash and token users Whereas Dade County is 49 percent Hispanic, 30 percent white, and 19 percent African-American, the cash/token survey indicates only 36 percent of the respondents are Hispanic, followed by 29 percent African-American and 24 percent white. Metropass survey results indicated that 46 percent of Metropass users are white, with 34 percent Hispanic and 13 percent African-American. The gender ratio of participants in the cash/token survey is similar to that of Dade County. The respondents are 46 percent male and 54 percent female, very close to the Dade County ratio of 48 percent male and 52 percent female. The Metropass male/female ratio was also 46 percent male and 54 percent female. Cash/token respondents have a lower level of vehicle ownership than Dade County as a whole. Thirty percent of the riders surveyed own no vehicle whereas among Dade County households only 16 percent do not own a vehicle. Interestingly, the Metropass survey results showed a higher percentage ( 42 percent) of respondents with no vehicle in their household. Overall household vehicle ownership levels are lower for transit users than for the general population. 40

PAGE 53

CHAPTER 4: CASH/TOKEN USER SURVE Y RESULTS The final question of the demographics section dealt with total annual household income. Thirty percent of survey respondents reported household incomes less than $10,000, much higher than Dade County's 19 percent, but similar to Metropass survey results (32 percent). Also notable is that 28 percent of the cash and token users surveyed fall within the $10,000 to $19,999 category Awareness of and Willingness to Use Metropass Three questions on the survey asked about Metropass awareness and previous usage. Whereas 70 percent of those surveyed have heard of the Metropass, only 21 percent of cash or token users have bought a Metropass previously (Figures 17 and 18). When asked why they are not currently using a Metropass, most either cited the high cost of the Metropass or infrequent transit use, as shown in Figure 19. These responses indicate that cost is a major factor in choosing not to purchase a Metropass. 41

PAGE 54

FARE CROSSELASTICITY A ND ALLOCATION Figure 15 Comparative Demographics: Casb Token Users v. Dade County Cash/ Token Users Age Dade County Olher 11% Tv.o 23% 1 5&under 18 to 19 201029 301039 401059 601064 8S&
PAGE 55

CHAPTER 4: CASHII'OKEN USER SURVEY RESULTS Figure 16 Comparative Demographics: Cash Token Users v. Metropass Users Cash/ Token Users """" 11% T.., 23% WhO 24% Mate 46% Three01More 12 % AQe 15&undet 18 to 19 201029 301039 40to 59 601084 85 & """ Ethnic Origin Gender Household Vehides 43 Metropass Users Hispanic 34% Two 24% NOf\O 42% 40% Whl!e 46% M alo -Th,..orMore 7%

PAGE 56

FAR CROSS-ELASTICITY AND ALLOCATION Figure 17 Awareness of Melropass v .. No Figure 18 Previous Purc hase of Metropass No 44 ... No RUCI,.,_ No-

PAGE 57

CHAPTER 4: CASH/TOKEN USER SURVEY RESUL TS Figure 19 Reasons for Not Using Metropass One Time Cost Too High Do Not Usa Transit Often Enou!Jh No Convenient Ptaoe to Buy Mettopass Prefer Using Tokens In the final battery of questions on user characteristics, riders were presented with three different scenarios on pricing of Metropass as well as other fares. These types of questions are known as "stated preference" questions because they ask respondents to make hypothetical choices as opposed to reporting actual behavior. Before and after studies have shown that respondents are approximately four times more likely to say that they will change their behavior than to actually change. Thus, results of stated preference questions must be interpreted with caution. In the first question, riders were asked if they would purchase Metropass if it were priced $10 per month less (i.e., $50 per month). Thirty-five percent indicated that they would definitely or probably purchase a Metropass at this price whereas 33 percent said they would definitely or probably not buy a pass. The second question asked if riders would be intere sted in purchasing a Metropass at $20 below the current monthly price ($40 per month). Survey responses indicated that 54 percent would definitely or probably purchase a Metropass whereas only 18 percent said they would probab l y or definitely not purchase a Metropass. 45

PAGE 58

FARE CROSS-ELASTICITY AND ALLOCATION The third question asked riders if they would purchase a Metropass if its price remained the same whereas cash and token prices were increased to $1.50 and $12 per ten tokens, respectively. Under this scenario, only 16 percent probably or definitely would purchase a Metropass whereas 52 percent said they would probably or definitely not purchase a Metropass. Figures 20 through 22 show the stated preference responses. The response to the fare increase scenario is in teresting Many transit systems have used a similar scenario to offer riders the chance to avoid paying an increased fare by choosing a multi-trip or monthly pass option. A literal interpretation of the response to this question suggests that riders are willing to pay higher cash and token fares rather than switch to the Metropass. One possible explanation is that respondents react negatively to any proposal including a faie increase. Another possibility is that a one-time $60 payment is perceived as too high regardless of the alternatives. Although the relative cost comparison between the monthly pass and single fares is favorable for the pass under this scenario, affordability becomes a major issue for low income riders, even those who travel frequently. Yes, Definitely Yes. Probably Maybe Probably No1 Oeflnftely Not Figure 20 Willinguess to Purchase a Metropass at SSO per Month 46

PAGE 59

Y ... Definitely Yos. Problbly Mayt>e Probably Not Definitely Not No Response Yes, Definitely Yes, Problbly Problbly Not Definitely Not CHAPTER 4: CASH/TOKEN USER SURVEY RESULTS Figure 21 Willingness to Pur
PAGE 60

FARE CROSS-ELASTICITY AND ALLOCATION The final question i n the user characteristics section asked riders what would encourage them to purchase a Metropass Results are presented in Figure 23. F ifty three percent responde d that a less expensive weekly pass option would suit their needs. Twenty percent said more convenient sales outlets would encourage a Metropass purchase, whereas 12 percent indicated that mail order purchases of Metropass would foster usage. The results of this question suggest that the cost of Metropass and the inconvenience of sales locations are major detriments to its use. Figure 23 Factors that Would Encourage Metropass Purchase 20% 12% Payroll Deduction Less Expensive Weekly 53% 40% 60% CONCLUSIONS In terms of demographics, cash and token users are typically of working age, with a very low percentage age 65 and over. The gender ratio is similar to Dade County's. Cash/token respondents are more likely to be African-American compared to Metropass users and to Dade County residents. The proportion of low income and zero-vehicle households was higher for 48

PAGE 61

CHAPTER 4 : CAS!I!TOK.EN USER SURVEY R.ESULTS cash/token users than for the county, but the differences in income and vehicle ownership levels between cash/token and pass users are not great. Despite the availability of a monthly Metropass and discounted tokens, cash is still the most common method of fare payment at MDT A The cash/to ken survey has revealed that the $60 per month cost of the Metropass and, related to this insuffic i ent use of transit to justify a pass purchase are major reasons why current cash and token users do not use Metropass. The s tated preference results indicated that one-third of cash/token users would switch to Metropass at a $50 cost, and one-half would switch at a $40 cost. These results shou ld be viewed w i th caution. however, since the number of people stati n g that they would change their behavior is far greater (by a factor of four, the literature suggests") than the numbe r who will actually change Cash and token users showed considerable interest in a weekly pass. Fifty-three percen t of all respo nde nts reported that a week ly pass would encourage them to purchase a pass A weekly pass could be priced within the rider's perceived threshold of v al u e and could encourage even more transit ridership since this new group of riders would have unlimited ride privileges for the duration of the pass The negative aspect of a week ly pass is the sharp in crease in ad m inis tra tive costs. A major pro ble m with pass or token sales are the sales locations themselves Many r ide r s listed the i n convenience of sales outlets as the reason they did not use Metropass, and one in five responde n ts reported that more convenient sales outlets would encourage them to purchase a Metropass. Major supermarket and convenience store chains need to be brought in t o t he sales campaign to facili t ate access to the riding public. 49

PAGE 62

this page IS blank

PAGE 63

CHAPTER 5: PREDICTING FAll PAYMENT CHANGES CHAPTERS Predicting Changes in Fare Payment Method Chapter 5 utilizes information from the cash/token user survey in developing a logit model to predict the method of fare payment. This chapter describes the model-building process and the predictions resulting from two binomial logit models: pass-versus-cash and pass-versus token. The outcomes of both models are used to predict changes in the shares of Metro-Dade Transit patrons paying with the Metropass, cash, and tokens. INTRODUCTION Chapter 5 describes the steps in building a log it model and reports the results to be used for predicting changes in fare payment method. Metro-Dade's logit model has been designed following the process used in the recently conducted model formulation for the Chicago Transit Authority." The data were obtained from the results of the cash/token user survey described in Chapter 4. A total of 374 surveys were returned. The nature of the data required that two separate binomiallogit models be calibrated: one for pass-versus-cash, and one for pass-versus-token. Most important in the model were responses to three survey questions con cern ing whether patrons would consider purchasing the M etropass at different price levels. Respondents were asked to answer whether they definitely would buy the pass, probably would buy it, maybe buy it, probably not buy the pass, or definitely not at each of the three price levels. The price levels were $40 (with current cash and token fares) $50 (again with other current fares), and $60 (the present price of the pass) with an increase i n cash fares to $1.50 and an increase in token fares to 10 for $12. TlJE DATA The data used in this model-building process were gathered from the results of th e cash/token user survey. Information such as fare type, payment method, frequency of use of the transit system, and income level was considered important for inclusion in the model. Most 5 1

PAGE 64

FARE CROS S-ELASTICITY AND AUOCATION important, however, were the questions posing three pricing scenarios. Each respondent was asked to react to three different price levels of the Metropass. Responses to these three questions can help to estimate the sensi ti vity of cash and token users' demand for the Metropass to changes in the price of the pass. A few reasonable assumptions were made about the data, for simplicity. First, if toke n users were to switch to the Metropass, they would only be eligible to purchase the full fare pass. Second, those who responded that they were making a transfer when they were surveyed were assumed to be in the full fare category. It was postulated, as in the Chicago study, that the utility of a particular method of payment or the degree of satisfaction experienced from choosing it, is based on the cost of that method. More specifically whether a person decides to switch from paying for transit with cash or tokens to the Metropass depends on the difference between the dollar cost he or she pays now and the dollar cost of the Metropass. Since the cost of the Metropass is worth one month of transit rides, the monthly cost of respondents who pay with cash or tokens must be calculated. First, the answer to the question, "How many trips using Metro-Dade Transit did you make yesterday?" was multiplied by five, which assumes (si nce the survey was carried out over two weekdays) that the respondent makes a consistent number of trips Monday through Friday. Then, the number of trips the respondent reported for Saturday and Sunday of the previous week was added This total number of trips is assumed to represent an average number of trips per week for each respondent and is used as such. Trips per week were multiplied by 4.333 to determine the total number of monthly trips. Using this number of trips as well as fare payment information monthly cost was calculated as follows: where: COST,= TRIPS, (CASHFARE, + .231 TRANFA.RE) TRIPS = the number of trips per month for the ith respondent; CASHFARE = that same respondent's dollar fare amount for one trip (either full fare or discount cash or token) ; TRANFARE = the cost of a transfer ($0 25 for full fare and $0.10 discounted). 52

PAGE 65

CHAPTER 5: PREDICTING FARE PAYME/YT CHANGES Tbe average share of trips requiring a transfer was determined from MDT A published data to be approximately 23 percent. The monthly cost was thus figured to take account of the average number of transfers a rider needs to make in a month. In order to maximize the amount of information accumulated from the stated preference questions, a technique known as exploding" the data set was employed. This procedure enables extra information to be exploited from the data to estimate the parameters of the logit model in a more efficient manner." The explosion process involves the breakdown of a ranked choice set into a succession of statistically independent choice sets. In this data set, each single survey re s pondent was treated as three identical respondents each answering to only one of the three questions. Thus, instead of having one respondent who answered according to three scenarios, each "new" respondent replied to one scenario. This process tripled the size of the original data set. THE MODEL In this study the binomial logit model is based on the utility functions of each method of payment (MOP): where: exp(uli/ityMoP/) ProbabilityMoPs = exp(utility.,10r1 ) .,. exp(utilityMon> Probability_,fOP = the probability of choosing a particular method of payment, or the share of those using that method of payment; and exp(uriliry,>r) = the exponent (antilog) of the utility function for the particular method of payment. According to the above equation, the probability of choosing a particular method of payment is equal to the exponent of the utility function for that method, divided by the sum of the exponents of the utility functions for each method of payment considered. Thus, alternative models can be calibrated by substituting different utility functions into the formula 53

PAGE 66

FARE CROSSELASTICrrY AND ALLOCATION By estimating utility functions, we can compute the probability of choosing a specific method of payment. There are two binomial Jogit models : one model for pass-versus-cash, and one for pass-versus-token. Results from these two models will be used to estimate changes in the shares among all three methods of payment. For this study since on ly cash and token users were surveyed, there was no control group of pass users. However, the estimated choice probabilities based on the responses to the stated preference questions were available. The estimates were produced by converting the responses to the future choice questions into choice probabilities. These probabilities were based on professional judgment, as in the Chicago srudy: Table 10 Estimated Cboiee Probabilities of Buying a Metropass Response. Yes. definitely .90 Yes, probably .70 Maybe .50 Probably not 30 Definitely not 10 Therefore, a person responding that he or she would definitely buy the Metropass g iven a certain price would have a 90 percent chance of acrually making the purchase. With these estimated choice probabilities, a linear transformation of the logit model seemed most appropriate. Since most interest lies w ith the probability of choosing the Metropass the probability and utility of the pass has been substituted for the more general "method of payment" (MOP): where: 54

PAGE 67

CHAPTER 5: PREDICTING FARE PAYMENT CHANGES = the ith respondent's probability of purchasing the Metropass; and utility = the utility respondent i associates with the method of payment. The difference between the utility functions takes the following form: where: p = paramete r s; x1 = various independent variables impacting utility recorded for each respondent; and e = random error term. In this model, the difference between the utility functions is equal to the natural log of the probability ratio The formula is now linear in the parameters and least squares regression can be appl i ed. This model is simply an algebraic manipulation of the standard logit mod e l ; there is no differ e nce between the t wo. There are, however, a few limitations to this form o f the logit model. To apply this form, estimates of t he probabilities must be known a priori. Usually, these probabilities are unknown, but there are way s they can be estimated. I n this case, the probabilities had already been estimated and, thus, were known and could be entered into the model. Also, it is e xpected that the calibration will result in unequal variances among the disturbance terrn s (e,). Homoscedastic variances can be restored, howev e r with an appropriate variance-stabilizing transformation of the data These transformations generally give less weight to the Iaeger variances and more weight to the smaller variances and, if successful allow the assumption of equal variances in least squares regress i on to be fulfilled. In order to apply the linearly transformed logit model the ut i lity functions must be developed Variables that are thought to impact utility were examin e d. T he monthly cost of the fare payment method was hypothesized to be the mo s t important variable in determining t h e method's utility. In addition, since use of the Metropass is the focus it s eems reasonable to incorporate the one-time monthly cost of the pass (full fare or discount at each price level in 55

PAGE 68

FARE CROSS-ELASTICITY AND ALLOCATION the survey) in the function The annual house hold income level of the respondents was also considered but was found to have no statistically significant impact on the utility. Income was subsequently removed from the model. The final form of the model is shown below: utility,,_-= (30 + (J1COSTDIFF1 + (J,PASSFAR1 + E 1 where: COSTDIFF1 = PASSFARE1 -MONTHLY COST, (monthly cost differenee for the ith respondent); PASSF ARE, = priee the ith respondent would have to pay if he or she bought the Metropass; 13o = constant term; 13, = parameter on COSTDIFF ; 13, = parameter on PASSFARE; and 61 = random error term. Clearly, the parameter estimates of the utility function for the Metropass will be different in each of the two models: pass-versus-cash and pass-versus-token. For the final results, the probabilities of choosing Metropass from both models (which also represent the share of those who will switch from cash or token to the pass) will be combined to determine the overall estimated change in pass use at each price alternative. FINAL MODELS The following model was calibrated using the least squares approach: LN(P1,.., I 1 P1 ,..,) = {30 + (J1COSTDIFF1 + (J,PASSFARE1 + E 1 Binomial pass versus -cas h and pass versus-token models were developed. As expected, the prob l em of unequal variances was present in the results of both models Several different 56

PAGE 69

CHAPTER 5 : PREDICTrNG FARE PAYMENT CHANGES weighting techniques failed to alleviate the heteroscedasticity completely. One particular transformation, which involved weighting both the independent and dependent variables by the predicted values of the probabilities (more specifically, the square root of [ P (I P)J, where P equals the predicted value of the probability of choosing Metropass) did at least reduce the problem. The final models incorporate this data transformation. The presence of heteroscedasticity in the final models does not bias the parameter estimates; that is, the expected value of the unbiased estimated parameters is equal to the true value of the parameter. However, these estimates will not have minimum variance Therefore the results should be interpreted with caution Pass-versus-Cash Model The final model for pass-versus-cash is as follows (!-statistics are in parentheses): LN(P,,_ /1 P,,=) = .993084 -.015431COSTDIFF1 -.03l635PASSFARE, (6.044) (-6.983) (-7.511) All statistics on the parameters are significant at the 95 percent confidence level. Also the F-statistic for overall fit is highly significant at 55.575. In addition, the signs of the estimates are what would be expected. It can be shown (proof omitted) that the sign of the quantity on the left-hand side of the above equation, LN(P, ,..,,II P, ,.,,), is the same as the sign of P,pas.,Thus, we can say that, as the difference between the ith respondent's current monthly cash cost and the cost he or she would incur if the pass were purchased decreases, the probability of switching to the Metropass increases. Likewise, if the cost of the pass itself decreases, the probability of purchasing it increases The R-square is low, at a value of .1312. However for this model, the R-square is not a reliable measure of the goodness of fit of the model. Generally, the R-square should not be used as a summary statistic in models with limited dependent variables." Because the value of the depend ent variable in such models can only take a limited number of values, it becomes very difficult to fit a good line to the data. Also, it follows from this model that the difference in utility between the Metropass and c ash is equal to the right-hand side of the above equation: 57

PAGE 70

FARE CROSS-ELAST/CfTY AND ALLOCATION utility,"""utiliJy1_. = .993084-.01S431COSTDIFF1 .03163SPASSFARE1 This function is valid only between pass and cash users. Pass-versus -Token Model The final model for pass-versus-token is shown below (!-statistics are in parentheses): LN(P,,..,/1-P,,.,,) = 1.012194-.017448COSTDIFF1 -.061101PASSFARE1 (6 .257) (-4.535) (-6.011) Once again, all !-statistics are significant, and the parameter estimates hav e the correct signs. Again there is a highly significant F-statistic (41.324), which tests for the overall usefulness of the model for predictive purposes. For this model, the R-square is larger than for the pass versus-cash model. This R-square has a value of .2285. However, as mentioned above R-square is not a reliable statistic for this model (linearly transformed logit model). The difference in utility function between the Metropass and tokens in this model is equal to the right-hand side of the above equation : utility,,.,.utility"" = l.Ol:Z194-.017448COSTDIFF1 -.061101PASSFARE1 This function is only valid between pass and token users. RESULTS Validation If current price levels are substituted into the models, the resulting shares between pass, cash, and token use should closely resemble the actual distribution of shares. If they do, one may have more confidence in the predictive power of the models. Using ridership data from June 1993, the month in which the survey was completed, the actual distribution between pass, cash, and token users was determined. Only data for 58

PAGE 71

CHAPTER 5: PREDIC11NG FARE PAYMENT CHANGES Metrobus and Metrorail were used. Table 11 shows the actual allocation between pass, cash, and tokens in June of 1993. Table 11 Distribution among Metropass, Cash and Token Users (June 1993) >' Metropass (Full Fare and Discount) 1,317,057 20.3% Cash (Full Fare and Discount) 4,725,168 72.7% Token 455,692 7.0% The shares in Table I I can be compared to the shares based on our survey sample population when current prices are entered int o the two binomial models The total numbe r of observations that were valid for calibrating the models was I ,021. Cash users consisted of 739 of the total observations, and token users accounted for 282 entries. The mean number of trips per month for each of the two groups was used to calculate the cost difference (mean number of monthly trips for cash users, 46.08022; for token users, 46.08834). The current fullfare pass, cash, and token prices were substituted into the models The r es ults are shown in Table 12. Table 12 Distribution among Metropass, Cash, and Token Users Compiled from the Pass-v.-Cash and Pass-v.-Token Models Fare Payment Method Number of Boardings by Payment Method Metropass (Full Fare and Discount) 229 Cash (Full Fare and Discount) 525 Token 267 59 Share 26.4% 51.4% 26.2%

PAGE 72

FARE CROSS-ELASTICITY AND ALLOCATION The Metropass share is overstated by our model (by approximately 6 percent). This is somewhat expected, however, since the model is based on respondents' answers to questions on whether they would buy the pass at various prices. People are always more inclined to respond that they would hypothetically buy something when, in fact, they may not be as likely to actually buy it This tendency is clearly evident in these models. The largest discrepancy between the actual shares and those predicted by the models is between the cash and token shares This inconsistency can be accounted for, however. Table 13 exhibits the allocation of shares in the cash and token survey sample population, as well as the shares between the actual cashand token-user population (based on data from June 1993). The survey population's distribution does not match the actual distribution: Token use is overstated. Thus, it would be expected that token use would be overestimated in the sample population. Table 13 Distribution among Cash and Token Users Fare Payment Method Survey Sample Share . Cash 72% 91% Token 28% 9% Predictions Probabilities for using the Metropass were calculated from the two binomial logit models. Using the ridership data from June 1993, the number of riders who would switch from either cash or token use to the pass was determined. Accordingly, a new distribution between the three payment methods was established, based on the total ridership on the Metrobus and Metrorail in June of 1993. Tables 14, 15 and 16 summarize these results. 60

PAGE 73

I CHAPTeR 5: PReDICTING FARe PAYMeNT CHANGCS Table 14 Distribution among Metropass, Cash, and Token Users Price of the Metropass = S40 Compiled from the Pass-v.-Cash and Pass-v.Token Models .. .,.it' New Share > Existing Share > Fare :'. (Currtntl'rlc:es). :' Pass Rrlce = $40 > Metropass (Full Fare) 20.3% 58.9% Cash (Full Fare) 72.7% 35.6% Token 7.0% 5.5% Table 15 Distribution among Metropass, Cash, and Token Users Price of the Metropass = SSO Compiled from the Pass -v.Cash and Pass-v.Token Models Fare Payment Method Existing Share New, Shant (Currant Prices) Pass Price = $50 Metropass (Full Fare) 20.3% 49.7% Cash (Full Fare) 72.7% 44.1% Token 7.0% 6.2% Table 16 Distribution Among Metropass, Cash, and Token Users Price of the Metropass = $60.00 Full Cash Fare = Sl.SO Token Fare = $1.20 Compiled from the Pass-v.-Cash and Pass -v Token Models Fare Payment Method Existing Share New Share (Currant Prices) Pass Price $60 Metropass (F ull Fare) 20.3% 44.4% Cash (Full Fare) 72.7% 49.0% Token 7.0% 6.6% 61 c ".p)BDIJ!' 38.6% -37 .1% -1.5% Change 29.4% -28.7% -0.8% Change 24.2% -23.8% -0.4%

PAGE 74

FARE CROSS-ELASTICITY AND AUOCATION Tables 14, I 5, and 16 demonstrate possible substantial increases in the use of the Metropass (mostly from cash users switching to the pass) i f its price is lowered by $10 or $20, or if cash and token fares are increased. To obtain a more realistic estimate of the number of cash and token users who would switch to the Metropass, the fact that true preferences are not always revealed through surveys must be considered. Previous studies have shown that only about one quarter of those who state they will conduct some furure action (such as purchasing the pass) actually do so. The results from T ables 14, 15, and 16 were weighted by a factor of four to adjust for the overestimation present in the stated preference responses. The normalized results are presented in Tables 17, 18, and 19. Table 17 Distribution among Metropass, Cash, and Token Users Price of the Metropass = $40 Compiled from the Pass-v.-Cash and Pass v. -Token Models Adjusted for Stated Preference Responses Metropass (Full Fare) Cash (Full Fare) Token Existing Share (CurrenfPrlces) 20 3% 72.7% 7.0% Table 18 .. ; . o\'d)!Jated Share . Pii,&S Price .. $40 29.9% 63.5% 6 .6% Distribution among Metropass, Cash, and Token Users Price of the Metropass = SSO Compiled from the Pass-v.-Casb and Pass-v.Token Models Adjusted for Stated Preference Responses Fare Payment Method Existing Share Adjusted Share (Cu"ent Prices) Pass Price = $50 < Metropass (Full Fare) 20.3% 27.6% Cash (Full Fare) 72.7% 65.6% Token 7.0% 6 .8% 62 .. . Change . ; . 9.6% -9.2% -0.4% Change 7.3% -7 .1 % 0 .2%

PAGE 75

CHAPTER 5: PREDICTING FARE PAYMENT CHANGES Table 19 Distribution among Metropass, Cash, and Token Users Price of the Metropass = $60.00 Full Cash Fare = $1.50 Token Fare = $1.20 Compiled from the Pass-v.-Cash and Pass-v.-Token Models Adjusted for Stated Preference Responses 1:' 1 r ; aiitiiJ s,, fl i'dju'is&d>Sh;,. ; ; .. ..... . y ... ; [ < Pass Price : c$60i. . . . Metropass (Full Fare) 20.3% 26 3% 6 .0% Cash (Full Fare) 72,7% 66,8% -5.9% Token 7.0% 6.9% -C.1% The adjusted results give more realistic estimates of i ncreases in pass use among current cash and token users. By adjusting the shares. the predicted changes in pass use take into account the limits of the stated preference responses. It could be assumed from these results that transit riders are sensitive to changes in the price of transit More specifically, cash and token users on the Metro-Dade Transit system are sen si tive to changes in the price of the Metropass. Recall however, that these results must be interpreted with caution. LIMITATIONS OF THE MODELS The power of the two models is restricted partially because of the nature of the data. It has been shown that, in a survey, respondents are much more likely to respond positively concerning a future action. In other words, the survey respondents in the cash/token user s urv ey were more likely to indicate that they would purchase the Metropass at each price level when, in reality they would not it People reac ted initially to the price difference, perhaps without considering other "costs" of using the pass, such as finding a convenient place to buy the pass. (In the survey, this was cited as one of the most commo n reasons for not purchasing the pass ) It is extremely difficult to get people to reveal their true preferences through survey questions and this has an effect on the re su lts of the study. 63

PAGE 76

FARE CROSS ELASTICITY AND ALLOCATION Additionally, the problem with unequal variances must be remembered This obstacle was addressed in the mode l -building process yet could not be completely overcome. Since heteroscedasticity does not bias the parameter estimates, the predictive power of the model should not be much altered. CONCLUSIONS The objective of this study was to formulate l ogit mode l s to predict changes in fare payment method between pass cash, and token users The data used in the modeling process were from the cash/token user survey, which accumulated information on demographic characteristics of cash and token users, as well as information on why these individuals are not currently using the Metropass and what factors might influence them to purchase it. The respondents were asked whether they would buy the pass under three different price situations: If the price of the pass was $40, $50, or $60 with increases in the cash fare to $1.50 and token prices to $1.20. Responses to these questions were particularly important for estimating choice probabilities, which were then used in calibrating linear transformations of binomial logit models : pass-versus cash and pass versus-token. Variables that were important for inclusion in the models were the difference between the one time monthly cost of the Metropass and the current monthly cost (utili z ing cash or tokens) of the respondents. The price of the pass was itself a significant variable as well. Results of the models, although most lik e ly overstating inc r eases in pass use, showed that cash and token users are sensitive to changes in the price of the Metropass. The results of these models may be used as a tool in estimating fare cross-e l asticities between fare payment me t hods. Although the pass-versus-cash and pass-versus token models in this study have their limitations, they nonetheless may be used as a starting point for subsequent calibrations 64

PAGE 77

APPENDIX: INSTRUMENTS

PAGE 78

this page IS blank

PAGE 79

MDTA Pass User Survey 67

PAGE 80

this page IS blank

PAGE 81

HETRO-DADE TRANSIT PASS USER SURVEY Dear Metropass User: Tell us how you use Metropass and help us improve your transit service. The first f)drt of the survey asks for a diary of your transit trips taken over a one week period The sey (see reverse side) asks for information about yourself and your use of Metro{Jdss. 100 pass users will receive a free M etrop.ass for participating In the suNey See reverse side for derails. THANK YOU/ TRIP DIARY INSTRUCTIONS: Please comp l ete this diary for the previous seven days. If you d o not have a May Metropass, p lease fill out the diary for the seven days in a row starting the f i rst day of June. Put a B (MetroBus). R (MetroRail) or M (Metro Mover) for each trip made using Metropass. Include all trips made using the pass. A TRIP occurs every t ime you board an MOTA vehicle. TIME STARTED refers to the time you boarded the Metro Bus, Metro Rail, or Metro Mover. & ampk. lime Started Monday 4:00a.m. 10:45 a m 1 0:46 a m 1 :45 p.m. 1 : 46 p m 6:45 p .m. 6:46 p .m. 3:00a.m. B .. MetroBus, R ... MetroRaiJ, M .. (Went to w01k by taking the Bus, transferred to Rail, and then to the Mover. ) (Went to lunch by using the Mover, and returned on the Mover .) (Reverse of morning trip to wo r k.) (Went to mall a n d bocl< by Bus.) t: :s Q 9i !S tl t: ... i ... .. 1! 10 "' ... .. ;: ... ... .:1 ... .:1 ... .. .. {! ... .. c 0 ::il ... .. c "' .. t: :! "' .. E ;: E E .. ci. ci. "' "" "' ':': ':': 0 . E E E "' ci. "' <0 <0 0 ':': 0 0 -" E .. 0 0 "' E ci. <0 .. .;; 1 I ::il j I "" i I .,

PAGE 82

1. Which Metropass do you use? _Regular Discount (Youth. Elderly Disabled, College Student) Corporate 2. How long have you been using Metropass? ------3 How many times have you bought a Metropass in the last six (6) months? __ 4. Did you purchase the $2.00 Metrorail Parking Permit this month? _Yes _No 5. Why do you use Metropass? (Check all that apply) _To save money No need for exact change _Avoid transfer charges _Other ____ 6. 7. Take extra rides I might not otherwise make Your age is: 15 or un de r 16to 19 _20 to 29 You are: Male _30 to 39 _40 to 49 50 to 59 Female _60to64 _65to74 75 or over 8. You r ethnic origin is: _White _African-American _Hispanic 9. How many vehicles are owned or leased by your household? None One Two _Three or more 10. Your total annual household income is: _Less than $10,000 $30,000 $39 ,999 $1 o.ocio$14,999 = $40,ooo-$49,999 -$15,000-$ 1 9,999 $50.000-$59.000 $20,000 $29,999 $60,000 and over Other 11. What is the z i p code of where you live? -----------12 Any comments: ----------------------To enter the free Metropass drawing, please write the pass number (from the middle of your pass) in the space provided below, from your just-purchased Metropass. A drawing will be held from all surveys received. and 100 passes will be given away. Nexr month look for a listing of the winners at your nearest pass outlet or other place/business of purchase. Show your M etropass with the winning numbers and you will rec eive a free pass on the spot. Thank you again for participat ing in this survey! June 1993 Metropass Number _________________ t>LEASE HELP USI Drop your surrey In the specYI/y boes ulect rMI stations, It to your lfdrobos drlrer, H retunt by N: Pass UHf' Surny Metro Dade Transit Agency Ill N. W. I st Street. Ste. 910 Mla111l, FL JJI18-1999

PAGE 83

MDT A Cashffoken Survey 71

PAGE 84

this page IS blank

PAGE 85

METR ODADE TRANS I T CAS H/T OK E N USER SURVE Y Dear R i der : Met r oD ad e T r a n s i t wou ld l ike informat i on a bout your use of the trans" system to h elp us improve your fare o ptio ns P l eas e he l p u s by complet i ng lhe survey and r e turn in g t o the su rv eyor on t h e bus th e s urvey return box at setect raj st a tions or t o Metro-Dade Trans at the lollowing add re ss : Cash/Token U ser S u rvey Me tro-Dade T ra ns i t Agency 111 N W 1 s t Street. Suit e 910 M iam i FL. 331 28 -1999 NOTE: I F Y OU ARE CUR R E N TLY USI N G A M O N THLY M ETRO P ASS PL E AS E DO NOT CO M PLETE THI S S U R V EY. 1 Wha t f are paymen t m e thod did you use whe n y o u boarded t his vehic le? F u ll Fare Cash D i sabled C ash To k e n S t uden t Cas h Senior Cash Transfe r 2 H ow ofte n do you rid e th e Me tro-Dade T ra n s i t system? 5 o r more day s per week 1 or 2 days per week 3 or 4 days per week L ess than once per w ee k 3 How ma n y trips usi n g Me t roD ade Tran s did you mak e : Yesterda y? Las t we e kend ( Sat and Sun)? (P l ease co unt e ach tim e you boarde d th e bus rai l o r move r as o n e trip) 4 Ha v e you heard of t he Metro p ass? Ye s No (If N o then go to exp l anati on abo v e question 7) 5 Hav e you e v er pu r c hased a M etropas s ? Y e s (if yes, n u mbe r of t i mes i n l a s t 6 m o nth s) ___ N o 6 Why are you n ot cu rrentl y us in g M et r opass ? (chec k a ll that a p pl y) One-time co s t per month i s too high Do not use th e tr a nsit sy s t e m often enough N o oonvenien t p lace to buy M etropass P r e f e r using tok en s Othe r ___ ________ ______________ ________ E N ESPANOL DEL OTRO LADO. T h e M etropaaa allo ws you to ride th e M etrobus, M etr o r a il a n d M et r omover a ll m o nth long, as much as you want. T h e r e is n o need t o p a y for a transfer. C u rre n U y a Metr opass costs $60 .00, w ith dis cou nted passes for el d erty disab led a nd s t udents lor $ 30 0 0 and a co l l ege s tudent pass for $ 4 5 00 7 Would you puncha se a mo nth ly M e tro pas s i f cost $ 50 .00 per month ( $25 00 for elderty disabled and s tude n t s $37 50 for college stude n t s ) ? Yes Defin"ely May be P robably not Yes Prob a b ly Defi n itely no t 8 Would y ou purchase a mont hly M etropass if It cost $ 4 0 00 per month ( $ 2 0 .00 for eld erly disab led and s t u den ts, $30 00 f o r co ll ege s tudents)? Yes, Definite l y _Mayb e P ro bably n o t Yes, P robably _Definite l y not 9 Would you purch a se a m ont h l y Met ropass If I t remained at $60 00 pe r month ( $3 0 00 for e l derl y disabled and s t ude nts, $45 .00 for co ll ege s tude nt s ) but regular cas h fa re s w ent u p from $ 1 .25 to $ 1 .50, and tok ens w en t up from 1 0 f or $ 10 to 10 for $ 1 2 7 Yes, Definitely Ma ybe _Prob ably not _Yes, Probably _Defi n i t ely not 10. Wha t oth er factors would encourage you to purchase a M etropass ? ( c heck a n that a pply) Convenie n t sales outle ts L ess ex pensive w eekly pas s M a ll orders _Oth er-------Payroll Deduction a t work 1 1 Y o u r age Is : 1 5 or under 30 to 39 1 6 t o 1 9 4 0 to 49 20 t o 29 50 t o 59 1 2 You are : M a l e F emale 13. Your e thni c origin is : Whi t e African-American Hispan ic 60 t o 64 6 5 to 7 4 7 5 or over O ther 1 4 How man y v e h icle s are owned or l eased by y o u r houae hold ? none one two three or more 15. Y oor total a n n ual househol d i ncome is: Less than $10 000 $ 30 ,000. $39, 999 --$10 ,000. $14 999 $40 ,000. $49 999 $15,000. $19. 999 $50 ,000. $ 5 9 9 99 $ 20,000 $29,999 $60 000 and ove r

PAGE 86

ENCUESTA DEL U S UARIO DE EFECTIVO/TOKEN DE METRO DADE T RANSIT Querido Usuari o : Metr o-Dade Transit qulslera saber acerca de usted y s u uso del sistema de transporte para que podamos mejorar las tarifas de los pasajes Para ello, complete esta encuesta y devuelvala a Ia persona encargada en et M etrobus, deposltela en Ia caja de encuestas ubicadas en las estaciones de tren 6 envlela por correo a Ia slgulente direcci6n de Me tro Dade Transit Cash/Token User Survey MetroDade Trans i t Agency 111 N W 1s t Street Suite 910 Mia mi, FL 33128-1999 NOTA : 51 USTE D EST A USANOO A CTUAL MENTE U N METROPASS MENSUAL HQ CO MPLETE ESTA ENCUESTA. 1 GRAC I AS I 1 <.Oue tipo de tarifa al aborda r este vehlcuto ? Tarifa Compiela (E fectivo) _tncapacttado (Efectivo) Token Estudianle (E fectivo ) Adu llo (Efectivo) Transferenc i a 6 conexi6n 2 <.Con que frecuen ci a uti l lza el s i stema de uans i to de Me tro-Dade? 5 o mas dlas por seman a 1 6 2 d l as por semana 3 0 4 dlas por semana M enos de una vez por semana 3 viajes hizo utillzando et sistema de transito de MetroDade? Ayer La semana pasada(Sab y Dom) Nota: cada vez que aborde el M etrobus, M etrorall 6 el Metromover, indlque lo como un v iaje 4. <.H a escuc h ado a l guna vez del Met ropass? S l N o ( SI ea no, vea I a expllcacl6n en I a pregunta 7 ) 5. <,Ha comprado a t guna vez e l M etropass? Si (indique cuantas veces en los uHimos 6 meses), ___ No 6 <.Porque no esta ullizando e l M etropass actualmente? (m arque las ciertas) El costo de una vez por mes es muy caro N o u tillza el s i stema de t ra ns lt o frecuentemente las taquillas de venta del M etropass no son convenientes Prefiere utilizar tokens Otro E N GLISH O N THE OTHER SIDE El M etropass i e permite abordar el M etrobus, e l M etrorail y e l Metromover todo el mes las veces que desee sin I a necesldad de pagar un ticket de t ra n sferencia E l M etropass regular cuesta $60 .00, los pases de descuento para ancianos, incapacitados y estudlantes cuestan $30.00 y los pases universitarios (college) cuestan $45.00. 7. (.Comprarla usted un M etropass mensual s l costase $50 .00 ($25 .00 para anclanos, incapacltados y estudian t es $37 .50 para estudiantes de college)? Sl, Definttivamente Quizas Probablemente no --_ Sl, P r obab temente Definitivamente no 8 <,Comprarla usted un Me tropass mensual si costase $40 00 ( $20 00 para ancianos i ncapacitados y estudiantes $30 00 para est udiantes de college )? S l, De finltiv amente Qulzjs Probablemen t e no S l, Proballlemente Definttivamente no 9 <.Comprarfa usted un Me tropass me nsu a l si a un costase $60 00 ( $30 .00 para ancianos, incapacitados y estudlantes $ 4 5 00 paoa estudiantes universitatios), pero las tarifas regulares en efectivo a u mentasen de $1. 25 a $1. 50 y si et costo de 10 to kens aumentase de $10 a $12? Sl, Definttivamente Quizas Probablemente no Sl, Proba b l emente Oefinitlvamente no 10. <.Oue otros factores le estimularlan a eomprar el M etropass? ( Marq ue las ciertas) Taqu i llas convenientes Pasaje semanal mas eeon6mioo O r de n es por co rreo Otro Deducci6n en el c hequ e de trabajo 11. Su edad esta entre los : 15 6 menos -16y 19 20y29 12. Sexo: M ascuiino Feme nino _30y 39 _60 y 64 _40y49 65 y 74 _5oy59_756mas 13 Su orlgen etnlco es:_ Blan co Afrloo-Amerlcano _Hispano_Otro 14. (.Cuantos vehlcu l os posee 6 Ilene alquilados en su resldencia ? ningu no uno dos tres 6 mas 15. Su i ngreso anuat total est{l eomprendido entre : Menos que $10 000 $30 000 $39,999 -$10, 000 $14 999 $4 0 000 $49 999 $ 1 5 ,000$ 19,999 $50 000 $59 999 --_ $20 000 $29 999 $60 000 y mas

PAGE 87

Sample Letter 75

PAGE 88

this page IS blank

PAGE 89

May II, 1993 Dear MOTA Pass & Token Sales Outlet Manager: The Metro-Dade Transi t Agency is in the process of surveying .June Metropass users to he lp us improve our transit service and fare payment methods. We are asking your cooperation in distributing the attached survey to purchasers of a June Metropass. If a person purchases more than one pass, please give them one survey for each pass bought. The Pass & Token Outlets will have no responsibility other than d istr ibuting the surveys. As an incentive to the pass user to fill out the survey, MOTA is offering a drawing of all surveys received and will award I 00 Metropasses free for the month of July. Along with your delivery of your allocated July Metropasses, you will receive a one page lis t ing of the I 00 pass numbers of winning e n tries. Please clearly disp la y this information and give winners a free June Metropass Winners will be required to turn in their June pass with the matching numbers. Keep the June winning Metropass(es) to be able to debit your account for the issuanc e of the free Metropasses. Th. ank you very much for assisting in this very important project!! Sincerely 77

PAGE 90

this page IS blank

PAGE 91

NOTES NOTES 1 Metro-Dade Transit Agency, MDTA Transit Rid!!rship Repcrt, Miami, FL: MOTA, Monthly Reports from OC1ober 1990 through June 1993. 1 Metro-Dade Transit Agency Summary of M on1h/y P= Sales by Pass Type and Outler : June 1993 Miami, FL: MOTA, 1993. J Metro-Dade Transit Agency, Ridership Technical Report: Metrobus, J\lletrorail, Metromo,,.er, Paratransit, June /993, Miami, FL: MOTA, 1993. Center for Urban Transportation Research. Fare Policy Work Order Technical J'4emorandum Number I, Tampa. FL : prepared for MOTA by CUTR, Ju ly 1993 s American Public Transit Assoeiation, Transit Fare Summary, Washington. D.C.: American Public Transit Association, January 1993. L.B Doxsey, "Demand for Unlimited Use Transit Passes," Journal of Transport Economics and Policy, Volume 18, January 198 3. 7 Richard L. Oram, Deep Discount Fares : Building Transit Productivity with lnnovaJive Pricing, Wash.in.gton D.C .: Ur ban Mass Transportation Administration, A ugust 1988. Metro-Dade Transit Agency, MDTA Transit Ridership Report: June 1993, Miami, FL: MOTA, 1993. 9 Daniel Fleishman, ConsumerBasedTran.sit Pricing at the Chicago Transit Authority, Volume I, Washington, D .C.: prepared for UMTA and CTA by Multisystems, Inc., October 1991. 10 Burak Jacobson, 1TC 1990 Fall Public Attitude Study: Attitudinal aM Ridership Analysis, Toronto: prepared for Toronto T .ransit Commission. January 1991. 11 Center for Urban Transportation Research, MDTA Survey Analysis: Draft Tampa, FL: pr

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 025524152
005 20110912120133.0
008 101006s1994 fluak l000 0 eng d
datafield ind1 8 ind2 024
subfield code a C01-00101
035
(OCoLC)668425533
040
FHM
c FHM
043
n-us-fl
090
HE4341
b .M48 1994
0 245
Metro-Dade Transit Agency fare cross-elasticity and allocation /
prepared by Center for Urban Transportation Research.
260
[Tampa, Fla. :
University of South Florida, Center for Urban Transportation Research,
1994].
300
x, 79 leaves :
ill., forms ;
28 cm.
500
"April 1994."
5 FTS
"Prepared for: Metro-Dade Transit Agency, Miami, Florida."
FTS
504
Contains bibliographic references (p. 79).
530
Also issued online.
FTS
650
Local transit
x Fares
z Florida
Miami-Dade County.
Local transit
Ridership
Florida
Miami-Dade County.
2 710
University of South Florida.
Center for Urban Transportation Research.
1 773
t Center for Urban Transportation Research Publications [USF].
4 856
u http://digital.lib.usf.edu/?c1.101
FTS
y USF ONLINE ACCESS
951
10
SFU01:002323940;
FTS