David Verbicha,b a YOU CAN’T ALWAYS GET WHAT YOU PAY FOR Averag Madhav G. Badami headways a 8-9 a.m. and Ahmed M. El-Geneidy 9-10 p.m. aSchool of Urban Planning, McGill University Transportation Research at McGill Toward a rapid assessment of public transit from multiple perspectives in North American cities bStantec Consulting Services, Inc.

INTRODUCTION CITIES AND AGENCIES AFFORDABILITY INDEX AFFORDABILITY TRADE-OFFS CONCLUSIONS

Public transit agencies must provide service that enables Affordability a. b. Transit agencies range widely in the service they offer Heavy rail Monthly Hourly Less affordable, More affordable, Less affordable, More affordable, unlimited minimum wage index better quality better quality more usage more usage riders to reach destinations easily, safely, quickly, and /streetcar Hours to earn affordably, while being financially sustainable monthly fare Affordability does not predict service quality—expensive agencies do not necessarily offer greater service quality and vice-versa Do agencies that are more expensive provide better Seattle Montreal C T A M H H A Population 3.7M Population 3.1M STM quality of service to their riders? Or, do some agencies Toronto S F SFMTA .00 10. .33 1.00 More Ridership 417M affordable, Ridership 101M Population 5.5M Less affordable, More affordable, Less affordable, less M STM .0 10.3 . 0.3 provide good service that is affordable? TTC lesser quality lesser quality less usage usage Our methodology provides a simple and replicable framework for Ridership 535M L A LACMTA .00 .00 .33 0. Average Average Passenger km ­evaluating service quality and affordability Boston S Kaccessibility C M to 1.00 .3 . divided0. by San Francisco Population 4.2M jobs by transit at 8-9 a.m. revenue km Population 3.3M Chicago MBTA B MBTA .00 .00 .3 0. We developed indicators to capture the points of view at AM peak & 9-10 p.m. SFMTA Population 8.6M Ridership 359M D DART 0.00 . 11.03 0. c. Ridership 155M CTA Averag (all modes) Less affordable, More affordable, Washington, D.C. 1. New York 8. Houston of transit riders, transit agencies, and society at large Ridership 514M headways C CTA 100.00 . 1.1 0. higher FRR higher FRR Population 4.5M 8-9 a.m. and 2. Los Angeles 9. Washington, D.C. T TTC RECOMMENDATIONS WMATA 9-10 p.m. 133. 11.00 1.1 0. 3. Chicago 10. Atlanta Ridership 409M New York 4. Toronto 11. Boston Dallas H M 0.00* . 1.1 0. Population 18.3M 5. Miami 12. Montreal Population 5.1M Altlanta We used multicriteria decision making analysis to study NYCT P SEPTA 1.00 . 1. 0. 6. Philadelphia 13. San Francisco DART Population 4.5M Ridership 3.6B Transit agencies should collect and make more data, like passenger Los Angeles Ridership 66.8M MARTA A MARTA .00 . 13.10 0. 7. Dallas 14. Seattle how agencies in large North American cities deal with Population 12.2M Ridership 128M loading, next-arrival information and other amenity details, publicly N Y NYCT 11.00 .00 1.00 0. LACMTA Unlinked trips (’000) Ridership 467M available trade-offs and conflicting goals Philadelphia M MDT 11.0 .3 1.1 0. 3,632,000 Houston Miami Population 5.4M ‡ Population 4.9M Population 5.5M SEPTA W WMATA 3.00 .0 . 0.00 Less affordable, More affordable, 535,000 Metro MDT Ridership 301M 100,000 Average 11. lower FRR lower FRR Affordability and service needs constant monitoring to ensure transit Most agencies provide service that is relatively afford- Ridership 72.7M Ridership 99.1M 66,000 Standard deviation . is serving marginalized riders Coefficient of variation 0.3 able but lower quality. While some agencies provide ex- Monthly Hourly Affordability cellent service, it is rather expensive. Two agencies pro- unlimited minimum wage index FIGURE 1 Cities,fare agencies, and modes analyzed. Population is of FIGURE 2 a, Affordability and service quality, agencies with the high qual- Future studies should consult different stakeholders to capture Hours to earn TABLE 2 Affordability index was defined as hours of minimum wage work * ity service are somewhat less affordable that agencies with lower quali- vide good and affordable­ service ­metropolitan areas, and ridership is only from analyzed modesmonthly of ­specified fare (pre-tax) to earn a monthly unlimited fare pass. Houston Metro does not of- ­different indicators and weigh them accordingly agencies. Sources: NTD 2014, CUTA 2014 Eval ua ting a bic ycle educty service. b, Affordabilitya tion and society p index (ridershipr ogram per capita), some for children : fer monthly unlimited passes, so daily unlimited fare ($3.00) was multiplied by 30 days. ‡WMATA has many monthly passes (by mode, maximum trip price, etc.), this affordable transit agencies have low usage. c, Affordability and agency Our findings indicate that paying high does not SERVICE QUALITY INDEX fare offers unlimited 30-day . Sources: APTA 2014, Government of Canada, index (FRR), cost recovery does not require unaffordable fares. translate into high service quality Doyle 2014 PERFORMANCE AND AFFORDABILITY Average Average Passenger km SOCIETY AND AGENCY INDICES accessibility to headway divided by METHODOLOGY jobs by transit at 8-9 a.m. revenue km at AM peak & 9-10 p.m. Society index Agency index (all modes) TRANSIT Ridership per Farebox Less affordable, More affordable, 1. New York 8. Houston Any metropolitan area in North America with over 3 million­ capita AGENCY recovery ratio better performance better performance 2. Los Angeles 9. Washington, D.C.

3. Chicago 10. Atlanta

4. Toronto 11. Boston people was included (see Figure 1) S Q

5. Miami 12. Montreal I

C T A T A R S I T A FRR A I R- 6. Philadelphia 13. San Francisco

A A A 3 H

7. Dallas 14. Seattle A NYCT . 1.00 TTC 0.1 1.00 C Only main transit service provider was analyzed and must STM 1. 0. STM 0. 0. N Y NYCT 10,1 1.00 11.0 0.3 .3 0.00 1.00 SFMTA 1. 0. NYCT 0.1 0. Unlinked trips (’000) operate two modes. was not analyzed C CTA ,11 0.0 11.0 0.3 1.31 0. 0.1 TTC 10.3 0. WMATA 0. 0. 3,632,000 T TTC 0, 0. . 1.00 3. 0. 0.1

CTA 10.0 0.3 CTA 0. 0. 535,000 S F SFMTA , 0. 11. 0.1 1. 0.3 0.1 Less affordable, More affordable, 100,000 WMATA 110.01 0. MBTA 0.0 0. 66,000 B MBTA ,3 0.1 11. 0.3 0. 0. 0.1 lower performance lower performance We used publicly available data (NTD; CUTA; GTFS) to SEPTA . 0.1 SEPTA 0.3 0.1 P SEPTA 3,31 0.1 1. 0. 1. 0. 0.0 MBTA .0 0.1 K C M 0.33 0.3 Photo credit: D. Verbich H M 1,1 0.0 1. 0. 10. 1.00 0. construct indicators. Indicators were normalized MARTA .1 0.1 MARTA 0.31 0.33 S K C M ,11 0.0 1.1 0. 1.11 0. 0. LACMTA .0 0.0 MDT 0. 0. M STM 0,3 0.31 10. 0. 3. 0. 0. K C M . 0.0 SFMTA 0. 0. Rider or service quality indicator consisted of average M MDT 1,333 0.0 1.3 0. 1. 0. 0. ACKNOWLEDGMENTS MDT 3.0 0.0 LACMTA 0. 0.3 FIGURE 3 Affordability and final composite indicator. Agencies like NYCT W DC WMATA , 0.0 1.0 0.3 1.0 0.3 0. accessibility, average headway (AM peak and evening), DART . 0.0 DART 0.1 0.0 (New York), TTC (Toronto), and CTA (Chicago) provide excellent service D DART 10,113 0.0 3.0 0.00 10. 0. 0.3 M 1. 0.00 M 0.1 0.00 and a comfort proxy L A LACMTA 3,30 0.1 1. 0. . 0.3 0. that is somewhat expensive to minimum wage earners. The STM (Montreal) We wish to thank Brendon Hemily for providing the fare Average 13. – 0.3 – A MARTA , 0.00 0. 0.1 1. 0. 0.00 and SFMTA (or Muni in San Francisco) provide good and affordable ser- database, Lauren Rudko for CUTA data on Canadian Standard deviation 10. – 0.1 – Average 0,30 – 1. – 1. – – vice. SEPTA (Philadelphia), MBTA (Boston), and King County Metro (Seattle) Transit agency indicator was farebox recovery ratio – Coefficient of variation 0. – 0. – agencies, Andrew Owen and David Levinson for ac- Standard deviation 0,0 – .3 – .0 – – provide adequate service that is also relatively affordable. MDT (Miami), cessibility measures for American cities, and Geneviève total fares divided by total operating cost Coefficient of variation 1.00 – 0.31 – 0. – – TABLE 3 Society index was derived by dividing annual ridership by popu- Metro (Houston), MARTA (Atlanta), and DART (Dallas) provide poorer ser- ­Boisjoly for help with calculating accessibility measures lation within service area of transit agency. This is the benefit to society by vice that is less affordable, while LACMTA (LA Metro) provides poorer yet for Toronto and Montreal. DV was supported by a TABLE 1 Service quality index was calculated by averaging and rescaling providing sustainable and collective . The agency index was cal- more affordable service. Finally, WMATA (Washington, D.C.) provides ser- vice that is expensive and performs low on the composite service indicator. SSHRC Joseph Armand Bombardier Canada Graduate Society indicator was ridership divided by service area accessibility index, headway index, and comfort index. Sources: NTD 2014, culated using farebox recovery ratio—the amount of operating cost cov- Sources: NTD 2014, ­Scholarship. CUTA 2014, GTFS (fall 2014), Access Across America (Owen and Levinson, 2014) ered by fare revenue, a measure of financial efficiency. population CUTA 2014