1 of 39

The challenges facing public transit before, during, and after the pandemic

Brian D. Taylor (with Jacob Wasserman) 21 March 2021

Professor of Urban Planning & Public Policy

Director, UCLA Institute of Transportation Studies

2 of 39

Ridership Trends

2

3 of 39

Transit use was down during the 2010s in California, though the absolute decline was modest

3

Data source: National Transit Database

4 of 39

But the fall in trips per resident was substantially greater: Mostly down since 2008

4

Data source: National Transit Database,

American Community Survey, and U.S. Census

5 of 39

And Lots of Variation across Modes

5

Data source: National Transit Database

6 of 39

A stunning collapse in early 2020, and a slow recovery since

6

Estimated change in transit demand, measured by Apple Maps routing requests. Details at https://covid19.apple.com/mobility

7 of 39

An even slower recovery in the US

7

Estimated change in transit demand, measured by Apple Maps routing requests. Details at https://covid19.apple.com/mobility

8 of 39

And slower still in California

8

Estimated change in transit demand, measured by Apple Maps routing requests. Details at https://covid19.apple.com/mobility

9 of 39

Transit-rich Bay Area saw the biggest per capita ridership drop in California

9

(January – August 2020)

Ridership visualized as Unlinked Passenger Trips/Capita, from FTA National Transit Database

Adapted from visualization available here

10 of 39

Asymmetries of Transit

10

11 of 39

Transit Concentration and Asymmetry

A few people make most of the trips
A few cities and neighborhoods generate most of the trips
A few operators (and lines on those operators) carry most of the passengers

These trends have been mostly magnified by the pandemic

11

12 of 39

Asymmetries of Transit

12

Data source: 2009 and 2017 National Household Travel Surveys

13 of 39

Transit users are more likely to be low-income and people of color during the pandemic

13

Data sources: APTA, “Who Rides Public Transportation” (2017), Transit App survey (April 2020) https://medium.com/transit-app/whos-left-riding-public-transit-hint-it-s-not-white-people-d43695b3974a

14 of 39

A Few People Make Most of the Trips

2% of Greater Los Angeles residents ride very frequently

~45 trips/month

20% ride occasionally

~12 trips/month

78% ride transit very little or not at all

< 1 trip/month

14

15 of 39

A Few Areas Generate Most of the Trips

15

Data source: Blumenberg et al., 2015

16 of 39

A Few Areas Generate Most of the Trips

16

17 of 39

A Few Operators and Lines Carry Most of the Passengers

Greater Los Angeles: The biggest, highest volume lines have lost of the most passengers

LA Metro alone accounted for 72% of CA ridership loss, 2010-2017.

Bay Area: Off-peak service and circumferential lines losing the most riders

17

* unlinked total MTC trips minus linked transbay BART trips

Data source: BART origin-destination matrices

and National Transit Database

18 of 39

Why ridership was and is down

18

19 of 39

External factors like increased auto access, high housing costs, and a global pandemic play a bigger role than changes in fares or service

19

20 of 39

Private was up, and is likely up further

20

Data source: IPUMS (Ruggles et al., 2019)

21 of 39

Increased vehicle access likely had a very large effect outside of the Bay Area

21

22 of 39

High housing costs, longer commutes are likely playing a role, especially in the Bay Area

Jobs-housing ratio becoming more imbalanced
Commute distances getting longer
Context: California housing affordability crisis

22

* straight line distance

Data source: LEHD Origin-Destination Employment Statistics

23 of 39

Ridehail likely had some effect before the pandemic

Very hard to say more without better TNC data

23

24 of 39

Rail and smaller systems cut more service than large bus operators

Bus

(< 10 million passengers)

Bus

(≥ 10 million passengers)

Rail

24

Averages

71%

82%

73%

Pct of Pre-Pandemic Service Levels Offered as of September 2020

Our survey respondents indicated that a lot of service has been cut (as of September 2020).
Most service is below 90% of pre-pandemic levels.
Bus service at smaller agencies (with fewer than 10 million UPT) is especially affected, though a handful of smaller agencies--and smaller agencies alone---also reported increasing service levels during the pandemic.
Larger agencies and rail-only agencies have mostly had more modest cuts in service.

[FROM SKIPPED SLIDE 21]

It is worth noting that, at most agencies, multiple changes have occurred over the course of the pandemic.
From our survey we learned:

95 percent of agencies responded that they at some point have modified service during the COVID-19 pandemic
Of those who operate bus service and adjusted service, 89 percent adjusted service more than once
About a quarter of respondents had 4 or more service adjustments.

25 of 39

Big changes to fare collections in the pandemic

25

	All Respondents		California Only
No change in fares	24	38%	9	31%
Fares no longer collected	20	31%	11	38%
Other	15	23%	6	21%
Fares are not expected but not expressly discontinued	4	6%	3	10%
Agency never collected fares	1	2%	0	0%

Most “Other” responses explained that fares were discontinued but have since been restored by survey day.

26 of 39

Big changes to fare collections in the pandemic

26

	All Respondents		California Only
No change in fares	24	38%	9	31%
Fares no longer collected	20	31%	11	38%
Other	15	23%	6	21%
Fares are not expected but not expressly discontinued	4	6%	3	10%
Agency never collected fares	1	2%	0	0%

Most “Other” responses explained that fares were discontinued but have since been restored by survey day.

27 of 39

Big changes to fare collections in the pandemic

27

	All Respondents		California Only
No change in fares	24	38%	9	31%
Fares no longer collected	20	31%	11	38%
Other	15	23%	6	21%
Fares are not expected but not expressly discontinued	4	6%	3	10%
Agency never collected fares	1	2%	0	0%

Most “Other” responses were that fares were discontinued but have since been restored

28 of 39

How is service changing in the pandemic?

28

29 of 39

Most transit systems are adjusting service based on observed changes in demand

29

Ridership data	85%
Direction from public health officials	53%
Crowding data	39%
Operating budget / finances	31%
Title VI considerations	27%
Geographic locations of essential jobs	24%
Direction from agency’s board	19%
Local COVID-19 testing data	19%
Employment data	18%
Fare revenues	11%
Non-fare revenues	6%

30 of 39

Ridership is increasingly concentrated in the poorest and non-white areas

30

Data sources: stop-level ridership data for October 2020, courtesy of LA Metro, LA Metro GTFS data.

October 2020 Service-Ridership Match

Darker colors = More riders than service
Ligher colors = More service than riders

31 of 39

The distribution of riders and service has grown more equal during the pandemic

31

Data sources: stop-level ridership data for October 2020, courtesy of LA Metro, LA Metro GTFS data.

October 2020 Service-Ridership Match

	LA Metro Bus Service	Houston Metro Bus Service
April 2020	0.49	0.41
October 2020	0.43	0.38

A “Gini coefficient” for service/ridership match

Lower coefficients imply that service and ridership closely align across all Census tracts

Service and ridership have become more closely aligned between April and October 2020, in both LA and Houston.

32 of 39

Smaller transit agencies and those in California report that financial shortfalls are affecting service during the pandemic

32

33 of 39

Takeaways

33

34 of 39

Takeaways

In spite of expanded and improved service, transit use was mostly down in the 2010s

Increased auto access likely the principal culprit outside of trips to/from downtown San Francisco
This may be an even bigger issue post-pandemic

Spring 2020 ridership declines were sudden and dramatic

Recovery has been slow and has largely stalled

The social service role for public transit is greater than perhaps ever before

Returning riders are even more likely low-income, people of color

34

Overall, transit has changed a lot since the pandemic began
Our research suggests it is continuing to change as the pandemic continues

These changes appear to be driven by considerations in ridership, crowding data, health guidance, and financial conditions

Where transit supply has changed varies significantly by agency size, geography, and--as we identified in our visualization--Census tract.

The effects of these changes on riders and former or would-be riders is the focus of our team’s next research paper, due in December.

Crowding could be a problem

It’s poorly defined
Many agencies appear to be poorly equipped to effectively communicate maximum capacities to their riders and operators.
Real-time data that could be useful for communicating crowding levels to riders before they board is not widely used and not used to its greatest potential where it is used
In some places, transit supply might not increase before demand does (due to financial and labor constraints) which would likely result in ridership loads above the socially-distanced maximum capacities, i.e., crowding.

Nevertheless, the public health literature suggests that transit is currently about as safe as other public settings. Much depends on the community’s public health response, however, and that is beyond the control of transit operators.

35 of 39

Takeaways

The future of concentrated activity centers (like CBDs, universities, and airports) will have a big effect on the future of public transit

If the activity density of such centers stays down, ridership will likely be very slow to recover
You will hear more about future scenarios for public transit in a bit

The December and, especially, March federal COVID relief bills have mostly stabilzed transit finances in the near term

Longer term funding from non-federal sources and from fares remain uncertain
We (at UCLA ITS) are launching a study to examine this

35

Overall, transit has changed a lot since the pandemic began
Our research suggests it is continuing to change as the pandemic continues

These changes appear to be driven by considerations in ridership, crowding data, health guidance, and financial conditions

Where transit supply has changed varies significantly by agency size, geography, and--as we identified in our visualization--Census tract.

The effects of these changes on riders and former or would-be riders is the focus of our team’s next research paper, due in December.

Crowding could be a problem

It’s poorly defined
Many agencies appear to be poorly equipped to effectively communicate maximum capacities to their riders and operators.
Real-time data that could be useful for communicating crowding levels to riders before they board is not widely used and not used to its greatest potential where it is used
In some places, transit supply might not increase before demand does (due to financial and labor constraints) which would likely result in ridership loads above the socially-distanced maximum capacities, i.e., crowding.

Nevertheless, the public health literature suggests that transit is currently about as safe as other public settings. Much depends on the community’s public health response, however, and that is beyond the control of transit operators.

36 of 39

Further information

36

37 of 39

For more information

Please see (http://its.ucla.edu/covid) for:

Visualization tools
Background data
Working papers
Policy briefs
Publications

37

38 of 39

Research Team Members

Yu Hong Hwang
Hannah King
Michael Manville
Julene Paul
Madeline Ruvolo
Andrew Schouten
Samuel Speroni
Brian D. Taylor
Jacob L. Wasserman

Evelyn Blumenberg
Tiffany Chu
Tianxing Dai
Eric Dasmalchi
Maya Desai
Hao Ding
John Gahbauer
Mark Garrett

This research was generously funded by:

38

39 of 39

39

Thank You

Brian D. Taylor, PhD, FAICP

btaylor@ucla.edu

(310) 903-3228

Director, Institute of Transportation Studies; Professor of Urban Planning and Public Policy, UCLA Luskin School of Public Affairs

Find research reports and policy briefs at its.ucla.edu

Transportation finance, public transit and innovative mobility