Understanding the Wider Value of Honor Based Ticketing in Light Rail

Professor Graham CurrieJames Reynolds

PUBLIC TRANSPORT RESEARCH GROUPInstitute of Transport StudiesMonash University, Australia

Light Rail 2016 – Transforming the urban transport landscapePullman Melbourne on the Park25th February 2016

1 Introduction

2 Research Context

3 Model Development

4 Results

5 Discussion and Conclusions

Agenda

2

• Passengers must have valid ticket

• Random ticket inspections for enforcement

• Allows passengers to board and alight at any door

Melbourne, like most Light Rail systems, uses HonorBased or Proof-of-Payment (POP) fare collection

3

All door boarding and alighting in Melbourne, Australia

This is often criticised in the media/community because of high fare evasion rates

4

0%

5%

10%

15%

20%

25%

0% 20% 40% 60% 80%

Far

e ev

asio

n r

ider

ship

sh

are

-w

ith

in

mo

de

Total lost revenue - share between modes

Lost Revenue

Tram

TrainBus

Source: ITS (Monash) analysis of the Fare Evasion and Valid Concession Percentage Survey - 2011

• Pay fare to driver on boarding, or show valid pass, transfer etc.

• Allows close monitoring of fare payment

• Requires all passengers to board by the front door

An alternative is Pay-on-Entry (POE) fare control such as adopted in Toronto

5

Front Door Boarding on a Toronto Transit Commission Streetcar

The trade-offs between POE and POP on LRT have not been fully explored

6

Pay-on-Entry (POE)Pay driverLonger boarding timesLower fare evasion

Honor Based / Proof-of-Payment (POP)Pre-purchase ticketShorter boarding timesHigher fare evasion

versus

Fare Revenue Increases?

Operational Costs / Savings?

Capital Costs?

Fare Evasion Losses?

Operational Costs / Savings?

Capital Savings?

THIS RESEARCH:

• Develops a model of the Melbourne Tram Network to directly compare the overall financial impacts of POE with POP, and

• Investigates how:

• Reduced fare evasion under POE fare control; compares to

• Costs due to slower boarding times compared to Honor Based/ POP system.

Currie G and Reynolds J (2016) ‘Evaluating Pay-on-Entry Versus Proof-of Payment Ticketing in Light Rail Transit’ Transportation Research Record - Transportation Research Board 95th Annual Meeting January 2016

1 Introduction

2 Research Context

3 Model Development

4 Results

5 Discussion and Conclusions

Agenda

77

Trams in Melbourne have open access, while trains and buses have more control of fare payment

8

Melbourne Rail Network

• Barrier gates at central stations

• Roving ticket inspectors

Melbourne Bus Network

• Enter by front door only

• Validate smartcard in front of driver

• Some roving ticket inspectors

Melbourne Tram Network

• Enter by any door (open access)

• No interaction with driver

• Roving ticket inspectors

Melbourne’s trams use the “myki” smartcard system, with passengers required to “touch on” after boarding

9

Touch On

myki Smartcard

Images source: www.ptv.vic.gov.au

Board

Ride

Touch Off

If travelling wholly in Zone 2

Alight

*Except in the FREE TRAM ZONE in the city center

Fare Evasion Trends

10

14% fare evasion

20% fare evasion

“myki” smartcard introduction period

Source: Public Transport VictoriaVictorian Official Fare Compliance Series May 2015

Victorian Fare Compliance Rate

6%

The Age Newspaper: October 3, 2011

The Age Newspaper: January 2, 2015

Would Melbourne Trams have been better off with POE instead of the myki POP system?

11

The Age Newspaper: June 2010, October 2011 and June 2011

Melbourne trams vs Toronto streetcars

12

Melbourne Toronto

• Some of the largest streetcar systems in the world

Melbourne = 167 kms (104 miles) Toronto = 71 kms (44 miles)

• Different Fare Control Systems

Melbourne = POP Toronto = mostly POE

Melbourne vs Toronto

13

Melbourne

– Tickets pre-purchased– Validated during or after

entry– No interaction with driver

Toronto

– Pay-on-Entry (POE) fare paid to farebox in front of driver

– Pass or transfer must be shown to driver

– Front door boarding only– Some Proof of Payment (POP)

zones but on only a limited number of routes

• Melbourne and Toronto dwell time surveys

• Regression models developed including one with a factor for fare control type

Previous Research - Currie, Delbosc and Reynolds (2012)

14

Dwell time = 3.7 + 0.9a + 0.7b + 13.4c – 6d + 3.4e + 9.8f

Where:a = Number of boardingsb = Number of alightingsc = 1 if 4 doors, else 0d = 1 if platform stop, else 0e = 1 if steps, 0 otherwise,f = 1 if pay-on-entry, 0 otherwise

Source: Currie, G., A. Delbosc, and J. Reynolds, Modeling Dwell Time for Streetcars in Melbourne, Australia, and Toronto, Canada. Transportation Research Record: Journal of the Transportation Research Board, 2012. 2275: p. 22-29.

• Model implies that average dwell time for POE is 9.8 seconds higher per stop than for POP

1 Introduction

2 Research Context

3 Model Development

4 Results

5 Discussion and Conclusions

Agenda

1515

16

Model Structure - Outlinevs

Operational

Impacts

Ridership and Fare Payment

Impacts

Annual Revenue /

Cost Impacts

Capital Cost

Impacts

Discount Cash Flow Analysis

Pay-On-Entry (POE) fare

control

Proof-of-Payment (POP)

fare control

• Compares POE fare control impacts with the (existing) POP across 22 of the 26 tram routes in Melbourne

• Determines operational, ridership and fare payment impacts

• Calculates capital cost and annual revenue / cost impacts

• Uses a Discount Cash Flow Analysis to calculate a BCR of switching to POE

Overall model - Detail

Fewer Ticket Validation Machines

Longer Dwell Times

Longer Passenger

Journey Times

Decreased Ridership

Longer Vehicle

Journey Times

Increased Vehicle

Requirements

Higher Operational hours & km

Increased Operating

Costs

Increased Capital Costs

Larger Vehicle Fleet

Reduction in Fare Evasion

Removal of Inspection

Staff

Increased Fare Payment

Decrease in Fines Levied

Decreased Revenue

Increased Revenue

Decreased Operating

Costs

Decreased Capital Costs

Better Financial

Performance

Worse Financial

Performance

Less Maintenance

vs Pay-On-Entry

(POE) fare control

Proof-of-Payment (POP)

fare control

17

1 Introduction

2 Research Context

3 Model Development

4 Results

5 Discussion and Conclusions

Agenda

1818

Impact of Conversion – Honor/POP vs POE

19

10%Ridership Decline

49 Additional

LRV’s Needed

8 163

66

-280

-230

-180

-130

-80

-30

20

70

Red

uced

Far

eE

vasi

on

Sav

ed A

OC

osts

Sav

edV

alid

ator

Cos

ts

Sav

edV

alid

atio

nM

achi

nes

Savings Resulting from Pay the Driver Ticketing

Cost Savings

Cha

nge

in C

osts

(M

)

Costs Resulting from Pay the Driver Ticketing

Cost Increases

-17 -9 -30 -276

-280

-230

-180

-130

-80

-30

20

70

Far

e R

educ

tion

Less

Rid

ers

Low

er F

E F

ine

Rev

enue

Incr

ease

d T

ram

Ope

ratin

gC

osts

Less

Tra

m/L

RV

Veh

icle

s

Operating/ Annual Costs Capital Costs

Operating/ Annual Costs Capital Costs

Open access saves $29M p.a. operating costs & $210M in Capital – increases ridership 10% and saves 49 LRVs

20

Melbourne – Open Access; Proof of Payment Ticketing

Toronto – Pay the Driver Ticketing

Source: Currie, G and Reynolds J (2016) ‘Evaluating Pay-on-Entry vs Proof-of Payment Ticketing in Light Rail Transit’ Transportation Research Record Journal of the Transportation Research Board - 95th Annual Meeting of the Transportation Research Board Washington DC January 2016

• +$29M p.a. in net Operating Costs each year

• +$210M in net Capital Costs• 10% less ridership due to delays• 49 additional LRV’s needed (+14%

of fleet)

21

Aggregate ResultsFinancial Analysis Annual Capital

POE Benefits ($AU)• Reduced fare evasion losses• Reduced staffing costs• Reduced maintenance of validation machines• Fewer ticket validation machines

8.1m15.8m3.1m

65.5m

POE Costs ($AU)• Lower fare revenue• Lower fine revenue• Increase vehicle operation costs• New vehicles

17.4m9.2m

29.8m276.0m

Total ($AU) Benefits – Costs -29.4m -210.5m

Discount Cash Flow Analysis BCR

• 30 years at 6% discount rate 0.44

22

Route Based Results

R² = 0.1851

0.3

0.35

0.4

0.45

0.5

0.55

0.6

0 5,000,000 10,000,000 15,000,000

BC

R

Annual Ridership

Benefit Cost Ratio vs Route Ridership

R² = 0.1202

0.3

0.35

0.4

0.45

0.5

0.55

0.6

5 6 7 8 9

BC

R

Stops per Route Kilometer

Benefit Cost Ratio vs Stops per Route Kilometer

• BCR of POE goes downas stops per kilometre goes up

• BCR of POE goes up as ridership increases

• However R2 is < 0.2

23

Sensitivity Analysis

0.30

0.35

0.40

0.45

0.50

0.55

0.60D

wel

l tim

e

AO

ann

ual c

ost

Ann

ual o

pera

ting

hour

s

Cre

w c

ost

per

vehi

cle

hour

Tic

ket

valid

ator

s pe

r tr

am

Rid

ersh

ip e

last

icity

Cap

ital c

ost p

er v

ehic

le

Far

e ev

asio

n ra

te

Cap

ital c

ost p

er t

icke

t va

lidat

or

Rev

enue

from

fin

es

Cos

t pe

r ve

hicl

e km

Ann

ual o

pera

ting

kms

Dis

coun

t rat

e

Val

idat

or m

aint

enan

ce s

taff

cos

t

Ann

ual r

ider

ship

Far

e re

venu

e pe

r pa

ssen

ger

Ben

efit-

Cos

t Rai

to

Most Sensitive to Dwell Time

1 Introduction

2 Research Context

3 Model Development

4 Results

5 Discussion and Conclusions

Agenda

2424

25

Findings• Melbourne trams have worse financial performance under POE than

POP

• $AU27.0m annual benefits and $AU65.5m capital savings

• But $AU56.4 annual costs and $AU276.0m capital expense

• BCR of only 0.44

• Costs associated with longer stop dwell times far outweigh the benefits of POE for reducing fare evasion and staffing costs

• Lower levels of ridership, increased fleet size and operating costs are significant financial penalties of operating a POE fare system

Toronto – should stop using POE!...

26

Front Door Boarding on a Toronto Transit Commission Streetcar

... AND THEY HAVE

27

www.worldtransitresearch.info

29

ALSO:

NEW PTRG WEBSITE

PTRG.INFO

Join the ITS (Monash) LinkedIn group to keep informed of our activities