Some toll economics

March 22, 2011

…in the context of a congestion charge…

Toll Revenues

They are mostly driven by the price elasticity of demand (PED). In a nutshell, it is about how the demand for a service (here road access) is affected by a price change.
The greater the decrease in demand for a given price increase, the greater the elasticity will be.

It is an important concept since it can allow to estimate what should be the price of a toll to reduce the congestion to a certain level or what could be the revenue of a tolled infrastructure at a given price.

It is a common occurrence to see the PED under estimated: it has been under estimated in London, in Stockholm, on the Golden Ears Bridge as well as on most of the tolled urban road infrastructure around the world.

According to [5], the elasticity to a congestion charge has been notoriously under estimated by the MOT in the case of Port Mann bridge study. One reason here like elsewhere is that the lower the elasticity is, the easier it is to justify a new infrastructure…while the greater the elasticity is, the lower the congestion charge need to be to address the congestion (or the more difficult it will be to charge high price on a new tolled infrastructure).

That is, the common denominator is that usually a congestion charge scheme is successful at achieving its main goal of reducing the congestion, but that also means it yield usually to less revenue than expected. Here after are some elasticity numbers. the bigger (in absolute), the easier it is to reduce congestion

Study Estimate Comment
Vancouver 2008 study [5] -0.25 Public transit as in 2008
Vancouver 2006 study [3] -0.32 from survey (variance -0.23 to -0.41)
Seattle [4] -0.9 to -1.6 from a pilot project (2003-2005)
Toronto 407 ETR [5] -0.7
Spain [5] -0.2 Congested bridge
France Expressway -0.3 Government recommendation
New York bridge [5] -0.1 Average parking price in Manhattan is $431/month
Singapore cordon [5] -0.19 to -0.58 average tax on car purchase is S$45,000 (~$37,000)
London Congestion area [4] -0.53 to -0.96 computed on the initial £5

Operation cost

A toll system incurs some operating, maintenance and capital cost. Usually it is a reason advanced to not implement it:
the most part of the revenue could be absorbed by the operating cos of the system”, and skeptic people will advance the case of London [9][10]. figures hereafter show that doesn’t necessarily need to be the case, like Stockholm or Singapore examples show

City Capital cost (M) Operating&maintenance cost (M) Revenue (M) Gantries Transponders
London [4] $180 $150 $280
Stockholm [4] $150 $26 $120 18 unknown
Singapore [6] $116 $10 $90 45 700,000
Golden Ears Bridge [1] $60 (8 years) 1 5,000
Port Mann Bridge [2] $18 (5 years) NA 1 200,000

But if we talk more of road pricing today than yesterday, it is not because there is more congestion today than yesterday but because it is dramatically much more cheaper to implement it nowadays.

Below are the answers to a recent procurement to equip a HOT lanes on the I85 by the Georgia DOT the terms of the contract being [7]

  • 410,000 sticker tag transponders ( 350,000 regular interior windshield tags + 60,000 external readers at 37 gantry locations
  • 37 readers at 37 gantrries location
Bidders TransCore neology SIRIT Kapsch
Transponders 1,005,300 1,076,500 1,697,500 8,680,000
RFID Reader Subsystems Equipment 134,665 175,299 297,66 1,600,975
Support Service 22,400 28,000 34,000 22,086
Total Bid Pric 1,165,235 1,279,799 2,019,163 10,303,061

Detail of the Transcore winning bid shows that the contract is based on

  • $1.59 per sticker tag transponders [8]
  • $3000 per RFID reader and matching antenna

What seems costly is the video identification (not included in the procurment above), but when a sticker (transponder) can cost as low as $1.59, one has to consider the alternative to outfit the whole vehicle park of its jurisdiction with it

That is what has been done in Singapore. Metro Vancouver is not a city state, but 65% of the BC vehicles are registered in Metro Vancouver and Vancouver could be considered isolated enough of out of Province road access, to make this option certainly worth of consideration for the whole province park.

Effect of the Congestion charge

By virtue of the PED, it reduces the congestion and make trip more predictable, for the remainding users, so those get some value (in time) for the toll.

That applies on Transit as well. bus speed has increased by 6% in London congestion charge area, and 3% in the surrounding (notice that is including an increasing dwelling time due to significant higher patronage, and previously existing bus lanes) [4], but more important is the very significant increase in reliability allowing the agency to reduce lay over and realize significant operating saving.

The shift on public transit can be very significant (numbers from [4])

  • In London, the bus service got a 37% patronage increase in the congestion charge area (notice that people also shift from subway to bus, eventually due to more attractive bus ride which became more predictable and faster after the CG came in service).
  • In Stockholm, patronage has increased 5% network wide, in a city already having a high modal split in favor of transit (30% trip on transit region wide [11])

Thought that could be the goal of the policy maker, the shift on the public transit system could be a cause of headache if he is not designed to absorb the increased patronage. In Stockholm, the introduction of the congestion charge has been coupled with a significant increase in the public transit offer.

The congestion charge in addition to reduce traffic, has also allowed to increase the road safety wherever applied, and obviously reduced the pollution and inherent health hazard due to it [4]

All in one, though initially not necessarily well received by the opinion, the congestion charge, wherever applied long enough, has sustained the test of time, and greater acceptance by the public [4], that eventually in recognition of the social benefits it is able to deliver.

[1] Open road tolling for BC Golden Ears Bridge – $60m contract TOLLROADSnews, feb 05, 2007.

[2] CS signs $18m toll system contract with BC gov toiler for Mt Mann Hwy 1 AET TOLLROADSnews May 11, 2010.

[3]Estimating commuter mode choice: A discrete choice analysis of the impact of road pricing and parking charges“, K. Wasbrook, W. Haider and M. Jaccard, Transportation (2006) 33:621–639.

[4] Road Congestion Pricing In Europe: Implications for the United States H. W. Richardson and C. H. C. Bae , Edward Elgar Publishing, 2008.

[5] From Freeway to feeway: Congestion pricing policies for BC’s Fraser River crossing, Peter Wightman, Simon Fraser University, 2008.

[6] Road Pricing Volume 9, Theory and Evidence, Georgina Santos, Elsevier 2004.

[7] TransCore win GA/I-85 HOT lanes tag-reader contract with $1.59/tag 6C GEN2 TOLLROADSnews, Sep 16, 2009.

[8] For matter of comparison, Translink leases Golden Ears bridge transponder at $1 per month

[9]Without denying the efficiency improvement, in direct economic term as well as in more indirect term (pollution…), some experts consider that those efficiency gains or social benefit, are not enough to offset the operating cost of the system [10]

[10] The London congestion charge: a tentative economic appraisal, R. Prud’homme and J.P. Bocarejo, Transport Policy Volume 12, Issue 3, May 2005, Pages 279-287.

[11] Transport planning in the Stockholm Region, Hans Hede, METREX International workshop, Moscow, June 2006

5 Responses to “Some toll economics”

  1. Rico Jorimann Says:

    Nice post,

    Good to see you are posting again.


  2. Voony Says:

    thanks. It eventually takes a bit of time to gather (and present information) like the one in this or previous post, and life sometimes doesn’t give you as much time as you wish.
    Anyway, after a long interruption, I hope to post more often.

  3. […] discussion below, grounded on a previous article as well as traffic data [6], starts of the viewpoint of what it takes to alleviate the congestion […]

  4. Thank you, I have recently been looking for info approximately this topic for a long time and
    yours is the greatest I’ve came upon so far. But, what about the conclusion? Are you sure concerning the supply?

  5. […] the congestion by 15 to 85% [1], the case is more moot for infrastructure tolling. A recurrent and unfounded critic is that congestion toll revenue doesn’t cover the toll collection operation. In that […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: