Friday 29 July 2011

San Francisco's innovative variable parking pricing scheme

It isn’t strictly road pricing in the usual sense of the word, but as it is pricing usage of the roads in one sense – parking – I thought I would write about it, particularly since the system in place in San Francisco is so revolutionary.

SFPark intelligent meter

It is called SFPark, and according to the New York Times it is intended to reduce congestion by managing parking (and information about parking) through varying pricing. It is a trial, but what it offers is the opportunity to transform how parking is managed in central cities.

Its website is comprehensive, with lots of useful information, and says:

SFpark is testing its new parking management system at 7,000 of San Francisco’s 28,800 metered spaces and 12,250 spaces in 15 of 20 City-owned parking garages. The pilot phase of SFpark will run until summer 2012. Federal funding through the Department of Transportation’s Urban Partnership Program pays for 80 percent of the SFpark project.

The NYT says:

The approach is twofold: to change the price of a parking space according to demand and thereby keep spaces open on every block, and to lead drivers to open spaces using an array of sensors, eliminating congestion caused by circling drivers

Prices are able to be varied from US$0.25 to US$6 an hour, with typical prices being between US$2 and US$3.50. The obvious implication is that when times are quiet, parking is cheap which may encourage people to enter the city by car. When busy, the price will be high, dissuading driving, but still managing congestion by ensuring there is always a space free.

The technology involved is for sensors at the car parks that detect when they are occupied. Motorists can use the website or the iPhone app to find parks and their prices. They pay using coins, credit or debit cards.

Prices are only to change once a month, by small increments, so it is not as if it is real time – although it could be. I’d suggest it may make a lot of sense to vary pricing on a daily basis, with algorithms developed to match days of the week, public holidays and special events.

If successful, it is possible to see this sort of trial being rolled out elsewhere.

Clearly it has the great benefit of actively managing the parking stock and optimising revenue by matching prices to usage. It also reduces congestion in two ways. Firstly, by matching pricing to demand, it means that demand to use the adjacent roads will be (in part, because private parking and use of roads as through routes is not affected) influenced by the pricing of the parking. Secondly, by guaranteeing a minimum level of availability, cars will not need to drive around searching for parks. If motorists are willing to pay the demand based priced, they will get parks and can go to them directly, saving them time and reducing usage of the roads.

It is not a substitute for full road pricing, because clearly this solution is only going to be effective in busy commercial districts, and it will not relieve busy highways or arterials of traffic in itself. However, it does have the potential to make a real difference. It has the added value of not penalising the businesses located in the area, because pricing matches demand. Presumably, if parking is expensive, there are lots of people seeking to go the commercial district, likewise if there are not, it will be cheap, making it more attractive.

Could this be an easier first step for many cities rather than congestion charging, in order to simply get better use and revenue out of their parking stock, as well as having positive effects on congestion?

Reactions to Australian tax reform discussion paper

As reported yesterday, the Australian Federal Government has floated both heavy vehicle road user charging and congestion charging as options for reform of motoring taxation in its latest discussion paper.

Initial reaction in Australia has been hardly positive:

Transport and Logistics News reports that the Australian Trucking Association wants a shift away from annual registration fees to higher fuel tax.  It argues against the high annual registration fees (which are designed to represent the difference between the damage larger trucks impose on the roads and the fuel tax they impose) for imposing high costs on small operators and not reflecting usage (which is correct).  However, it also says there is no link between where trucks are used and where revenue is spent.   Fuel tax is incapable of doing this as well.  Chief Executive Stuart St. Clair argues against a distance based charge because:

Mass-distance-location pricing would increase freight costs in rural areas. It would be a compliance nightmare for trucking operators, with enormous administrative and billing costs. It would also breach one of the fundamental principles of tax reform: simplicity

It would only increase costs for operations by trucks that have very high usage over long distances, but then again the long thin rural networks are the most expensive to maintain.   The "compliance nightmare" is ridiculously overstated, when it could be a simple monthly direct debit system if done electronically.   As for simplicity, why not argue that truck operators get paid by the government, which imposes a flat tax on all goods movement based on time it takes to move it?  How about charging all freight according to weight?  The market isn't simple, and neither should roads be treated in the same way as defence.

He also opposes congestion charging because he claims politicians would only impose it on trucks, which would be pointless.  He's right, in that a congestion charge on trucks would not be worth it (and would not deliver the benefits to trucks paying), but it is not a reason to scrap the idea.

Meanwhile, according to the Sydney Morning Herald, Prime Minister Julia Gillard has now denied the Government is even interested in congestion pricing, but said it was up to state governments.

This may contradict a report from the same newspaper that her Treasurer, Wayne Swan, says everything is up for grabs (in this area).

According to the Rouse Hill Times, NSW Roads Minister Duncan Gay argued in favour of more Federal funds for a rail link, instead of a congestion tax (showing his political nouse if not economic).  He said (according to Byron News):

it would be the very worst thing to do to the commuters of Sydney, who have no alternative in many cases but to use their cars... This is a state without an alternative form of public transport and if you haven't got alternative, good, public transport, in many areas of the city you are condemning people to pay twice"

The "paying twice" point is difficult to understand, but he also either doesn't understand that congestion charging doesn't need other modes to generate benefits, or doesn't think he could ever sell the idea.

Of course the real problem is that congestion taxes may only be able to get public support in most cases if they replace other taxes.  Then a conversation can be more easily had.

What clouds all of this debate is the very unpopular move by the Federal Government to push for a carbon tax, that would impose higher charges on the entire transport sector.  Given that Government's apparent penchance for increasing taxes, it is likely that many will see it as not being as trustworthy as it may wish in terms of tax reform that could be seen as revenue neutral.

Australian Federal Government floats road pricing in new tax discussion paper

After a cool reception for the Henry Tax Review, commissioned by the previous Labor (Federal) Government in Australia (under Prime Minister Kevin Rudd), the current Labor minority Government (supported by the Greens and three Independent MPs) has released a tax discussion paper called "Tax Reform: Next Steps for Australia" (Full PDF file)

Among other things it proposes:
- Heavy vehicle road user charging based on mass, distance and types of roads used (implying a shift from vehicle registration and fuel taxes) to be developed through the state level Council of Australian Governments;
- Congestion charging variable by time and location.

It also notes that technology may allow road pricing to be implemented more widely than those examples.
Page 29 of the discussion paper explains the thinking, none of which should be a major surprise.  It criticises the current system (which is widely implemented globally) as follows:

current arrangements provide inadequate incentives to operators to choose routes and vehicle configurations that minimise road damage and costs on others. The current charging system also results in significant cross subsidies between different types of heavy vehicle operators.

Quite. Vehicle configurations are taxed based on annual registration fees, and fuel tax only encourages road use that minimises fuel consumption, which may not minimise road damage.  In addition, it is clear that the heaviest longest hauling trucks are likely to be cross subsidised by the shorter haul vehicles, because fuel tax over long distances means less charged per km than for short or urban trips, because fuel consumption is so much more efficient.  Road costs do not decrease though.   In addition, with registration fees reflecting weight, the vehicles that travel the least distance effectively cross subsidise the ones that travel the most.   

On congestion charging, the key point is the limits to building new infrastructure:

Building more roads is one way to try to address congestion concerns, but equally important is better utilisation of existing infrastructure. There are physical, environmental and financial limits on the construction of new transport routes.

What does it mean? Well a Tax Forum is to be held to discuss all of the issues in the paper (and most are unrelated to transport).  Questions offered for discussion are:
  • Should Australia consider ways to more closely link road charging to the impact users have on the condition and upkeep of roads?
  • Is there a case to more closely link road charging to the impact users have on the level of congestion on particular roads?
My answer to both is "yes", but this is about convincing the transport sector and the public.  For it is a prelude to potentially wide ranging changes in how roads are charged for in Australia.
If I was to lay a bet on it, I'd say Australia will first implement heavy vehicle charging, because there are clear economic benefits in doing so, the political cost is not particularly high, and it is a significant first step to prove a system or systems can work, be user friendly and efficient.  Congestion charging in cities may be harder, because most of the cities don't lend themselves to compact easy to implement solutions, although the existing toll roads in many major cities could be used as a first step with peak pricing (which already happens on Sydney Harbour Bridge).

However, given the importance of the Greens in supporting the current government, I wouldn't be surprised if some incentives were created to promote congestion charging.  In addition, as it is a couple of years till the next election, it is easier to be brave this early in a Parliamentary term.

So the big question is, will Australians accept a shift in how vehicles are charged in the coming years?  At least the discussion is being had, and it appears the terms of reference for it are quite rational.

Thursday 28 July 2011

Tel Aviv to host distance based congestion charging trials

Traffic Technology and Diamond News both report that the Israeli highway operator, Ayalon Highways, had a tender on its website for a pilot programme to test congestion charging in Tel Aviv. The deadline for participation in the trial was 12 April.

The report says:

- The Israeli government is considering reducing the purchase tax on new vehicles and to replace it with user fees;
- Congestion fees would be “between NIS 0.10 and NIS1.00 (US$0.03-US$0.30) per kilometer of travel inside the city, based on traffic congestion”;
- A pilot programme would have 1200 vehicles, measuring travel by distance, time of day and location; and
- The pilot will involve the following incentive “A motor vehicle that does not travel to the Tel Aviv city center during rush hour and whose congestion charge does not exceed NIS100 ($30), will receive an NIS 250 ($75) credit. The fee, according to the pilot scheme, will not be paid to a driver who disconnects from the system during the trial period.”

Obviously, this sort of trial will be very interesting, and politically contentious. With the successful HOT lane already established from Tel Aviv to the airport, it certainly puts Israel at the leading edge of road pricing in the Middle East, in technical, economic, but most importantly in policy terms in delivering results for users, not just showing off technology.

If GPS based congestion charging proves technically viable and acceptable to users (the offer of credits is quite lucrative) in this trial, then Tel Aviv could be in the race to be the first city to ever implement time, distance, place based congestion charging (Singapore is the other key contender at present).

It is fairly obvious from a policy perspective that this is the most optimal solution for any city, but the key limitation will be ensuring all users are equipped, at a reasonable price. Before they are, any such system must have a backup based on ANPR to capture those who are not equipped. However, it is early days yet. I look forward to seeing the results of the Tel Aviv trial, and hope that it forms the basis for an implementation that can deliver positive results for network management and users.

RACQ supports discussion on congestion charging in Brisbane

A Queensland University Ph.D student’s study on the effects of introducing a congestion charge on Brisbane has provoked a cautiously positive response from a group that might otherwise be thought of as being opposed – the motorist advocacy group,  Royal Automobile Club of Queensland (RACQ), according to a report in the Brisbane Times.

Student Jake Whitehead modelled a Stockholm like cordon option for Brisbane that would have a cordon initially circling the CBD and Spring Hill, followed by a widening of the cordon to circle South Bank, Woolloongabba and West End by 2026. The charge proposed was A$3 (US$3.30) for entering between 0630 and 1830 Monday to Friday. It isn’t clear whether he meant an area charge or cordon, because a cordon would exempt anyone within it from a charge. Widening a cordon will increase congestion, but two cordons would address that.

Anyway, Whitehead claims A$250 million (US$275 million) could be raised per annum by 2026 (he is presuming it not be an offset to other taxes) and traffic would drop by up to 20%. He suggests the money be spent on a rail scheme, which is another issue (as motorists tend to want to see at least some of the money collected spent on roads).

The political response from Main Roads Minister Craig Wallace has been to say no to any congestion charge.  He obviously sees any positive discussion being political poison.

However, Paul Turner, General Manager of External Affairs at the RACQ is welcoming the debate, saying that tolls on existing roads within any cordon would have to be abolished to “make it work”. He welcomes a start of the debate saying:

"I don't think politically, or as a constituent base Brisbane people are ready for a congestion tax yet," …But it is something we really need to consider, because the fundamental ways of funding road and rail infrastructure are diminishing”

This is a very constructive attitude indeed. Whilst I disagree with Mr Whitehead’s concept for a congestion charge (it look a little like a duplication of Stockholm rather than a bespoke design developed specifically to address congestion in Brisbane), I applaud his efforts in raising the debate, and applaud the RACQ for wanting the debate.

My view is that talk of a Brisbane congestion charge needs to be in the context of considering road charging as a whole, and obviously concessions related to toll roads in Brisbane would be drastically affected. Meanwhile, the mere presence of toll roads in Brisbane is positive, in that it has enabled financing of new infrastructure and exposed people to paying directly for road use. In the longer term Paul Turner raises the real issue, which is the diminishing ability of fuel tax to be a sustainable way of charging for road use.

Could payroll tax be cut in exchange for New York traffic pricing?

Could congestion pricing for New York return because of a deal to use the idea not for new revenue, but to replace an existing tax. A Gotham Gazette report claims a group of New York state Republican senators want to drastically cut back the payroll tax that helps pay for public transport. This would make it acceptable for them to support tolls on the East River bridges as a replacement source of revenue, and to help ease congestion in Manhattan.

The payroll tax applies to New York City and the counties serviced by the New York MTA including Putnam, Westchester, Rockland, Orange, Suffolk, Dutchess and Nassau. The tax raises US$1.34 billion annually for the MTA. The tax is at US$0.33 cents on every $100 dollars paid by employers, but it is unpopular with many employers who say their employees and businesses do not benefit from the public transport services concerned.

Sen. John Bonacic has introduced a bill that would do away with the payroll tax in all counties except New York City. He would allow (but not demand) the New York City Council to set tolls on the East River Bridges: the Brooklyn Bridge, the Manhattan Bridge, the Williamsburg Bridge and the Ed Koch Queensboro Bridge of between US$2.50 and US$5.

Some concerns are:
- The toll might be too low, reducing the total amount of money available to the MTA; and
- The MTA needs new money, so it shouldn’t be about replacing an existing tax.

However, given Republicans opposed the Bloomberg congestion charge plan, this one which would have a similar effect, but as a tax replacement should have more traction. Then issues about raising money and how the MTA spends money can be addressed in parallel, whilst New York benefits from reduced congestion.

Why does San Diego want to buy a bankrupt toll road?

The South Bay Expressway has been a financial disaster in the world of private concession toll roads. It opened in 2007 as a new 10 mile stretch of tolled highway, but has never generated enough demand to be profitable enough to service its debts.  TollRoadsNews has a long list of reports on the project.

Where is it?

Beginning at SR-905 in Otay Mesa, the Expressway provides fast and convenient access to and from Mexico and communities across San Diego County.

It filed for bankruptcy in March 2010 and since then has been owned by the banks and federal government, which supplied the debt for the road. However, the San Diego Association of Governments is seeking to buy it. This is rather strange.

Why? Because the road is there now, working and providing service to users. It isn’t going to close down. The concession agreement hands it back to state ownership in 2042 regardless. In addition, SANDAG would presumably have to borrow money or raise taxes to buy it back. Unless it believes reducing or eliminating tolls on the road would create net economic benefits that outweigh the purchase, it seems like a very strange move indeed for a public body.

Another report quotes Gary Gallegos, Executive Director, saying the road is strategic and demand will be increasing – implying that it would be money making. He also said money to buy it back could come from an existing sales tax.

Bankruptcy court apparently valued the road at US$287 million, but it is apparently now generating an operating surplus. SANDAG thinks that as a result it wont be a bad investment, although I’d have thought it would not be a good investment if the return on capital is less than the cost of debt to buy it (or the interest gained from a bank deposit).  However, it is likely that the consortia of 10 banks that own 68% of the road will want out of it quickly, as will the US Department of Transportation which holds the other 32% because of loans it made to the previous owners.

Note: This was one dud investment for the Macquarie Group, which was at one time a 50% shareholder in the road.  It wrote off the entire value of the investment in 2009 according to the Bond Buyer.

French Ecotax truck toll appeal successful

It appears that France's distance based truck toll system is going to roll, once more.

Italian toll road operator, Atlantia, has produced a press release announcing that the French Council of State (appeal court) has overturned the decision by the lower court that had annulled the tender for the proposed "Ecotax" distance based toll for trucks to apply to government owned untolled motorways.

The Ecotax is intended to apply to all trucks over 3.5 tonnes in weight, on 12,000 km of highways and 2,000 of secondary roads, and would mean around 600,000 French trucks and 200,000 foreign trucks expected to have accounts.  The system was meant to be operational from 2013, but obviously now faces a delay of around 6 months.  Estimated revenue was around 1.2 billion Euro per annum, with charges based on distance at rates of between 8-14 Euro cents per km (US$0.11-US$0.20).

As a result, the Autostrade led consortium is the winning tenderer, as had been originally the case and confirmed the tender process was valid after previous claims of a conflict of interest between advisors RAPP Trans who were working for the government, but who had also undertaken work for Autostrade.  The contract is expected to be for 13 years and Autostrade will now be negotiating the final terms before signing the contract.

No doubt the French Government will be relieved.  The Ecotax is a new charge, so will be new revenue and the sooner it is operational the better for the government.

Of note is that the Ecotax will be a distance charge, with distances measured using GPS technology.  This will make it the third system based on GPS to charge for distance (after Germany and Slovakia), and the fifth system using GPS either to supplement another form of distance measurement (Switzerland) or as a voluntary option (New Zealand).   The objective of the French government is, in part, to rebalance demand between tolled and untolled highways, as there has been an issue of trucks choosing not to use the private and state owned toll roads in favour of the untolled parallel routes.

UPDATE:  More details on France's Ecotax here.

A replacement for fuel tax?

Eric Jaffe in the Infrastructurist discusses two options raised for replacing "the gas tax", which are in fact just one option that varies by scale.  He says:

In a recent article (PDF) in BNA’s Daily Report for Executives, Jack Schenendorf (an occasional contributor to National Journal’s Transportation blog) and Elizabeth Bell of the Covington & Burling law firm propose two alternatives to the gasoline tax: a ”Federal Interstate User Fee” for all interstate drivers, and a “Federal Motor Carrier User Fee” limited to commercial trucks.

The fuel tax wouldn't be replaced, certainly not initially, but supplemented by these fees. The proposals as Jaffe presents them are curious for their lack of detail and lack of consideration of consequences.

The technology proposed is essentially DSRC, as it it proposed that a tag and beacon system (EZPass being the one quoted) would be implemented, with charging at on and off ramps of all interstate highways.  It seems somewhat like the heavy vehicle charging systems already operational in Austria, the Czech Republic and Poland, which use overhead gantries to detect vehicles with tags (and also take number plate images).

Now just think about the scale involved in introducing such equipment on all of these across the United States.  It is technically feasible, but there are some major drawbacks:

1. Cost of installing gantries to do DSRC tolling on such a wide scale. Given the cost of a single gantry (which presumably has to have ANPR cameras to catch those with non functioning tags) could be anything from US$50,000 to US$150,000, the implementation costs on a nationwide scale are enormous, and could easily approach a billion dollars when communications and other infrastructure are included.

2. Diversion impacts of charging only on interstates. Obviously the answer to that is for states to be incentivised to charge their roads as well, to avoid diversion of traffic, but the negative short term impacts are considerable.  The obvious need is to charge across networks, and if states need to implement duplicate charging systems on their roads, it would easily be billions more.

3. If fuel tax remains, would the toll be high enough to make it worthwhile to collect?  Operating costs on a wide scale would inevitably decline, but a small increment of less than 2c/mile would almost be hardly worth it.

4. What of those not equipped for the toll?  The gantries would also need to include ANPR, so that number plates could be matched against those not paying through the tags.  As a result, there would be a huge exercise in chasing up unpaid bills across the country, and dealing with DMVs with poor quality vehicle ownership databases.

Certainly a transition from fuel tax is highly desirable, but a huge federal government run system would appear to be fraught with difficulties, particularly on such a scale.  It would be a nightmare to even do this for heavy vehicles alone.

Would it be better implemented by devolving power to collect revenue to the states, whether it be by fuel tax or a per mileage charge? Bear with me, because constitutionally what I am suggesting may not be possible, it’s just an idea.

At that point, states would be told that they would be responsible for collecting a set amount of revenue, based on statistics collected about actual traffic volumes on interstates. The states would be empowered to charge for distance, with a minimum amount to be paid to the federal government per mile (per type of vehicle even). If a state wanted to charge for its own roads then it could do as well, and of course it could contract out the collection and customer service aspects. If a state was uninterested in a distance tax, then it would still have to pay the money and collect it however it wished (e.g. fuel tax or annual registration fees). If the state refused to collect the money, then the federal government could contract another to implement it on federal highways.

In essence, I think the problem is that the network is too big and on too grand a scale to implement a single solution. I would strongly support a heavy vehicle only distance charge to be implemented nationwide, but initially as a voluntary option (with refunds of the fuel tax). With competitive service providers, it could be a way of managing risk (without being technology specific), but also provide a means to shift eventually to replacing fuel tax. Light vehicles would be the next, big step.

An alternative is for the states to pursue such charging and to piggyback off that, a reverse of what I described above.

In conclusion, what was suggested in the article is not ground breaking, but while barely thought out at all, it is in principle correct. A shift toward direct charging is desirable. Yet it is important for those who advocate that to not seek a silver bullet, for the transition away from fuel tax is going to be a long road.  The logical step would be to provide options for motorists to transition, to mandate an alternative for new vehicles, and to allow for diversity of options for payment and collection, as long as the right amount was collected.

Otherwise, it may be simpler to break up the Interstate highway network into a series of commercial companies, fund them with revenue from fuel tax now and empower them to introduce taxes or replacement means of paying for road use, then privatise them.   After all, isn't the USA meant to be the home of private enterprise, entrepreneurship and free markets?

UPDATE:  The article quoted at the top of this post is available here as a PDF.

Wednesday 27 July 2011

AECOM faces lawsuit over Clem7 traffic modelling forecasts

The Sydney Morning Herald reports that engineering consultancy and traffic modeller, AECOM, is facing a large class action lawsuit over its forecasts for the Clem 7 toll tunnel in Brisbane. This tunnel will be familiar to regular readers of this blog as I have written twice on it before.

Previously, the Brisbane Courier Mail reported AECOM claimed it had indemnity against the inaccuracy of its forecasts to be a liability of no more than A$500,000 (US$553,000), and it was paid A$1.5 million for the work (US$1.66 million).  Clearly investors think that this does not apply.

The suit involves a claim of A$700 million (US$774 million). The claim comes from a wide range of investors who lost heavily when demand on the road was around a third of forecasts. One concern is that AECOM did work on the traffic demand study for both the Product Disclosure Statement (the document intended to outline the project as an investment) and the Environmental Impact Statement (the document intended to obtain planning approval).

AECOM is denying the claim is valid and is responding by asserting that conditions have changed since the work was done in 2006, that the projects and traffic modelling for the two statements are different.

The Sydney Morning Herald continues:

In the PDS, AECOM estimated the average daily number of vehicles using the Clem7 tunnel would be more than 100,000 by 2011…

In a study done 18 months earlier for Brisbane City Council's environmental impact study, AECOM estimated traffic volumes would hit 57,000 a day by 2011, assuming a $3.30 toll.

The key concern being that if there had been transparency about those latter figures, investors would have been far more cautious and reluctant to invest.

IMF believes investors may have rights to recover their losses in RiverCity under the Corporations Act 2001, because it alleges that AECOM's statements in the PDS were misleading and omitted to provide investors with critical information relevant to their decision.

I am sure AECOM will defend this vigorously, as it means that the work of traffic modellers is, in fact, far more valuable than may typically be thought if the reliance on the work is made into a liability for the companies involved in this activity (companies that tend to concentrate on civil engineering consultancy, not investment bank level analysis). If AECOM succeeds, it will likely raise the issue as to whether investors will want a serious audit and due diligence done of such forecasts for future toll road. If the investors succeed, the premium on traffic forecasting will rise accordingly, as firms decide how the likely increase in corporate liability insurance required for such work affects their pricing.

Australian Professor suggests comprehensive road pricing reform

My attention was drawn to an article in Australian website Business Spectator by John Stanley, Adjunct Professor, Institute of Transport and Logistics Studies, Faculty of Economics and Business at University of Sydney.

He calls for radical road pricing reform for Australia saying:

My strong preference is for a broadly based road pricing reform agenda, which includes pricing for congestion and also for other identifiable costs of road use, such as road damage, various emissions and accident costs that are not adequately covered by existing arrangements. This could be done by a GPS-based pricing system that includes a variable road use charge to cover costs such as air pollution and accident costs that are not met by current charging/insurance arrangements and base road damage costs, levied through a vehicle kilometre charge.

In addition such a scheme would include tonne-kilometre (mass-distance-location) charging for additional road damage costs of heavy vehicles and congestion pricing by time and place.

I don’t agree that accident costs should be charged for, because that implies all motorists, those who do not have accidents included, pay for the costs imposed by those who do have accidents.  This is a matter for insurance premiums to reflect.  I also question whether the right approach to air pollution is to charge for it, when other activities are not also charged for creating air pollution (and when regulatory measures may be more effective).

However, he wisely notes that pricing shouldn't be introduced in isolation of offsetting complementary policies (the type that are essential for public acceptability):

As an offset, the reform should include abolition of existing excise and registration charges. Revenue should be hypothecated to a transport trust fund, with transport (road and public transport) and related environmental improvements being eligible for funding from this trust fund.

Reforming road pricing would also provide the opportunity to review public transport fares, many of which are currently artificially low to compensate for inadequate road pricing.

In other words a shift from vehicle ownership and fuel taxes to road pricing. He also notably recommends that public transport fares are too low because road pricing is inadequate, and that this should be corrected.

Beyond the detail, this is a good summary of the economic policy points around road pricing. It can better recover costs imposed by road use, and should be considered as a replacement for the legion of “second best” policies that have been introduced to tax fuel and vehicle ownership, and subsidise public transport as an alternative. It is a view that deserves more discussion and debate, and more forward thinking than simply seeing road pricing as a revenue raiser or as a way of suppressing traffic.

More GNSS road pricing trials, but is much really learnt?

The European GNSS Agency (GSA) essentially exists to promote the EU’s competition for GPS, Galileo, by encouraging applications of GNSS (Global Navigation Satellite System) technology more generally. It funded a research project to identify how the road transport sector could use the technology, understandably it ended up with road pricing.

Inside GNSS reports on a project, called GINA (GNSS for Innovative road Applications – yes EU agencies aren’t great at acronyms, just Google GSA!) involved another (there have been many examples of this) set of field tests for the use of GNSS technology to charge for road use. Given the technology is used already in Switzerland, Germany, Slovakia and New Zealand to actually charge vehicles, you might wonder how much more is to be learned.

The project started in March 2009 and ended this year, with full details available on its website. The prime driver behind the project is listed as:

GINA proposes to address the existing obstacles preventing a large scale take-off of road pricing and other services (VAS) based on the use of EGNOS/Galileo by:
• Proving the technical feasibility and economical viability of a large scale, GNSS-only road pricing scheme (alternatives to GNSS are being considered by authorities)
• Fully identifying the potential benefits related to congestion and pollution management
• Demonstrating the possibility of VAS running on a road pricing operating platform, as well as the use of the same on board equipment shared by different applications
• Understanding the added value of the EU technology in terms of performance or cost improvement compared to GPS-only.

There are extensive newsletters and the like on the website, so I wont go into detail. However, it appears not to have been driven so much by policy makers, but by GNSS technicians. Take this point:

A particular focus is on protecting users against the incorrect charging that can sometimes arise from large position errors and service interruptions resulting from the use of a single unaugmented GNSS system, such as the current GPS or GLONASS civil services, which lack a built-in integrity capability.

It implies that anyone implementing road pricing would simply use GPS without any other form of measurement of distance or correlation to location (such as map matching). It is simply designed to prove how augmenting GPS will improve accuracy, which is a fait accompli. 

The summary result is that it came up with what it claims is an inexpensive on board unit that doesn’t need an on board map to measure distance. By response to that is “so what”? There already is one operating in New Zealand manufactured by a company called ERoad that does just that, with GPS.

The REAL benefit that comes from any form of GNSS based distance charging is being able to distinguish location and time of day anywhere in a network, that includes not charging for being off road. It means that roads can be priced accurately, everywhere.

The report here suggests that the trials would allow this, only by creating “geofences” as wide cordons, so that vehicles crossing certain co-ordinate defined points, could be charged differently.

My bigger question is whether there really is anything new learnt from any of this? The study was commissioned by a body with a vested interest in promoting Galileo and EGNOS, all very well. Certainly the greater accuracy Galileo and EGNOS can bring will be welcomed, but the issues with rolling out GNSS road pricing are not mainly around accuracy of signals. They are around:
- Logistics and costs of installing OBUs on large fleets;
- Public acceptability of OBUs, especially around privacy;
- Managing the unequipped user efficiently and effectively.

However, GSA doesn’t exist to deal with those issues, it only wants to prove GPS isn’t good enough – when, in fact, it has proven that it is.

China to use tolls to do more than raise revenue

China's international radio service, China Radio International, reports a Chinese Ministry of Transport spokesman saying that China is considering building two road networks: the tolled expressways and untolled “common roads”. The expectation is that the government will pay for the latter, but tolls for the expressways. It is interesting that it noted tolls would not just be to pay for the roads, but to “regulate traffic flow” and “optimise modes”.

This implies some form of peak charging when congestion appear on the expressways, which is sensible. However with the statement “It's not reasonable to move large volumes of goods for long distances on roads” suggests the intention is to dis-incentivise heavy freight movement by long distance on roads. As much as China is using market mechanisms and private enterprise to build and manage expressways, it is still interested in central planning, presumably to maintain usage of railways for high volumes of freight.

Short items on the Philippines, El Paso and York

Philippines to build more toll roads

Agence-France Presse reports that Filipino President, Benigno Aquino wants the private sector to build three new major toll highways in the country. He listed them as:

- A link between North Luzon and South Luzon highways in Manila;
- Manila airport highway; and
- New access route to South Luzon tollway.

Of course the key problem faced by the Philippines is with congestion at toll booths near Manila.  Longer term it may want to relieve the congestion on parallel routes, but for now tolls are an effective way to help fund a major expansion of infrastructure.

El Paso to build new lanes as HOT lanes

The El Paso Times reports that El Paso County, Texas is intending to build HOT lanes parallel to the Border Highway.  It involves the construction of two new toll lanes, one in each direction, and an upgrade of the highway's four existing lanes, about 8.7 miles from U.S. 54 to Zaragoza Road.  The cost is estimated at US$80.2 million, with state bonds paying US$74 million and federal funds another US$6.2 million with the project expected to be completed in 2013.  No new land will be needed to build the extra lanes. What isn't clear is whether the revenue will be used to pay back the bonds, or be "free revenue" for the County, as it is claimed the "revenues raised here will stay in the region".   Nice deal for the County it would appear. The intention is that the lanes be fully electronic free flow and have peak and off peak charges.
A previous El Paso Times poll showed that most people did not plan to use the toll lanes, but motorists will have the option of using the express toll road or the non-toll part of the road.

It was also recommended that the toll lanes revert to non-toll lanes if the project does not pay for itself by 2028.  The proposed toll would be a base of US$0.10 per miles off peak US$0.13 per mile peak.

York rejects congestion charge

The Yorkshire Post reports that York Council has expressly rejected congestion charging for its Local Transport Plan. Executive Member for City Strategy, Steve Galloway said:

“The present council is very much opposed to congestion charging. It would impact on the city’s key economic drivers in the tourism and retail sectors. “If the Government does introduce a road pricing system across the country, then things could change. But as things stand, we do not want to be in competition with other cities with people perceiving York as somewhere that is more expensive to visit".

However, I don't think that the small congestion charge in Durham has had a negative effect. After all, York could well have a system that is targeted for its needs on a scale not much different from Durham. However, given the debacle of the last government’s mishandling of the issue of road pricing, it is hardly surprising it is seen as a poison pill by British councils outside London.

India's burgeoning toll road sector

Indian site Business Today has an excellent article summarising many of the ins and outs of the toll road sector in India.  Some of the key points being:
- The National Highway Authority of India (NHAI) estimates around 15% of revenue (US$68 million) per annum is lost due to pilferage at manual toll booths;
- The Ministry of Roads and Transport has proposed RFID (let's call it a form of DSRC) to resolve this, with rollout to START by May 2012 (why so slow? Is there an intention to have a nationwide interoperable network of clearing houses?)
- The rate of return for BOT PPP projects is estimated at around 14%;
- 80 PPPs have been commissioned in the past decade, with around 47,000km of additional road to be built under this model in the next few years;
- 170 toll plazas across India, of which 100 are government run by NHAI;
- An emerging market is the operation of tolling systems for BOT concessionaires (Egis being a partner in the first joint venture doing just that).

Clearly India is building new highways at a pace perhaps only rivalled by China, but the spread of tolling suggests that intercity road transport at least, is subject to the vagaries of pricing to recover infrastructure costs.  The biggest issue is, as it is in many countries, the issue of fraud and theft around manual toll booths.

Tuesday 26 July 2011

Sri Lanka's first toll road about to open, but already criticised

The first stage of Sri Lanka's Southern Expressway is due to open in the next week and will be the country's first toll road.

It runs from Pinnaduwa in Galle to Kottawa in Colombo district, with details on the Sri Lanka Road Development Authority (RDA) website here.  It is essentially the main highway south from outer Colombo to the southern tip of the country.  

Slow vehicles (tractor and rickshaws) will be banned on the 126 km expressway, which will include strict enforcement of speeds (100km/h) and overloading to ensure it is a safe route. 

The latest report (from Lanka Business Online) is that the tolls are to be collected manually, using tickets so that there is a ticket given at the start of the trip and used at the end to pay the toll.  It is noted that electronic tolling will be introduced later, which may reflect the low cost of manual tolls in the country, although the other point made by the article is that in India up to 20% of revenue is reportedly lost due to "leakage" (people not paying) with manual tolls.

Lanka Business Online suggests that the decision to introduce manual tolls was foolish, because it feeds the black economy.  It also notes the inevitable congestion with manual tolls saying:

A traditional staffed toll lane typically processes 300- 350 vehicles per hour whereas an electronic toll lane can handle from 1,000 vehicles per hour in a dedicated toll lane within a conventional toll plaza and even as high as 1,800 vehicles per hour in an open highway configuration.

The ticket system is damned as being slow and inefficient, when it would be preferable to have electronic tolls, or at least a flat rate charge to pay to speed up throughput.  

I wouldn't be surprised if the thought of electronic tolls, and the cost of equipment, and systems to detect offenders caused fear at the RDA.  However, there is little reason to not introduce barrier based electronic tolls as a starter, to accelerate throughput and provide a mechanism to detect and count throughput efficiently (to reduce leakage).   There are plenty of international experts who could have helped with this, and saved much money in the longer term.

Electronic free flow would require the number plate and vehicle databases to be fit for purpose, and I have no idea if they are or not.   However, that would be a future step.  For now, it would be nice if Sri Lanka did the basics right.  Maybe an audit of the manual toll system and business rules would be a start?

Clem7 toll road bankruptcy has not been disaster for Brisbane

The Clem 7 urban inner city toll tunnel in Brisbane was doomed from the start financially, not least because projects associated with it were not completed on time, and it was built on the back of grossly optimistic demand forecasts. It is an impressive piece of urban infrastructure, as I reported previously here, with 4.8 km of tunnel on a fully electronic free flow inner city highway. 

However, forecasts of 60,000 vehicles a day did not come to pass, with demand peaking at 27,000 a day at best. The A$3 billion (US$3.25 billion) project simply ran out of cash, as revenue couldn’t meet operating costs and debt repayments, so the owners – RiverCity Motorway group – went bankrupt owing A$1.3 billion (US$1.41 billion). One big problem for the road was that the Airport Link motorway, which would feed traffic from the airport onto the project, is not due to open until 2012. That is also a toll road project, and given RiverCity’s massive writedown of the value of Clem 7 from A$1.56 billion (US$1.69 billion) to A$258 million (US$280 million), Airport Link faces a similar risk.  RiverCity was publicly floated in 2006 at a price of A$1, but its final trading value was A$0.01 before going into administration. The risk with Airport Link is the Queensland government is a part investor in the company responsible for that A$5 billion (US$5.4 billion) project. According to the Sydney Morning Herald, ARUP is responsible for the demand forecasts for that road.

Hedge funds now own a third of the debt associated with the toll road, but the road still functions. The website is here with all the details about how to pay the toll, noting that it is operating under administration with the firm Korda Mentha. One report notes that once the Airport Link is opened, it is a chance to bail out of the road and sell it on to realise what little value is left. However, the big losses have been taken, and investors have been looking for those to blame for the overly optimistic demand forecasts. Engineering consultancy AECOM is being blamed because it was responsible for two sets of forecasts of demand, which reportedly gave different results to different client.

Lord Mayor of Brisbane, Campbell Newman is buoyant though as was reported by the Australian Broadcasting Corporation:

"My reaction is of course that I'm very sad for small investors who've lost money, but having said that, the fact remains that this project, this tunnel will be open for 100 years," he said.
"It'll deliver congestion-busting benefits for many generations of Brisbane motorists.
"It can't be taken away, it can't be shut down or closed, so it will be here delivering what it's meant to deliver in the future."

He's right of course, still, the bankruptcy is expected to reduce interest by investors in new Australian toll roads according to Queensland Treasurer Andrew Fraser, simply by the fact that it raises the risk profile (but should also raise the due diligence activity) around such investments. You see, paying for road use is not quite like paying for other activities!

Yet all of this vindicates the policy of putting this out to the private sector. The capital cost has not been born by taxpayers, but by private investors who, in this case, made a mistake. The risk has indeed been successfully transferred, and a bankrupt owner has still left a first class asset for motorists and the city.

New Zealand's toll road that probably shouldn't have been one

The Northern Gateway toll road north of Auckland is New Zealand’s first electronic free flow toll road, using only ANPR technology to detect vehicles. The road comprises an extension to Auckland’s Northern Motorway which is the main corridor between the city and its northern suburbs and the region of Northland. The extension is a bypass of seaside suburbs Orewa and Waiwera, saving time and relieving those suburbs of considerable congestion. So the economic (and environmental) benefits are considerable for those not even using it.  As with many toll roads, this is partly reflected in the fact the road is only 50% toll funded (and because if it was any more the road would be largely empty).

Northern Gateway is basically a bypass
The road is not a PPP.  The road and the tolling system is run entirely by the government agency responsible for the state highway network and the tolling system – the New Zealand Transport Agency (NZTA). The toll system has had a host of teething problems, largely related to human behaviour as people did not exactly act as was expected (in terms of how they pay or their willingness to establish accounts).   This is unsurprising, as NZTA has never had experience with free flow tolling so was hardly in a position to make such predictions.

Embarrassment over this have meant that the NZTA has taken a fairly relaxed attitude to non-compliance.

So it is not surprising that the NZ Herald reports that no violators have ever been referred to the courts, with unpaid tolls only being referred to debt collection agencies. The worst violator owes almost NZ$1050 in tolls (US$909). The reason for not forwarding to the courts is the cost of doing so, which does raise the wider issue around the incentives on NZTA to get this right.  Compliance and enforcement should be part of a comprehensive strategy, something that is commonplace on toll roads in other countries, but new to New Zealand.  A private concessionaire may have known better what to do, but the previous Labour government figured it could borrow money cheaper than any private concessionaire, and there were insufficient advantages in getting the private sector involved to offset the higher interest cost expected from private investors compared to the state.

For the road itself, non-compliance was at 3.7% in December 2010, which after two years of operation is a little high, but not catastrophically so. It has been on a downward path, but the level of non-compliance in part reflects the road being a rural intercity motorway, not an urban motorway, so the proportion of regular traffic is lower than in cities.

The Automobile Association claims one reason for non-compliance is that it is difficult to pay the toll, because roadside kiosks see “queues” as people stop to pay, and online and phone based options are not “user friendly”. Given that phone payment is only possible from 0830-1700 Monday to Friday, it’s hardly surprising. Likewise there is no easy SMS payment option like in some countries.

NZTA has since announced new “admin charges” for payment options, ranging from zero for account holders and online payment to NZ$3.80 (U$3.20) for phone payment. The latter suggests the system used for phone payment is remarkably labour intensive and well out of step with systems overseas (call centres should not cost that much and automation could reduce this).

The entire system is crying out for privatisation or to be opened up to competition from service providers who may do it more efficiently. The underlying problem is that the politics around toll roads clouded the finances.

The previous government passed legislation allowing for tolling of new roads, and the now-defunct highways authority (misleadingly called Transit New Zealand) was keen to be in charge of the first such road (beating any local authorities), so decided that the long delayed Northern Gateway project (or ALPURT B2 as it was then called) would be the first, because no other likely projects were in the pipeline. 

This was despite analysis showing it would not make much money if you took into account the costs of setting up the tolling system. So Transit convinced the then government that it was ok for NON-tolled road users to pay for the multi-million dollar tolling system through motoring taxes, to subsidise tolls. This was before there was any realistic chance of there being more than perhaps two other toll roads that could have been opened in the depreciated lifespan of the tolling system (which is about seven years).

In other words, the tolls were introduced in spite of the finances not really adding up, because the government wanted a toll road to show its legislation was not in vain. In truth, the Northern Gateway neither has the volumes of traffic nor the level of regular users to make tolling anything more than marginally useful to the finances of the project, even though its tolls are purportedly paying half of the cost of the project, the system to collect the tolls is 100% subsidised by other road users (and the road itself is only 50% toll funded).   

Of the possible policy reasons to introduce tolls, affirming a government policy is not a good one.  Neither is "making sure we are first and set the standards for toll roads" which was the reason for Transit New Zealand pushing tolls.  The proper reason is to accelerate a road project with new money paid for by users who value the benefits from using the road greater than what they pay through conventional motoring taxes.

Given that the likelihood is that there may only be two more such toll roads in the next ten years (only one has been funded), then some serious questions ought to be asked about reform, especially as the entire system will need replacing in around five or so years.  It needs to be investigated by those who have experience with such systems, and there are plenty in Australia.  It needs to not be shadowed by politics and concern by those in the NZTA who fear for their jobs, or fear the inevitable finger pointing about the project.   What is needed is some financial honesty and transparency about what went wrong, and how to proceed with the tolls (now that they are there) on this road and future ones.   The focus should be on maximising net revenue with significant improvements to customer service.

Dare I suggest this reform be done before the next toll road, the Tauranga Eastern Link, is completed?

Spain's gloomy economy affecting toll road demand

Euro Weekly News reports that the Alicante Ring Road is experiencing its lowest levels of traffic since it opened, and that other Alicante toll roads are at volumes of traffic not seen since 2000.

Last year, traffic on Alicante Province’s toll roads fell by five per cent against 2009 figures which had already dropped 15 per cent against the previous year. The drop is mainly on Alicante’s second ring road between El Campello and Elche, according to the Ministry of Development.

The ring road is the AP7, as seen below, providing a major bypass to the holiday city with has a normal population of over 770,000. 

View Larger Map

All of this doesn't bode well for any of Spain's private toll road concessionaires, which are facing severe challenges to their own viability as many are facing chewing up significant reserves whilst demand is well below forecasts (and what is necessary to service their debts).

Mozambique to push forward with private toll roads

Times Live South Africa reports that Nuno Mapossa of the Investment Promotion Centre in Mozambique has said that a toll road will be built between Mozambique’s capital Maputo and the KwaZulu-Natal northern coast. It would greatly enhance access which is currently available by road only on a 130km dirt route.

Chairman of the Mozambique Roads Fund, Elias Paulo, was reported by Macauhub as saying that guidelines for three new private concession toll roads will be prepared before the end of 2011. Furthermore:

According to Mozambican daily newspaper Notícias, the government plans to grant concessions on the Maputo-Maxixe section of National Road 1, as well as National Road 6 linking Beira to Machipanda, on the border with Zimbabwe and National Road 7, which links Vanduzi in Manica province, to Changara in Tete province.

Mozambique’s first private concession highway, EN4 from Maputo to the South African border, has been a success, with 35,000 vehicles a day compared to forecasts of only 13,000.  Mozambique has had a hypothecated roads fund since 1990, which has successfully ensured that taxes collected from road users (e.g. from fuel taxes) are dedicated to maintaining the road network.  This is inadequate to fund major improvements, which has encouraged the government to use tolling to allow major strategic highway improvements to be self funding over time.

Monday 25 July 2011

Weekend shorts: India, London and Transurban


In Thane District, Mumbai, India, there is a political problem with tolls. People don’t know what they are paid for. So, according to the Times of India, the State Excise Minister, Ganesh Naik, has decided that at every toll plaza electronic signs should be installed to tell people where the money is going. He accepted the suggestion by MP Eknath Shinde who said:

I suggest that toll operators must put up computerized display boards listing out the total cost of the project, the toll collected till date, and the number of vehicles passing through the toll plaza and their daily collection data

This is intended to give people confidence that the tolls aren’t being wasted, and I suggest provide an audit to ensure toll plazas are collecting what they should.

The report continued:

Thane district has a total of 14 toll plazas and residents of Thane, Navi Mumbai, Kalyan have to pay anywhere between Rs 30(US$0.68) to Rs 50 (US$1.13) towards toll every time they drive in or out of the city.

The Minister asked for an investigation into whether any toll plazas might be eliminated. Now if ever there was a country crying out for more up to date tolling, it would be India. Manual tolls always carry risks of corruption and theft, and there are relatively easy ways of addressing this.


Guardian journalist Dave Hill asked Transport for London for a breakdown of the estimated £55 million p.a. net loss in revenue from the closure at the end of 2010 of the Western extension to the London Congestion Charge zone. It replied that it would lose £65 million gross and save £10 million p.a. in operating costs. Of the gross revenue, one third is from enforcement revenue. Yes, about £21 million p.a. was made from the Western extension in fines through Penalty Charge Notices. In other words, the money is made from people not complying with the law. Now it doesn’t take too many people to make that much money. Given the fine is £60 if you pay within 14 days (£120 if you pay within 28 days), compared to what was an £8 charge, you need 9 times as many law-abiding users to make as much money as one violator (operating costs for pursuing violators are higher, and let’s assume quite a few don’t pay in time). Any urban road pricing scheme ought not to be THAT dependent on violation revenue, but it should be a gentle notice to those considering it. The real money can be made from enforcement, as the fines have to be high to deter people, and you need few violators to make a lot of money. I gather non-compliance with the London scheme is less than 5%, so less than 1 in 20 vehicles that should pay, don’t pay. That is about right, but from that 5% one can make over a third of revenue.

AND just to show you shouldn't believe everything you read, a website called Ars technica (!) claimed this:

"the government built an extensive network of surveillance cameras with license-plate-reading technology as a way to fight car bombings by the Irish Republican Army. The network proved useful a decade later when London introduced a congestion charge (now £10) for entry into central London. The cameras snap pictures of all vehicles entering and leaving the congestion zone and match their license plates against government records."

Except that it wasn't the same network. A bespoke network of cameras were introduced that would take two images - a context one, and an ANPR one - were introduced.

Then it makes another mistake saying:

We've already mentioned the London congestion tolling system, and Singapore has had a similar system, known as cordon tolling, in place since the 1970s.

No, Singapore had a paper based system called Area Licencing Scheme since the 1970s, and replaced it with a non-cordon system called Electronic Road Pricing, using quite different technology (involving prepaid smartcards) that applies varying prices on different roads at different times. Similar? Barely.

Then it concludes:

Unfortunately, the people building these systems don't seem to be making privacy a priority.

I guess if you don't bother looking at how Singapore DOES make privacy a priority (prepaid anonymous smartcards), or Eroads in New Zealand does (measuring distance only against a prepaid licence), or the Swiss system (which only measures distance), then you'll make this mistake. However, if you only talk to people in your own country, then it's hardly surprising you'll get a rather distorted view.


Australian toll road investor Transurban is a shining star in a world of toll road investors that have been burnt by the poor economic conditions in Spain, Portugal, Ireland and the US. According to The Australian, just under half of Transurban’s revenues come from one road – Melbourne Citylink. It isn’t difficult to understand why. Citylink includes the main highway from Melbourne’s main airport to the city (and there is no parallel rail link), as a result is a north-south bypass of the city, and an east-west one as well, providing connections from the south and east to the airport. It is in such a strategic position it is bound to do well, and its revenues grew 15.4% in the 6 months to December 2010. It bought Sydney’s Lane Cove Tunnel, which has tremendous potential as Sydney’s primary artery north. The company has been wise to not jump on more questionable assets, like Brisbane’s Clem 7 tunnel, but will have its eyes on roads with potential for growth. In that respect, I’d have thought it would have been disappointed that the British Government is not now selling the Dartford Crossing. Its share price is A$5.15 as of last quote, and it has been trading within a range of A$5.07-$A5.50 during the last 6 months.

Saturday 23 July 2011

Luxembourg city Mayor rules out congestion charging for now

Some of us working in the road pricing field are always looking to see if largely ignored countries or cities might be considering introducing some form of pricing.  The fact that Valletta, Malta and Znojmo, Czech Republic both have forms of urban road pricing is more often than not ignored.  So what about Luxembourg?  One of the smallest independent countries in Europe, surrounded by fellow Schengen area countries, it is difficult to determine exactly where you enter and exit the country, as there are no border controls or remnants of border controls.

In 2010, Luxembourg city was rated the 10th most congested city in Europe by satellite navigation provider TomTom. The key reason being that Luxembourg (the country) has the highest per capita car ownership in the world, yet a third of Luxembourg city households are car free.

In an interview with website 352LuxMag, Mayor Paul Helminger says that 40% of car trips in the city are for journeys of less than 2 km. The city’s strategy to reduce congestion has been to reduce car parks, reserve parking for residents and promote walking. Five park and ride facilities have also been built linking to rail and bus services, with two more to be built. Trams are being reintroduced to the city after an absence of fifty years.

However, when asked about congestion charging, the Mayor said no. He believed that if people simply were more aware that they need not use a car to get around, it would manage congestion “for a while”. In short, other measures have not yet been tried sufficiently to take such a politically bold step!

Florida moves forward to more electronic free flow tolling

Florida Turnpike’s Homestead extension went all electronic earlier this year according to the Miami Herald. 80% of users were already paying using the DRSC/tag and beacon “Sunpass” system (which is owned by the Florida Department of Transportation and usable on 600 miles of toll roads in the state). Those not equipped will have their number plates detected using ANPR technology, which when matched to vehicle ownership databases, will see a bill sent in the mail to the registered owner, with a US$2.50 surcharge. This is obviously intended to incentivise greater takeup of the Sunpass. The Florida's Turnpike Enterprise and the Miami-Dade Expressway Authority are seeking to convert all toll roads in southern Florida to be electronic free flow to reduce operating costs, congestion at toll booths and accidents from vehicles at toll plazas.

So far the following toll roads have gone electronic free flow:
- Homestead extension;

To convert in the next three years are:

What more can be said but to encourage Florida to continue joining the 21st century with modern tolling technology. What may be more interesting is seeing how easily they can enforce out of state vehicles that fail to pay tolls. That is perhaps one of the biggest risks for some states, which for south Florida is unlikely to be a significant issue.

California toll roads hit by decline in demand

The LA Times reports that usage of Orange County’s toll road/toll lane network has dropped, apparently due to people being less willing to pay for the time savings that the roads offer. Comparing 2007 to 2010, usage of the San Joaquin toll road declined by 19%, whereas usage of the Foothill and Eastern toll roads declined by 17%. The 91 Express Lanes also had a decline in usage, but has had usage partly recover as of late. The most recent attempt to change this has been a “raffle” with prizes of credit to toll accounts, to attract users.

Whilst people are being more prudent about whether to spend money on saving time, compared to other expenditure, it is also likely that the time savings offered have reduced, because with declining travel patterns, congestion on untolled roads will have declined making them more attractive. When the economy recovers, and employment picks up, congestion will increase, as will the disposable incomes of motorists, so it would be reasonable to expect some recovery in demand. However, as long as there are untolled alternatives, toll roads and lanes will always be seen as a “premium” product, which is easy to surrender when times are tight.

Israel gets windfall in privatisation of Cross Israel Highway

The Israeli government had indicated some time ago that it intended to embark on a programme of privatisation, including its 49% stake in Road 6 franchisee Derech Eretz Highways (1997) Ltd. The road, known variously as Road 6, the Cross Israel Highway or the Yitzhak Rabin Highway, is Israel's largest toll road, employing fully electronic free flow technology, with tolls measured according to the number of gantry points a vehicle passes.

Wikipedia describes the highway here, with the official website here in Hebrew.  It is 104 km long, started construction in 2002 and has a franchise through to 2029.

Derech Eretz's 51% private stake is held in equal proportions by Israeli companies Israel Infrastructure Fund and Shikun & Binui.  According to Israeli business website Globes, the government's stake had been valued at up to NIS550 million (US$162 million).

The price obtained is far in excess of that.  Globes now reports that Noy Infrastructure and Energy Fund has agreed to pay NIS1.4 billion (US$412 million) for the road.  It is the first investment from the fund which was established on 1 May. 

Globes reports further on the composition of Noy:

Noy Fund's partners in its winning bid for Road 6 include top investment institutions Psagot Investment House Ltd., Menorah Mivtachim Holdings Ltd. (TASE: MORA), Clal Insurance Enterprises Holdings Ltd. (TASE: CLIS), DS Apex Holdings Ltd. (TASE:DSAP), The Phoenix Holdings Ltd. (TASE: PHOE1;PHOE5), and Eliyahu Insurance Company. 

The other bidders oddly only wanted to buy part of the concession, making Noy the clear winner, and making it suddenly a key strategic investor in infrastructure.  

What must happen now is for the other two private investors to respond.  They must decide if they will match the offer of Noy to keep their respective proportions of shares.  They can do nothing, sell their stake to Noy Fund, or match it.  None of these will concern the government.  Indeed they will probably see the Noy bid as a boost in confidence in the future viability of the toll road.

Clearly, the view is that there is money to be made from this critical highway, which is Israel's major north-south artery bypassing Tel Aviv.  My question is whether Noy will be looking elsewhere to make further investments in toll roads.

Friday 22 July 2011

Tolls not yet finalised for South Africa's Gauteng network project

South Africa has one of the best managed highway networks in the developing world, and indeed can teach more than few developed country jurisdictions how to manage networks (yes that means cities and states in the USA, and local authorities in the UK among others). SANRAL (South African National Roads Agency Limited) is considered by the World Bank to be an example of international best practice in developing country highways management. As a limited liability company, it shows that a national highway network can be managed on a commercial basis, which would be considered almost revolutionary in the USA, if not in France, Italy and Japan where this is quite the norm.

One of South Africa's biggest highway improvement projects since the end of Apartheid has been the Gauteng Freeway Improvement Project. The project involves 185km of new and upgraded roads, including a ring bypass of Johannesburg, and tollways between Pretoria and Johannesburg and Pretoria and Johanneburg’s Oliver Tambo International Airport.  It is being funded through borrowing, with tolls to pay off the debt.  It is NOT a PPP, as the borrowing is being undertaken by SANRAL, as a state owned company, but is all being handled commercially.  In essence, many of the advantages of a PPP are being realised, because the government operates roads in a commercial manner.

The tolling system that is being introduced will use electronic tolling gantries (using DSRC/tag and beacon technology to the 5.8GHz EU standard). Customer service options will include being able to establish and pay accounts online, by phone, at roadside kiosk or at retail customer service centres. Tags are acquired for a deposit of 50 Rand (US$7.30), which is credited into the prepaid account.

The proposal has been controversial, because it is charging existing as well as new roads.  It is across a network, as can be seen on the map on this pageI wrote previously about the controversy around the scheme, as some motorists are upset that they must have a tag to get a discount, and some see barrier-free tolling as being easy to evade (probably around concerns that South Africa's vehicle licence plate database may have a significant number of inaccuracies regarding ownership details and addresses).

Today the big issue is the level of charges. It will be a form of distance charging, but by using DSRC and ANPR cameras, the distance is measured by passing various points on the network, not the constant measurement of distance as seen in GPS based systems.  This is akin to the heavy vehicle tolling system operated on Austrian motorways by ASFINAG (the Austrian government state owned enterprise responsible for the motorways).

The Wheels 24 website reported in February that the prices without tags were expected to be:
- Cars R0.66/km (about US$0.10)
- Heavy trucks R3.96/km (about US$0.58)

With tags the prices would be:
- Cars R0.495/km (about US$0.07)
- Heavy trucks R2.97/km (about US$0.43)

The price schedule includes a range of vehicle types. However, tags generally offer a 25% discount. There will be differential pricing by time of day, with discounts at night times. Buses may get a 50% discount (presuming they are registered legal scheduled services), and there are frequent user discounts (something unknown in the tolling world).  South Africa's Independent Online reported claims from SANRAL Chief Executive Nazir Alli that "if your travel costs are between R380 and R550, you would get a 15% discount and, if the value of your travel is greater than R900 at any given point in time, you qualify for 50% discount."  Such a discount may well be crucial in gaining grudging support.

A full map showing the detailed prices according to segment is located here, although it indicates that the Minister suspended the prices and has yet to announce new ones.

Revenues have been forecast at R300 million per month (US$44 million) intended to pay back loans of R20 billion (US$2.9 billion) to finance the upgrades.

Business Day reported that a dedicated police enforcement unit is being established to enforce the tolls, by identifying vehicles that are repeat offenders and having the power to stop vehicles and arrest the driver. Clearly the need to use force is seen to be necessary!  Given concerns about the risks of evasion, this is not surprising, if enforcing fines by post and court order is problematic.

However, the big question remains the toll rates. According to the South Africa National Roads Authority (SANRAL), the Minister of Transport created a Steering Committee to decide on the toll rates. The toll system was meant to commence operation in June, but it has been decided tolls will not be collected until all of the road upgrades are complete. Clearly the South African government thinks that the revenue from tolls is not as precious as ensuring political acceptability of toll levels.

It also appears that a significant body of opposition has appeared, last minute, to oppose it, with motorist lobbyists and trade unions opposing the tolls.  The Mail and Guardian Online quotes COSATU (Congress of South African Trade Unions) spokesman Patrick Craven claiming the tolls will generate "grotesque profits".   This claim seems utterly absurd, when the tolls will be dedicated to repaying debt incurred to upgrade the highway network.  SANRAL as a company is state owned, and wont be profiting anyone personally from the tolls.  FEDUSA (Federation of Unions of South Africa) agrees with COSATU and sees it as "privatisation" according to the Times in South Africa, both supporting a general strike in protest.

The South African AA, and "economist Mike Schussler" claim fuel taxes are better, presumably because it means everyone driving pays for the cost of upgrading a small part of the network.  None answer why this is so, or what happens as vehicles get increasingly fuel efficient (and the poorer tend to not be able to afford the newest most fuel efficient cars).   Eyewitness News reports how more members of the public are outraged at the tolls.  Fin 24 also reported how the AA believes it will result in alternative routes being gridlocked.

capita GDP (PPP basis) of just over US$10,000 a year, it means an average 17% cut in income.   The alternative is to go by the about to be opened, Gautrain, rapid commuter rail service (which is brand new and will operate at speeds of at least 160km/h (100mph)) between the cities, or competing bus and conventional rail services.

Surely what this all demonstrates is the need to lead with communications on the tolling right from the start.  It is always very difficult to toll existing roads, rather than simply new roads.  What South Africans face is new tolls on existing roads to pay for improvements to these roads, improvements that many clearly don't think are worth paying for.   It is fundamentally the risk of having fuel taxes and tolls on the same stretch of road, with no untolled alternative.

Ultimately, the South African government will almost certainly have to implement the tolls, probably at a greater discount initially, to get the necessary revenue, but what this should provoke in the country is a more fundamental debate about the future of road funding.   It means that the role of fuel tax vs. tolls be investigated, which should also include considering whether fuel tax may be better replaced with some form of wider network tolling system.   It is exactly the debate going on in the US, and in a different form in the Netherlands and Australia.  

Given how committed South Africa is to the Gauteng project, I suspect it will be implemented largely as planned (possibly with some short term compromise on price), but a new wider strategy is needed.  That strategy should be developed bottom up, be objectives led, take into account technological options, but more importantly the economics of road funding and financing over the longer term.  South Africa already has a world class model for road management, it now needs to develop a similar approach to road charging and more importantly, get motorists on board with a new deal - one that ensures they do not feel ripped off or cheated, where they see they get value for what they pay.  I don't think the country is far away from being able to do that, but right now the Minister of Transport must be wondering quite what to do - none of the options are going to be cheap, but all will be better than raising fuel tax.

UPDATE One writer to News 24 argues that the tolls are good value, and opponents are wrong to think someone is profiting heavily from money collected to pay off a substantial debt over more than ten years.