Oregon set on slow move to statewide GPS tolls

An Oregon state agency has proposed a 20-year-plus transition to mileage-based tolls as the principal means of financing and managing the state's roads as fuel taxes decline. The Road User Fee Task Force (RUFTF pron 'rufftuff') was appointed by the legislature in Nov 2001 and given eight years to design and implement a new revenue collection system to take the place of fuel taxes. RUFTF consists of four legislators, two transport commissioners, a mayor, a judge, an academic,a businessman, a thinktanker and a consultant. It is managed by a prominent Oregon power broker James M. Whitty.

Its major conclusion on technology was that a GPS receiver and a digital odometer are the best combination of technologies for recording road use, and a short range (DSRC) transponder is the best means of uploading trip data for toll collection. It leaves open the details of toll collection which would mostly occur on the roads via readers at fuel (gas) stations, though there would also be a number of readers and enforcement equipment on roadway gantries.

In its first report to the state assembly RUFTF suggested a transition period in which the fuel tax and tolls would operate in parallel. Tolls would be collected at first at a flat rate of 0.8c/km (1.25c/mile). Later the system might be adapted to charge different toll rates by time of day and congestion - "congestion pricing." A policy might be adopted of tolling new roadway facilities such as extra lanes or new roads or bridges, and the system could be adapted to that.

"Any new roads, bridges or extended lanes should be paid for, at least in part, through tolling when practicable. Tolling will make newer, and often sorely needed, large projects more likely to be built and likely on shorter timeframes."

If Oregon proceeds with these plans it could be the first US state to follow the path being laid down by last year's 15-US state DOT/FHWA report "A New Approach to Assessing Road Use Charges."

Germany is presently in the final stages of implementing the world's largest toll system for trucks over 12t based on vehicle positioning equipment using in-vehicle units that will synthesize data from GPS, digital odometers (or tachographs as they call them) plus DSRC roadside equipment, cameras and tickets. Switzerland has a system in operation for trucks using GPS and other vehicle positioning systems. The UK is planning the world's most elaborate nationwide tolling, for implementation in 2006.

The Oregon RUFTF looked at a six year phase-in. It estimated that a GPS-transponder based system would cost about $1 billion to set up for the whole vehicle fleet of the state. An odometer-transponder system would cost in the range $300m to $350m. The 6-year conversion to GPS would require a two-thirds increase in charges to fund the scheme and the 6-year odometer-based system would require a one-fifth increase. This was too expensive, the task force concluded. [ Administrative costs of collecting the fuel tax, by contrast, are put at only $1m or 0.25% of revenue raised.]

The task force settled on a more economical but much slower transition - applying the VMT toll to new vehicles only. All new vehicles sold in the state would be required to incorporate the toll system. But based on the normal turnover of vehicles it would be about 20 years before 90% of vehicles in the state were equipped. Indeed even then a fuel tax would be continued for out-of-state vehicles until some nationwide interstate compact is adopted to manage mileage tolls by out-of-state vehicles.

"The task force recommends application of a mileage fee only to new or newly registered vehicles. This will require a mandate for the new vehicles to enter Oregon properly equipped with the necessary technology and for the newly registered vehicles to be retrofitted with the necessary technology.... The task force accepts a phase-in period for a mileage fee, possibly 20 years, as a practical necessity."

Costing of collection methods by consultants suggested a centralized system could be quite expensive - $49m/year - whereas a system of using fuel stations as the primary points for vehicles to upload toll data would be much cheaper. Use of mobile telephone sideband as being used in Europe is reported as expensive also - costing $55m/year.

"Centralized data and fee collection, whether private or public, may be cost prohibitive unless ways are found to reduce costs to acceptable levels. For this reason, the task force will continue to investigate ways to reduce the administrative cost of collection centers while focusing more intensively on data and fee collection at service stations."

Toll collection is recommended to be contracted out to private operators.

On privacy issues the report says: "To provide technological safeguards to Oregonians, the task force recommends only the transmission of summary mileage data and not transmission of continuous vehicle locations. The task force also recommends transmission via radio frequency intermittently over only a few feet to local readers. The task force rejects transmission of data via cellular technology over a wider area. The task force also recommends legal safeguards be built in to any GPS-based mileage fee to prevent anyone other than the vehicle owner/operator from knowing the vehicle's movements without the consent of the vehicle owner/operator."

Demise of fuel tax the premise

The new road use fee scheme is based on the belief that new vehicle technologies will "dramatically improve the average fuel efficiency" of the state vehicle fleet seriously undermining the yield of fuel taxes. It cites hybrid-electric vehicles with the ability to get 1.5 to 2.5 times the fuel efficiency - 40 to 70 miles per gallon (16 to 28km/l.) Additionally the increased use of lighter and stronger materials such as composites will further erode fuel usage and tax revenues. Further on fuel cells will take fuel economy to new levels. It then claims the US will run into scarcity of oil and higher underlying costs will encourage the substitution of lower or untaxed fuels for taxed gasoline and diesel.

"Oregon fuel tax revenues from the sale of gasoline are likely to level off and then permanently drop during the next ten years." (p14)

The past several decades have seen a four-fold increase in the intensity of use of the state highway system from about 0.25m vehicle-miles per lane mile to over 1 million. Miles driven per year is similar in urban and rural areas.

The group found no technological impediment to a mileage-based toll with multiple technologies available including satellite positioning (GPS), transponder systems, wide area wireless, and license plate recognition with cameras. This should be designed to implement a user-pays principle. The system must be capable of being used to support the whole road system and multiple road operating agencies.

The group proposed consideration of six approaches:

(1) Transponders used to transmit mileage data derived from in-vehicle GPS and/or odometer. Data transmitted by dedicated short range communications (DSRC) then by landline to billing center.

(2) Gas stations to get mileage data and toll due since last fueling. Toll added to fuel bill and fuel tax credited.

(3) Gas station pumps with choice. Switch will enable motorist to choose to pay fuel tax along with fuel price, or download toll due data to pump.

(4) Mileage data calculated according to fuel used and miles-per-gallon rating for vehicle type.

(5) Department of Motor Vehicles to receive uploads of mileage traveled and toll due and levies against the motorist after gas tax credit.

(6) Statewide network of DSRC toll points with enforcement cameras and issue of transponders statewide.

The first or preliminary report of the task force said the most functional combination is GPS and a DSRC transponder with a network of over-the-road readers. It rejected reliance on odometers alone because they are unable to distinguish high value/higher toll rate trips from lower rated trips, but the final report this year seems to retreat from emphasizing the need for GPS.

Equity

One comparison could weigh politically against the toll versus the tax. That is a comparison of payments by different kinds of cars (Appendix CC). It looks at seven vehicle makes and the comparative payments for 250 miles buying gasoline at $1.50/gal of which the OR tax is 24c compared with a VMT fee of 1.25c/mi. The Ford Taurus and the Dodge Caravan as your average cars are pretty much a wash. The Taurus will pay $16.82 in VMT tolls versus $16.30 in fuel tax. The Caravan pays $19.70 tolls vs $19.74 taxes. The political problem is that the environmentally virtuous vehicles pay a lot more in the VMT-toll than in taxes - The Toyota Corolla $12.39 vs $11.03 and the hybrid Prius $9.69 vs $7.81. At the same time the Lincoln Navigator SUVpays only $27.36 tolls vs $28.85 taxes, and the Lamborghini gas guzzler (11mpg) $31.76 tolls vs $34.09 taxes.

Now it can be argued that the road use charge is not the place to reward environmental virtue... but this is a political problem for VMT tolls.

The earlier report by RUFTF suggested as a remedy a fuel-efficiency adjustment to tolls to provide a discount on the base rates. It also suggests that congestion pricing can be implemented as adjustments to base toll rates. Under the heading "geography" – "politics" would have been a better heading – it suggested motorists with addresses in certain areas could be given adjustments on the base rate tolls. We wonder if any of the geographic "adjustments" would be up?

In the later report the group rejects all but congestion pricing adjustments saying rates should not be complicated further.

Allocation of funds

The group recommended that congestion pricing funds should be used for road modernization within the tolled area. Where peakhour tolls are imposed on a particular facility it suggested they should be devoted to transport modernization of that corridor including other parallel routes within the corridor. Congestion pricing would have to wait for the "20 years plus" of high penetration of GPS positioning units, the report says.

It is unclear why transponders or a call-in booking system with camera enforcement like London could not suffice in Oregon to support congestion pricing. The task force does not seem to have thought this through. Call-in booking plus cameras will be necessary even when 100 percent of Oregon vehicles have GPS units - to cope with vehicles with defective or disabled GPS units, and with Californians, Washingtonians, and other visitors. So why not use the cameras for tolling the Oregonians without GPS units during the period when GPS is only partially in use?

Pilot program

The latest report lays out a pilot program to support adoption of a

mileage fee with a time-of-day price adjustment option. The first step will

involve small-scale testing of what the report calls "an electronic odometer, either global positioning system or odometer tag for mileage data collection; radio frequency technology for summary mileage data transmission; and related technology to support a mileage fee." Got that? The second step will involve using the same technology for "large-scale testing of the behavioral elements of a time-of-day component to a mileage fee (congestion pricing) as well as the mileage fee." HB 3946 requires ODOT to commence the pilot test program no later than July 1, 2003. The pilot program will operate for about three years.

"GPS systems Big Brother"

    Accounts of GPS (global positioning system) often describe it as a tracking system implying that it can track vehicles with GPS devices. Wrong. It doesn't track anything. It allows people with GPS receivers to establish their own position.(see http://www.colorado.edu/geography/gcraft/notes/gps/gps_f.html)

CNS News quoted Chris Edwards of the Cato Institute as saying: "I don't think Americans are ready to be subjected to that type of civil liberties intrusion, where government tracks them around wherever they drive. I think it's nutty." (CNSnews20030102)

Nutty is Edwards' misunderstanding of GPS.

The Associated Press in its account of the Oregon RUFTF proposal said the in-vehicle GPS unit works by "bouncing signals off satellites." (AP20021231) More nuttiness. GPS units are purely receivers. They receive radio data signals being broadcast continuously by orbiting Navstar satellites. Managed by an atomic clock each satellite sends out a stream data bits each tagged to the nearest milli-second. The GPS receiver has its own clock and by comparing the time each signal was transmitted with its arrival time the GPS unit in the vehicle can estimate the travel time of the signal, which allows it to calculate its distance from each of the satellites in view - usually about four satellites. Each satellite can describe a circle of potential positions for the receiver and with three satellites calculating circles in space the GPS receiver is able to estimate estimate its position at a point of intersection of three circles - a kind of triangulation. A fourth satellite signal can be used to synchronize the clock of the receiver with the satellite clock system. No one is being tracked.

Now of course to be of any use for tolling the position data gets time-dated and stored in an in-vehicle computer, and routes and distance traveled can be estimated by reference to a digital map, and a record of travel generated. No one outside the vehicle can gain information about the travel of that vehicle unless they can get to the computer or until the data is uploaded. In principle if there is a "Big Brother" concern the travel data could be applied to lookup tables of toll rates stored in the computer and then the uploads would be merely $-owing. In practice as with transponder accounts most people want the tollster to have records of trips made in order to be able to contest mistaken billings. But if customers want to sacrifice that element of customer service their travel can be anonymous with GPS.

(SEMANTICS: the Oregon report confusingly uses the term AVI when it clearly means a DSRC or short range wireless transponder system. Automatic Vehicle Identification or AVI is a function not a technology. The function of AVI can be achieved by cameras, DSRC transponder, laser communications, barcode reading, and other technologies. Most DSRC transponders actually identify a pre-arranged toll account, not a vehicle. They are provided with a velcro attachment so they are intended to be moved between different vehicles on the one account. They are performing an Automatic Account Identification function. Any device intended for vehicle identification must be physically attached to that vehicle, as proposed for the next generation of DSRC at 5.9GHz.) (contact Jim.Whitty@odot.state.or.us see www.odot.state.or.us/ruftf/)

BACKGROUND: Oregon enacted the first fuel tax in the US in 1919. In recent years voters have repeatedly rejected gas tax increases in ballots. Instead the legislature increased the gas tax from 6c in 1980 to 23c in 1990. The yield rose far less because of increases in fuel economy, and it took a heavy political toll. Throughout the 1990s the fuel tax rate was frozen. Oregon raises about $620m in road revenues, split between fuel taxes $390m (62%), a weight-distance tax on trucks $180m (29%) and registration and title fees $50m (9%). It gets some $350m in federal grants. State law requires the yield of the weight-distance tax to be maintained in proportion to the fuel taxes. The fuel tax paid per vehicle-mile traveled is about 1.25c whereas in 1960 it had been over 3c in 2002-$s. TRnews 2003-05-28