- Tolls and Public-Private Partnerships on California's Agenda
- House Bill a Mixed Bag for Tolling
- Building Electronic Toll Collection into Vehicles
- Getting Clear on Federal Transportation Subsidies
- U.K. Tackles "Optimism Bias"
- Speaking Engagements
As previewed in our last issue, Reason Foundation released our white paper on California's highway finance crisis on January 20th. The paper argued that things are not as hopeless as they seem, in preparing to keep people and goods moving as the state adds 16 million people over the next 25 years. Carefully targeted mega-projects could make a real difference, and although lack of funding has made them seem fanciful, the report builds a case that there is so much demand for better mobility that such projects could be funded largely or entirely with tolls.
In addition to citing toll-funded mega-projects of this type overseas (Paris, Birmingham, Toronto, Sydney, Melbourne), and showing how long-term public-private partnerships can shift major risks (cost-overruns, inadequate traffic) from taxpayers to investors, the report presents four detailed case studies of potential California projects. One is a $10 billion network of congestion-relief managed lanes for San Diego . Another is a $3 billion tunnel connecting Glendale with Palmdale and its would-be international airport. And the other two are $9-10 billion toll truckways, one each in the greater Los Angeles and greater San Francisco metro areas. The bottom line of the report is that California needs a state-of-the-art tolling and PPP enabling act, comparable in scope to those now in operation in Texas and Virginia.
We expected that a lot of attention would be paid to this report, and we weren't disappointed. Every major paper in the state did stories, and several wrote editorials in support. Every one of our stakeholder briefings around the state was sell-attended, including senior Caltrans people, people from the relevant metropolitan planning organization, and staffers from the governor's office. Speaking of whom, we are eagerly awaiting the imminent release of Gov. Schwarzenegger's "GoCalifornia" plan, which will reportedly include plans for tolling and public-private partnerships. We have also been contacted by several state legislators who want our advice on the provisions that should be in enabling legislation.
What's especially encouraging is that the legislative interest is bipartisan. To be sure, some on the left have taken the usual shots at tolling and privatization. But both Assembly Speaker Fabian Nunez and Senate Majority Leader Don Perata have been quoted as saying that tolling and public-private partnerships need to be considered, as California copes with its worst-ever transportation funding crisis. My Reason colleagues and I are guardedly optimistic. Needless to say, I will keep you posted.
By the way, our work on this project was made possible by several key supporters, whom I would like to thank. William E. Simon, Jr. is honorary chairman of Reason's Building for the Future Project and continues to speak out on these issues. AECOM and Granite Construction underwrote much of the research. And Nossaman, Guthner, Knox & Elliott provided in-kind legal research assistance.
The House Transportation & Infrastructure Committee has introduced its new bill to reauthorize the federal surface transportation program. Unfortunately, its tolling/pricing provisions leave much to be desired.
First, instead of mainstreaming the value pricing concept, allowing all states to use it where appropriate on Interstate highways (as the Administration's previous SAFETEA bill did), it actually shrinks the scope of the previous (and very successful) Value Pricing Pilot Program. Whereas the former allowed 15 states to become "project partners" and do any number of pricing projects, the new bill would allow a total of just 25 projects nationwide, including those already in operation. This would be a real step backwards.
Second, there is still no provision for private activity bonds, which were included in last year's Senate bill and are strongly supported by the Administration. The idea here is to end the long-standing tax-code discrimination against private developer/operators of toll roads, under which they can issue only taxable toll revenue bonds while government agencies can finance comparable projects at tax-exempt rates. This biases decisions against the very kind of long-term public-private partnerships that effectively shift risk from taxpayers to investors. (At least the bill is consistent; in the sections that permit up to six pilot projects to build or rebuild Interstates with tolls, the state must "give preference to a public toll agency with demonstrated capability.")
Third, every place the bill discusses tolls or pricing, it mandates that low-income people be charged lower rates than everyone else. This would be a first of its kind in federal transportation policy. Amtrak does not give low-income discounts, nor do federally subsidized mass transit agencies. Nor do the airlines participating in the federal subsidy program for rural communities. Why single out toll projects for this kind of social engineering?
Fourth, although as in last year's bill there is a proposed pilot program for dedicated truck lanes, the language of this section makes no mention of tolls, and no mention of relaxing federal limits on size and weight that would make such a program attractive to the trucking industry. (Many would gladly pay tolls if it meant carrying much higher payloads.)
Instead, the bill confronts the trucking industry with even more of what it fought so hard against last year: pilot programs to build and rebuild Interstates with tolls. In addition to the previous pilot allowing up to three states to rebuild an Interstate with tolls, it would add a second pilot program allowing three new Interstates to be built with tolls. While I'm glad to see greater use of tolls, given the overall shortfall in highway funding, this approach needlessly sets up further battles with trucking, when a value-added toll truck lane program would be likely to gain their cooperation, instead.
Finally, on the subject of HOT lanes, the bill is a mixed bag, like its predecessor. It does allow HOV lanes to be converted to HOT, but now would require variable pricing (a good idea, but does it have to be a mandate?). It also requires that the performance of such HOV facilities not be "seriously degraded" by selling space to paying customers. It's hard to tell what that means, but it might make it difficult or impossible for local entities that want to change the definition of "high occupancy" from two to three, or from three to six or eight (van-pools, say) to do so. Why can't the definition of what is high occupancy be left strictly to the community in question?
On the other hand, the language on allowing hybrids (low-emission and energy-efficient vehicles) into HOV lanes has been changed from last year. Now such vehicles may be allowed in if they are charged a toll (albeit a discounted toll). So whereas the previous provision risked flooding HOV lanes with hybrids and thereby congesting them, the new provision backs off from that and appears to permit these kinds of vehicles only if the lane is converted to a HOT lane. That does seem to be an improvement—and it would mean that the recent raft of state laws to allow hybrids into HOV lanes could not legally go into effect unless the lanes were converted.
Right now, road pricing is only practical on limited-access roadways like Interstates and freeways. That's because today's electronic toll collection technology requires significant investment in gantries at tolling points, and a limited-access system means you have only a limited number of access points to equip with such gantries. But what if every vehicle came equipped with a GPS device (so that its location was known precisely), an on-board computer (to keep track of where it was driven), and a two-way communications capability (so it could download that information periodically)? That would make it relatively easy to charge directly for the use of any roadway, anywhere in the country. The potential feasibility of such universal road-charging systems is under study by the Oregon DOT and by a multi-state project based at the University of Iowa.
But how likely is it that cars will someday be for sale with all that gadgetry built-in? The trail toward that goal is being blazed by a year-old effort called the Vehicle Infrastructure Integration (VII) Initiative. It's a joint effort of the Federal Highway Administration, 10 state DOTs, most of the major auto manufacturers, and the organizations representing local government agencies. They comprise the VII Coalition, which had its first public event earlier this month in San Francisco. You can find all the presentations at www.itsa.org/vii_meeting.html.
The VII effort is being driven by two major concerns: safety and congestion. On the former, the idea seems to be that better technology at signalized intersections (for example) can prevent many crashes. And it�s pretty clear that, in addition to facilitating value pricing to address congestion, better information and communications could help highway managers respond faster and more effectively to traffic-snarling accidents.
So VII envisions three major components that need to be developed and installed:
- The in-vehicle components, mentioned above;
- Roadside infrastructure that communicates with vehicles, and;
- A network that links everyone, including car manufacturers (like GM's Onstar), transportation managers, and even commercial service providers (like, where is the nearest Starbucks?).
So far, the Federal Communications Commission has allocated spectrum in the 5.9 GHz band for these purposes, and the VII effort has defined a basic architecture for the three components. The next step is "deployment analysis," figuring out where it would make sense to put various pieces, what it would cost, and who would pay. There are six task forces at work, including one on policy issues such as privacy, obviously a major issue.
These are early days for this highly ambitious effort. Everyone interested in road pricing should be paying attention and, where feasible, getting involved.
The Bush Administration's proposal to eliminate operating subsidies for Amtrak has rekindled debate on how much subsidy goes to each mode. I was on an hour-long NPR affiliate radio talk show on Amtrak last week, and was appalled to hear the director of a transportation think tank claim that highways and airlines get more federal subsidies than Amtrak. This was either unconscionable ignorance or deliberate deception.
What many people do in such debates is a semantic sleight-of-hand. They equate the number of federal dollars spent with the amount of federal subsidy. What this ignores is that the users of airlines and highways pay user taxes that are restricted by law to paying for the infrastructure those transportation modes use. Amtrak users pay no such user taxes. The amount of federal subsidy is the difference between what a mode brings into the federal treasury in user fees/taxes and what the feds spend on that mode.
Fortunately, the US DOT's Bureau of Transportation Statistics has come to the rescue, with its December 2004 report, "Federal Subsidies to Passenger Transportation." This very useful report compiles data for highway, air, transit and rail (Amtrak) modes, from 1990 through 2003. Highway is broken down into four categories of user: autos and other light vehicles, school buses, transit buses, and intercity buses. Air has two categories: commercial (airline) aviation and general aviation. As the report defines its terms, "Net federal subsidy is estimated as federal outlays minus federal receipts from transportation taxes and user fees." The tables include both the overall numbers for each year and, to facilitate comparison, normalized figures in which the total net subsidy in a given year is divided by the number of passenger miles for that mode in that year.
The results should put to rest spurious claims that highways are more subsidized than Amtrak. On the basis of net federal subsidy per thousand passenger miles, Amtrak finished first, at $186 per thousand, over the 1990-2002 period. In second place was transit, at $118 per thousand. For highways overall, the comparable figure was minus $2—i.e, highway users paid in more than they got back. And in aviation, the airline subsidy was $6 per thousand, while general aviation�s was $90 per thousand.
And in case you are wondering what might be obscured by the normalization by passenger miles, let's also look at the total dollar amounts of net federal subsidy. By this metric, urban transit got the most, averaging $5.1 billion a year (in 2000 dollars) during this time period. Airlines were second, averaging $1.9 billion a year (mostly for FAA safety-related costs covered by general fund appropriations). Amtrak was in third place, averaging just over $1 billion per year over this 13-year period. And over this same period, the highway system was a net provider of federal funds to the tune of $7.4 billion per year (though that�s been trending downward since it peaked at $11.7 billion in 1998).
In their outstanding 2003 book Megaprojects and Risk, Danish researcher Bent Flyvbjerg and colleagues documented the miserable track record of large-scale transportation infrastructure projects. Again and again, they come in way over budget and frequently attract significantly less traffic than forecast. Flyvbjerg explained the flawed incentive structure of conventional governmental procurement models, which makes it in the interest of project promoters (both private and governmental) to minimize projected costs and maximize projected traffic, in order to get to a decision to build the project.
Recently, the UK Department for Transport contracted with Flyvbjerg to develop ways to counter this "optimism bias" in planning large projects. The report was published last year as "Procedures for Dealing with Optimism Bias in Transport Planning."
The basic idea is to counter optimism bias in project cost estimates by applying an upward-adjustment factor to such estimates, at the time a decision is made to build the project. Based on empirical data, the �uplifts� (to have a 50% probability of achieving the cost target) are 15% for highway projects, 40% for rail projects, 23% for bridges and tunnels, etc. Higher uplifts are required if a higher probability of meeting the target is desired.
The problem with this approach is that project developers who know that an uplift factor will be applied to their cost estimate may further reduce the cost estimate they submit, in hopes of being selected and being able to make up for the under-estimate later on, as they do today. The report recognizes this possibility, and says that the use of uplifts should be supported by things like requiring local co-funding of cost over-runs where possible, formal requirements for cost- and risk-assessment at the business-case stage of the project, and introducing independent appraisals.
While those measures may help, it would be far more effective to genuinely change the incentives of the players. If the developer is also to be the owner/operator of the project, and must pay for it largely or entirely out of fees paid by its users, then the risk of cost overruns is no longer the government's problem. And the developer/operator's interest is to get the project built on time (so it can start generating revenues) and on-budget (so that those revenues will be enough to repay the investment). That's the long-term public-private partnership model that seems to be working well for European and Australian toll projects. The U.K. has used this model for just one project: the M6Toll motorway bypassing Birmingham . It should use if far more widely.
I'm on the road again this month, and again in April. Here's what's on my speaking calendar:
Feb. 24: Breakfast talk on California highway finance, in Oakland, sponsored by BR&H (RSVP to Sheila Garvey, email@example.com).
Feb. 25: Panel at West Coast Corridor Coalition/Cascadia Leadership Forum in Redmond, WA (www.cascadiaproject.org).
April 13: Managed Lanes presentation at Team Florida meeting in Tallahassee.
April 18-19: Managed Lanes/BRT presentation at TRB 12th International HOV Systems, Pricing, and Managed Lanes conference, Houston.
April 26: highway finance discussion at Construction Industry Round Table conference, Washington, DC.
"Californians can't get from place to place on little fairy wings. This is a car-centered state. We need roads. Like Gov. Pat Brown before me, I intend to see that the government builds the roads that Californians need." —Gov. Arnold Schwarzenegger, State of the State speech, January 5, 2005.
"Nothing could be more natural than for the transport modes that best meet the demand to develop more than the others, and for the latter to dwindle. This is true in all sectors of the economy, and it is difficult to see why transport should be the exception to this rule." —Christian Gerondeau, in Railways, Public Transport, and Others: the Republic's Wastes.