National Road Pricing Conference

Proceedings
Friday, June 4, 2010
SESSION 5
The Future of Pricing and Potential New Applications

Moderator: Bob Poole, Reason Foundation

Federal Perspective on Future Road Pricing
Wayne Berman, Federal Highway Administration

I would like to give the federal view on the future of road pricing. It is hard to look at the future without looking at where we are today and the lessons learned before determining our future. SAFETEA-LU has provided legal authority for the federal sector to provide tolling and pricing on highways. We have financed construction, reconstruction, efficiently reduced traffic congestion, and improved air quality. These six programs are available:

• Express lanes demonstration program (managed by myself),
• High performance vehicle facilities program (Jesse Yung),
• Value pricing pilot program (Angela Jacobs),
• Interstate system construction toll pilot program (Greg Wolf),
• New interstate construction co-pilot program and interstate reconstruction co-pilot program (Greg Wolf); and,
• Section 129 toll agreements (Greg Wolf).

Programs 4-6 are all managed by Greg Wolf from our Office of Infrastructure. Patrick DeCorla-Souza leads the tolling and pricing team which heads all of these programs. Patrick is the go-to-guy to express interest in tolling and pricing. We try to organize ourselves to support state and local efforts to advance pricing.

The types of priced lanes we have discussed at this conference include HOT lanes (with HOV 2, HOV 3 lanes) and express lanes. The different between the two is a very minor nuance: where-in the HOT lanes HOVs go free and in the express lanes the HOVs could be charged a fee. Whether HOVs are charged or not depends on the program and the nature of the project.

We have a number of operating HOT 2 lanes around and another coming on line soon on I-110 in Los Angeles. We toured Houston's HOT lane yesterday. There is the new I-95 express lane program in Miami that went from and HOV 2 to an HOV 3 which was part of the urban partnership program. Another number of larger projects are in the process of being built: the I-495 HOT lane in north Virginia, another partnership in Atlanta, the one in Los Angeles on I-10 and the extension of I-95 are all urban partnership projects. On the express lane side there is SR-91, I-95 in Baltimore which is under construction now, the I-595 project and a number of others all supported in our express lane demonstration program.

Priced lanes and pricing have come out of the closet and it is no longer a discussion among academics. Pricing is here, it is being implemented and it is real. There are toll lanes and express lanes that are operating and lanes that are under construction. Most of the experiences are with the conversion of HOV facilities to HOT facilities. These transitions allow a shift from congested lanes to tolled lanes but limits the degree of shifting by pricing. These projects preserve or incentivize transit and carpool use. These congestion pricing projects have been valuable because the public has seen them developed and seen them work. Now the public is more accepting of them and willing to develop them. The projects have:

• shown that reliable travel times are possible;
• shown priced lanes enhance public transportation;
• shown the public that there really is a choice and everyone can use the priced facilities;
• demonstrated technical feasibility;
• shown a change in travel behavior;
• proven that travelers are happy to have the option to buy into a reliable trip. When they see advantages, they are happy to support it and use it.

What have we learned is that these projects affect behavior, volume, and speeds. There was success with the I-15 project in San Diego that increased usage. The SR-91 express lane is a good model that shows these projects work.

Transit ridership is a key piece of the pricing concept. We have seen in a study of out New York that 20 percent of riders shifted to transit after variable pricing was implemented. The express lanes on I-95 in Miami also showed an increase in transit use with the pricing program. Although air quality is an important issue, it has not been studied to the extent that we would like to see it done. We believe that because there have been increased benefits to the traffic flow, there are indeed opportunities to calculate improvement to air quality as a result of the pricing programs. There is public acceptance as shown in the San Diego and Minnesota experiences. When people see that it works people generally support it. Initially, equity issues would come up before lanes are built. The HOT lane conversions were seen as being built for the rich but these concerns diminished as they began operation and it was seen that anyone can benefit from them especially when there is good public transit system available. We have seen technology development that has occurred with the advance of these pricing programs.

Where do we go in the future? Reauthorization is out there. The long term future in pricing is unknown, but in the short term I think there will be more HOT lanes and express lanes tied with transit especially with BRT services like in Tampa. Networks of priced lanes will be emerging like in Seattle, Miami and Minneapolis. Pricing strategies will become more a part in the metropolitan planning process. We may see more priced shoulder lanes as in the Minneapolis experience. In the long term there will be the move towards full facility pricing. We have seen a little of this with the I-520 project in Seattle and the inter-county connector in Maryland. There are lots of benefits with full facility pricing but it is still unacceptable at this time. I believe there is potential for it in the future. The most important thing to leave you with is that planning for congestion pricing is the most important thing for areas to do.

Arterial Managed Lanes
Chris Swenson, Wilbur Smith and Associates

We are going to talk about how you manage arterial lanes in the same way as managing freeways, even though there are a lot of access points on arterials. Why manage arterials? This started through the Office of Operations with Wayne Berman and staff. We know that in the metropolitan areas freeways are over-crowded and people are already moving to the arterials. We even see this in small to medium size areas such as Lee County where the arterials are the backbone of the transportation system more so than your limited access facilities. So there is a real need to price arterials even in smaller areas.

We need to look at efficient priced capacity to keep new lanes uncongested and improve services. We have to get the message out to the public that priced lanes offer much greater throughput; that we will not be robbed of our capacity during the peak hour, when we need it the most, if we are pricing at a level to manage those lanes. Self generated revenue means we can do projects.

We know how to price for a limited access facility, but for arterials we manage through queue jumps. Queue jumps recognize that arterial capacity is defined by intersection capacity. A grade separation queue jump, allows for the bypass of the traffic signal. The driver has the option to take the queue jump and pay or stay in the traffic and face the congestion. The way the queue jump works is that the lane itself becomes a natural collection point with all electronic tolling. Queue jumps can be over passes or underpasses. Bob Poole suggested queue jumps could be underpasses and I told him, sure if we issue everyone a submarine. We have found that underpass queue jumps can be a viable idea and the cost was not that much greater than over pass construction and drainage was not an issue. Hurricane evacuation is not a problem because the weather is usually fine prior to a storm and we are given ample warning, usually 24-72 hours notice, of a storm.

Benefits of queue jumps versus standard lanes:

• they are self-funding through toll revenues;
• they create a greater capacity from existing lanes;
• the right-of-way takes are not as bad;
• you don't have to add an additional lane anywhere except the intersections;
• they facilitate region-wide high speed bus rapid transit; and,
• reduce congestion for all drivers, not only those who take the lane but those who do not and those on the cross streets.
• they are not as restrictive as full limited access facilities.
• queue jumps obviously lend themselves to networks.

The synergy of queue jumps, bus ramps and transit creates the opportunity for virtual exclusive busways (VEBs). The facilities should have pricing limits that allow flow to what is comparable to a level of service C, just like on managed lanes. It is important to remember that the lanes between these queue jumps can handle a lot of traffic; the delays on roadways come from the intersections, not the roadways. A reliable uncongested speed is sustained due to long term effects of pricing.

We are also looking at the feasibility of VEBs. A VEB facilitates a region-wide high speed express bus service. In small to medium urban areas fixed rail gets to be an issue because you don't have the density to support it. These small to medium urban areas would benefit from queue jumps to promote BRT.

How much do queue jumps cost? The basic queue jump is $35 million. The 5.6 mile example corridor has six queue jumps and the total cost of the project came in at $277 million. The cost was higher in our project due to real world problems such as intersections being too close together, rail road tracks close to the intersection and other factors that made us extend the length of the queue jump. The project did not require a lot of additional right of way.

What are the revenue assumptions? The peak hour toll is $0.45 in 2007 dollars. The shoulder hour (hours just before and after peak hours) toll is $0.35. The off peak toll is $0.20 and the weekend toll is $0.25. There is a 60 percent peak usage rate and drivers are gaining 2-3 minutes per queue jump. There are about 50 percent shoulder hour tolls, 35 percent off peak tolls and 40 percent weekend tolls.

We assumed our typical operating costs at 20 percent of our revenues, which we think is high. Overpass queue jumps operating cost is 30 percent. Underpass operating cost is obviously more because of possible drainage issues. Consumer Price Index at four percent, discount rate at six percent, total revenue over 30 years is $790 million and that gives the project a present value at $285 million.

In closing, former Secretary for Transportation, Norma Mineta, said "congestion is not a scientific mystery" because we know what causes it and we (in this room) know how to get rid of it. It is not an uncontrollable force, but it has resulted in some poor policy choices and a failure to separate the solutions that are effective from those that are not.

Tolled Truckways
Annie Nam, Southern California Association of Governments

The Southern California Association of Governments (SCAG) comprises six southern California counties and has 189 member cities. SCAG is the largest MPO in the US in terms of population and square mileage, serving 18.6 million people and 38,000 square miles. Four of the top 30 US gateways are located in Southern California including the Port of Los Angeles-Long Beach. SCAG regional population is expected to grow by nearly six million more people by 2035. The San Pedro Bay Ports recent cargo forecast projects reaching 43.2 million TEUs (twenty-foot equivalent units) by 2035. These are recently updated projections. SCAG's recent warehouse demand and supply analyses show that port and non-port related demand will absorb available space (suitably zoned vacant land) by 2028. By the year 2035 warehouse space is projected to be short by 288 million square feet. Additionally, segments of our highways are facing substantial truck volumes. Projections show volume as high as 90,000 trucks per day on some corridors.

The 2008 Regional Transportation Plan included the regional toll truck lane concept with a dedicated and separated toll truck lane system from the Ports of LA/Long Beach along the I-710, and extending further east and then up along the 1-15 (to Victorville)— linking the ports to major warehousing and distribution centers further inland. An alignment for the east-west corridor is still in the process of being further defined. Nevertheless, the initial segment of a potential regional system has progressed with an Environment Impact Report (EIR) and Environmental Impact Statement (EIS) currently underway for the I-710. Financial feasibility may be a significant issue for implementation of a system as costs are estimated to be significant.

How can road pricing and truck tolling systems help? The region has taken a serious look at pricing/tolling as a means to facilitate financing. SCAG looked at the potential benefits of a truck tolling system by conducting a very high-level analysis using the SCAG Regional Travel Demand Model to generate year 2030 estimates of truck toll facility demand. SCAG employs a separate model component for Heavy Duty Truck trip generation and distribution.

We assumed a per-mile toll cost (based on average rates for express lanes throughout the region), estimated travel time savings, buffer time savings (depending on the likelihood that unexpected traffic or accidents would increase the typical trip times, a factor was added for trip planning time or "buffer time" to ensure that goods reach their destination on schedule), and an estimated industry value of time to assess capital costs recovery—assumed at $73 per hour. Using this model, the 2030 time period was used to estimate the impact on time and reliability if separate truck lanes were built. The model compared this situation to that faced by cargo carriers if no transportation projects other than those currently planned were built. Savings were measured in terms of travel minutes saved as well as planning minutes saved due to not making such large contingency or buffer allowances. The time savings in minutes were converted to fractions of an hour and multiplied by $73 to estimate the cost savings to a shipper from having the separate truck lanes available to them.

The estimated potential value ranged from $103 for a trip from downtown to the port, up to $345 for a trip from the port up to Victorville–suggesting that truck lanes could potentially offer shippers and trucking companies a significant value proposition. This analysis was conducted a few years ago and at a high-level. We are continuing to conduct extensive outreach with industry representatives to obtain more data and to refine some of the analysis.

Additionally, SCAG is in the process of releasing the updated Port and Modal Elasticity Study. Findings to date provide critical information about the implications of pricing on the industry. Although pricing scenarios evaluated involve fees in the form of levies on containers rather than tolls, the cost of transport more generally can have significant impacts on the freight industry—most notably our analysis focused on the potential for cargo diversion from the Ports of LA/Long Beach (San Pedro Bay ports) due the imposition of hypothetical container fees.

Compared to the 2005 analysis (Phase I study), the elasticity of imports via San Pedro Bay to potential container fees increased markedly due to unfavorable changes in transportation rates, including aggressive rate competition from other North American ports. The analyses of various scenario runs highlight several key points as follows:

• The elasticity of imported cargo to potential fees is much more sensitive than previously thought.
• There are disparate elasticities depending on the various categories of import volumes routed via the San Pedro Bay Ports.
• The three basic categories of imports include: (1) local imports, consisting of imports consumed within the greater region (southwest region) (2) direct-shipping imports (Inland Point Intermodal or IPI), consisting of imports destined to other regions which simply pass through Southern California while remaining intact in the marine box coming from Asia; and (3) trans-loaded imports, which are imports consumed in other regions that are unloaded from the marine box in Southern California, possibly receiving value-added services and ultimately re-loaded into domestic containers or trailers for re-shipment to other regions.
• There are various factors that could considerably impact the elasticity of imported cargo, e.g., transportation rates, market share of large nation-wide retailers in Southern California, and infrastructure investments.
• Infrastructure improvements (such as the regional trucklane system) that reduce the container travel time may be a value proposition for the trans-load business (large, nation-wide importers and importers of high-value goods).

A number of studies are currently underway at SCAG, including our Comprehensive Regional Goods Movement Strategy and Implementation Plan—to develop a detailed goods movement action plan for the region. As a part of the effort, SCAG is also analyzing the impact of pricing in more detail, including road pricing, on commercial vehicles—to consider how shippers and commercial vehicle operators may respond to pricing proposals and how these responses may affect the transportation system and economic outcomes.

A New Approach to Freeway Congestion Pricing
Bob Poole, Reason Foundation

I will discuss the politics of freeway congestion pricing, as in whole-freeway pricing. Why would it make sense, and why is there such political opposition to it? In principle, there is a strong case for pricing. There is a large supply and demand imbalance on the freeway systems in our large metropolitan areas. Pricing can reduce or eliminate congestion during peak periods, generate revenue, and target investments to areas where additional capacity is most needed, particularly where bottlenecks exist.

But there are serious political obstacles to charging for something that has always been "free." In Pennsylvania, when the state wanted to put tolls on Interstate 80, highway users and trucking groups mobilized and fought very hard to stop tolling. They argued it was double taxation or "paying twice." Elected officials take this very seriously, in part because 91 percent of U.S. households have one or more cars, with most having two or more. That means "voters" equals "motorists." Putting a price on an un-improved freeway can be seen as monopoly pricing (exploiting people).

Federal efforts to promote road pricing date back to UMTA (Urban Mass Transportation Administration, the predecessor of the current FTA) which made planning grants for road pricing back in the 1970s, They offered more money to cities that would implement pricing but got no takers. After the Value Pricing Pilot Program began in the 1990s, one grant to Los Angeles funded a 50-member task force to consider peak-period pricing on the freeways in Los Angeles. After one and half years of planning and discussion, including traffic and revenue modeling by Wilbur Smith Associates, the group concluded there was no chance to do this (politically), so our "second-best" recommendation was to move forward with HOT lanes. As you may know, there are still no HOT lanes in Los Angeles County; however, two HOV to HOT conversions are now under way, 15 years later. So you can see how long it has taken to overcome political opposition even to that limited form of pricing.

The Urban Partnership Agreements program a few years ago was based in part on hopes within the Office of the Secretary that one or more jurisdictions would propose pricing an existing freeway--but no one bit on that. The only place that proposed pricing existing capacity was Manhattan, but that fell through. The I-520 bridge project will toll the bridge but only to pay for a new bridge, so that's really new capacity, not existing capacity. Overseas we have only seen three successes in 30 years: Singapore, London, and Stockholm, with many more failed attempts, such as Hong Kong, Cambridge, Manchester, and The Netherlands. All these overseas cities which have lower car ownership, better transit, and denser land uses than nearly all U.S. cities, but it's still very difficult to put pricing on their roads.

In thinking about political opposition, we need to face up to the reality that congestion pricing will produce losers as well as winners. The people who pay the tolls will be winners due to time savings. Some people will be losers because they would not use the facility if they have to pay and will divert to the arterials (because in most spread-out cities such as Houston and Dallas transit is not a viable option for them). Another set of losers will be the existing arterial users because the "tolled-off" will add to the existing congestion on arterials. So, overall, the losers would probably be a larger group than the winners. Because congestion is so high in areas that need ;pricing the most, the prices needed to restore free flow would be very high. And if the revenues are not used to expand the freeways and eliminate bottlenecks, customers will feel they are facing monopoly pricing and will oppose it. That's especially true in places where transit is not a viable option for most people.

There have been many proposals for overcoming political opposition to pricing. Ken Small, a researcher at the University of California at Irvine, in 1992 suggested using most of the revenues as tax rebates and about one-third of the revenue for transit and highway improvements. Most other proposals suggest using revenues to expand transit. But in most U.S. metro areas the transit mode share is five percent or less, so even if transit were doubled, this would still leave 90 percent mode share on the highways. Right now, the average door-to-door transit trip in the US takes twice as long as the average single-occupant vehicle trips, even with the current state of congestion.

It is time to rethink the model we have in our heads about how we would price freeways. I want to question two assumptions in our standard model: the single-price assumption and the GP lanes assumption. The single-price assumption is that the same variable price would apply to all vehicles. But Ken Small and others have found there is a huge variability in both value of time and value of reliability. Charging a single price is not going to get it right for most people. A single-price system will most likely charge too little to people with high values of time and reliability and too much to those with low values. Obviously a freeway cannot have hundreds of prices but a two-price model is workable--a premium price for those who have a high value of time and reliability and a much lower price that still spreads out the peaks for everyone else. When Small and colleagues modeled that, it produced much greater social welfare than a single-price model.

The GP-lanes assumption is based on the standard engineering case for GP rather than specialized lanes. A multi-lane road has a higher throughput than one with physically separated lanes because faster vehicles can pass slower-moving ones in a multi-lane context. If you decide to have specialized lanes you cannot have 1.5 lanes for trucks, you either have one or two lanes. But we have learned with HOT lanes that a special type lane in certain situations can be kept full and flowing, thanks to pricing. And there are even special cases where it looks as if exclusive truck lanes would be justified.

Pulling these points together, I believe our revised freeway pricing goal should be:

• Premium lanes with premium prices that essentially offer a money-back guarantee of Level of Service B or C at rush hour;
• General purpose lanes with modest peak pricing, designed to spread out the peaks; and,
• Truck lanes in select corridors (and only select corridors),with truck-specific pricing.

I think this scenario would produce more winners than losers. And therefore this scenario stands a better chance of happening than the single priced systems we have in our heads right now.

This two price system can and should be introduced in an evolutionary approach where each step is justified solely on its own merits, with no announced plans for any further steps, unless support develops for those steps:

• Step 1: We expand on early HOT lanes to create HOT networks, mostly one-lane-per direction, converting existing HOV lanes and adding missing links and flyover connectors at interchanges. There will be considerable capital costs because there are a lot of links to fill in and flyovers to build.
• Step 2: After the HOT network is functioning and people appreciate its benefits, that would be the time to propose expanding it to two lanes per direction, in most cases by converting an adjacent general purpose lanes (since that would be far less costly than building new capacity, and by then people will have seen the benefits of pricing and understand that if there is more priced capacity, prices could be lower and more people could use it).
• Step 3: Once Step 2 is in place and working, at that point propose adding modest peak pricing on the remaining general purpose lanes and perhaps add a truck-only lane in a few specific corridors. Of course a truck lane would not have to wait until this step, it could come earlier.

The benefit of the evolutionary approach is that each step is justified on its own merits, and does not require anyone to commit to a future step that most people would be skeptical of at that point. This approach relies on people learning from their experience with pricing.

My conclusion is that we should understand and accept that the conventional pricing approach to freeway pricing is not making headway. But HOT lanes and HOT networks are making headway. We need to build on the current HOT successes in an evolutionary fashion, taking far more seriously that different people have different values of time and using that knowledge to develop a more evolutionary approach. The big bang approach, where the entire freeway is priced all at once, most likely will not happen. But I think the evolutionary approach outlined here has a much greater chance of being accepted.

Audience Questions

• We all know it is difficult to get support for pricing, yet Craig Stone with WashDOT indicates the PSRC, with a 98 percent vote, has embraced the program which will go from HOT lanes to full freeway pricing. As Congress is trying to restrict and/or stop tolling programs, these elected officials (in PSRC) are embracing tolling systems. I think that what will drive this is that it will be revenue. I believe the very nature of our interstate system, which was the federal government paid for it, handed it over to the states and now sits there and says it is already paid for and watches it fall apart and does not want to implement tolling, is stupid. Now that the freeway system needs excessive repairs, the government has cut funding, is reducing their role in the system. If the federal government cannot solve the problems with the system, it needs to move out of the way and this could be an argument that the states could use to implement tolling. I think the idea of revenue from pricing may become more attractive to the federal government as an option. I would just like to hear anyone's comments on this idea of these contradictions in public policies.

  Annie Nam: I recently proposed pricing to the Southern California Regional Council and it did not go over well. They didn't understand why we would want to do this. I do think the revenue potential is driving some level of discussion. What is somewhat hopeful is the fact that in California we have SB 375 which requires the state to meet greenhouse gas reduction targets. Due to SB 375, we spent a considerable amount of time discussing pricing and its role in the target reductions. I think in the future there may be more of a willingness to explore various aspects of pricing beyond HOT lanes.

  Chris Swenson: In Florida on the I-75 project (which had four free lanes) we did focus groups to look at the possibility of having two HOT lanes and two general purpose free lanes (each way) versus having only one HOT lane and two free general purpose lanes. In the groups, seven out of eight wanted the two additional HOT lanes. The one person that did not want the additional toll lane was only concerned with the additional 12 months of construction and not the toll. I'm seeing a lot of evidence that the elected officials are not always in tune with what the people want.

• What drove the PSRC to get to where they are and the elected officials to support it?

  Craig Stone: We have a legacy of steps along the way. Each of the alternative transportation plans has included a pricing idea. We have a pollution management act and a very heightened environmental culture in PSRC. The key part is there is a culture of asking, "How do you bring this together?" We also have bright individuals managing the process, who are building a consensus and posing questions to elected officials about financing. We were able to bring transit people into the process because they see the tax base revenue going down and they are looking for sustainable revenue. We were able to bring in the greenhouse gases people (Choices people) because we have a state law and they keep saying we can't keep building freeways. We were able to find common ground by going around to each subject matter expert and asking "how will you solve this?" The most promising answer would come from pricing and we found when we brought pricing into the picture, it helped balance our system. We had older statesmen who were well respected start buying into the idea and started showing leadership for the idea. It was a long process of getting the support for the project. All of this movement led to the vote that provided the 98 percent support for it. There are still nay-sayers out there but there will always be some with any project. It is a combination of revenue, balance, and education that worked to convince them.

  Wayne Berman comment: Those six programs in the federal government are confusing and limiting and in some sense hampering to pricing. There are limits to the number of slots and we cannot always decide who to give them to. My read is that at least in the urban areas it streamlines what can and cannot be priced in the urban area.

  Bob Poole: Rather than the status quo we need to think of liberalization of programs. Right now we are still in a learning-by-doing phase.

• Did you guys study queue jumping at grade with the use of signal coordination instead of grade separation?

  Chris Swenson: Yes, there is some possibility but we did not include it in our study because in this case we felt that physical separation was necessary in trying to minimize the right of way.

  Bob Poole: In regards to the evolutionary approach, you may only be able to do a few queue jumps at a time and maybe you could coordinate signals to work with them in order to get the project up and running. If you start with a few queue jumps, it may make it easier to convert to a more traditional system of queue jumps down the road.

• A few years ago when HOV lanes came in, the objective was to increase throughput and vehicle occupancy. We are seeing a struggle with HOT lanes because a lot of people are saying why don't you just toll everyone? What is the future and role of pricing in encouraging increased vehicle occupancy? Does pricing have different objectives now? Should we trying to get off of the whole HOV system?

  Chris Swenson: We are looking at moving people from one place to another, not cars. There are two hindrances with the high occupancy factor. The first is that we do not have the technology to enforce occupancy verification but we do have the technology to toll (which is easier). Right now it is not efficient to manage high occupancy lanes, so right now we are headed towards managed lanes instead of HOT lanes. I think when the technology comes around we will be able to put the "HO" (high occupancy) back into HOT lanes. The second thing is that I think we have another mode that pricing is pushing us into is the electronic mode. I am talking about telecommuting. People will stop and think, if it is going to cost me more to get to work, do I really need to be there? Telecommuting is a greener approach and a more cost-efficient approach. I think that until we can double transit and add a lot of remote working, we really won't be able to implement pricing completely.

  Bob Poole: We have gotten a lot better documentation in the last couple of years that shows that two person carpools are largely family members not reducing vehicles off the road. We should encourage more vanpools and express buses in response to priced lanes because these modes could reduce the number of vehicles on the lanes more than the existing carpools.

  Wayne Berman: We have not generated the carpools to fill the lanes. If we had promoted more employer-based programs that allowed employees much more flexibility to carpool, we might not be talking about this now. It is the employer-based programs that have not encouraged more carpooling and vanpooling activity. Shared riding is stagnant and I don't see that pricing is going to change that.

• I like your phasing approach (to Bob Poole). I think that HOT lanes are causing a two class system which lead people to weave in and out and change lanes. Some studies have shown that these lanes can lead to hypercongestion. These studies suggest that it would be better to have all lanes unpriced rather than a two class system. We see now that the government is only putting in around 20 to 25 percent of the funding needed to implement new HOT lanes, so how long will it take raising taxes to accumulate enough revenue to get to the second stage?

  Bob Poole: That is a very good question.

• Pricing will change behavior. Chris, if you want people to work in their pajamas you don't have to legislate it, pricing will do that. The pitfall, it seems like, in your evolution is that you are wanting to build your new capacity as HOT. By doing this you are creating the political hurdle of taking away free use of those lanes. Why not build the new capacity as full-priced managed lanes and do the conversions from HOV to HOT?

  Bob Poole: That is what I was thinking, I'm sorry if it did not come across that way. All new capacity would be fully-priced express lanes.