This tension has existed from the very beginning of HOT lanes, at their birth in California. America's first two HOT lanes, though both using value pricing, represent very different business models. The 91 Express Lanes were developed as Express Toll Lanes, but required by the terms of the franchise agreement to give discounts to HOV-3s. Despite that modest revenue leakage, they produced enough revenue to pay for not only their construction but all operations and maintenance, highway patrol services, and even local property taxes.
The I-15 Express Lanes in San Diego already existed as HOV-2 lanes, but were then allowed to sell their excess capacity to paying (single-occupant vehicle) customers. Toll rates are kept high enough to permit extensive levels of car-pooling, and the modest toll revenues are used to heavily subsidize bus service in the corridor. San Diego has strong local support for expanding the project, going from two lanes to four and from eight miles to 20. But because they are committed to the carpools-dominant model, the projected toll revenues are insufficient to support revenue bonds to pay for the $660 million expansion project. Instead, funding will come largely from gas taxes and the local transportation sales tax.
Some HOV to HOT conversions are going forward using the San Diego model (e.g., in Denver and Minneapolis). But there are important straws in the wind for rethinking who should get free access to congestion-relief lanes. Houston's $1.1 billion project rebuilding the Katy Freeway is adding four HOT lanes that will give free passage only to HOV-3s and buses, replacing the current HOV-2 policy. In Tel Aviv, a planned HOT lanes facility will offer free passage only to HOV-4 and higher. And now Maryland has announced its preference for pure Express Toll Lanes (everyone pays), to be added potentially to all the major limited-access facilities in the state's urban areas.
There are three good reasons for moving to pure Express Toll Lanes. The first is the cost of new congestion-relief lanes. Many of the most congested urban freeways (I-880 in Oakland, US 101 in Los Angeles, the Washington Beltway) don't have HOV lanes now and likely never will have them. Why? Because it's simply too expensive to condemn land for widening or to build elevated lanes in these built-out urban areas. The only way new congestion-relievers can be added is with premium tolls to pay the high costs. But if you give away half or two-thirds of the capacity to two-person "fam-pools," you will never have enough revenues to pay the $10-30 million per lane mile needed to build them.
The second reason is the value these lanes provide. Relief from rush-hour congestion is very valuable, as evidenced by people's willingness to pay 50-60 cents/mile in San Diego and Orange County. It simply makes no sense to give away this extremely valuable space, whether it cost a fortune to add or was already there and converted from an underutilized HOV lane. This is very basic economics; failing to charge is like leaving $100 bills on the sidewalk.
The third reason is enforcement. Let's be honest: As of today, no technology exists to rigorously enforce a random mixture of carpools and paying vehicles. To be sure, extensive patrolling by officers in black-and-white cars can deter violations to some extent, especially if coupled with California-type $271 fines. But there is no technology that can reliably tell how many people are in a car when it's zipping past stop-and-go traffic at 65 mph, mixed in with transponder-equipped paying vehicles. Second-generation HOT lane facilities, with multiple entry and exit points, are going to have serious enforcement problems.
By contrast, a policy of only granting free passage to "super-HOVs" is compatible with such vehicles having to register as, say, company sponsored van-pools or authorized transit vehicles. Such vehicles would be given a transponder permitting them to operate in the Express Toll Lanes. Enforcement can then be completely automated. Every vehicle must have a transponder; a vehicle without one is automatically a violator, easily identified and videotaped for follow-up enforcement action.
In Reason Public Policy Institute's 2003 HOT Networks policy study, we estimated the cost of building seamless networks of congestion-relief lanes in eight of America's most congested cities. The total cost was $43 billion, mostly due to the very high cost of adding elevated lanes to the older, central-area freeways and of building flyover connectors for the premium lanes at freeway interchanges. And our estimate of toll revenue bonds that could be issued to support such construction was $29 billion — about 2/3 of the total. And that was with a policy that only vanpools and buses, not carpools, would go free.
So we face a very real policy choice in urban transportation. We can continue to give free rides to two-person carpools, but the cost of doing so will be (1) the HOV lanes fill up and cease to provide time savings, and (2) we forego the revenue needed to create a complete network of congestion-relief lanes.
Yes, car-pooling is a nice thing. But is it really so important that we should allow it to trump other transportation goals? It's time for serious debate on this issue.
Robert W. Poole Jr. is director of transportation studies and founder of the Reason Foundation.