Public Works Financing

Keys to Financing Managed Lane Networks

The greatest challenge in implementing such networks is not political acceptance; far more difficult will be paying for them

By now it’s pretty clear that most of America’s major urban areas hope to implement region-wide networks of priced (“managed”) lanes on their freeway systems. Plans for such networks exist in places as small as Minneapolis/St. Paul and as large as greater Los Angeles. The greatest challenge in implementing such networks is not political acceptance; far more difficult will be paying for these vitally important networks. And most transportation planners have not faced up to the kinds of trade-offs this will involve.

Last month I addressed this question in a webinar organized by the folks at Atkins, who gave me free rein to present my thoughts (and who may or may not agree with what I write here!). My key message is that the policies adopted for a managed lane network must focus on revenue generation to a greater extent than many people have come to grips with.

To preface this discussion, remember that two perspectives on managed lanes and money are in wide circulation. One says that managed lanes cannot pay for themselves out of toll revenue and are doing well if they cover their operating and maintenance costs. The other view, embraced by public-private partnership (PPP) advocates in particular, is that the majority of capital and operating costs can be financed out of toll revenues.

Ironically, both perspectives are correct. But they refer to two fundamentally different business models. Type 1 projects are viewed and operated as high occupancy vehicle (HOV) lanes that sell any excess capacity to willing customers. They are nearly all conversions of existing HOV lanes, and nearly all permit HOV-2s at no charge. By contrast, Type 2 projects are express toll lanes selling faster and more reliable trips to paying customers, with some exceptions for non-paying vehicles. These differences have a profound impact, especially on revenue generation. For the webinar, I assembled data on representative projects of each type: the average annual revenue of the Type 1 set was $3 million compared with the Type 2 average of $26 million.

So the major point in my webinar presentation was that if you aim to create a network, Type 2 is the only sensible way to go. That is because a network will involve many billions of dollars in capital costs, and the most realistic source of those billions is toll-revenue financing. Also, a number of details of design and policy can contribute significantly to revenue maximization. Let’s look more closely at several of these points.

On the very high cost of a managed lanes (ML) network, the reality is that most of the candidate metro areas (except for Los Angeles and San Francisco) have HOV lanes on only a small fraction of their freeway systems. And even those two have large stretches without HOV lanes to convert. So reason #1 for high cost is that in most cases the large majority of the network will be lane additions, not lane conversions. Secondly, if the business model is to deliver fast and reliable trips during peak periods, the network must include seamless ML to ML connectors at freeway interchanges--and those cost several times as much per lane-mile as ordinary lane additions. Third, in the largest and most congested metro areas, two MLs per direction, not one, is likely to be optimal, so the lane addition costs must reflect that reality. Fourth, there are safety and traffic flow reasons to prefer direct-access ramps rather than slip-ramp access to and from the MLs and the regular lanes; which again means higher capital costs.

It is factors such as these that lead to estimates in the $16 billion range in network studies for Atlanta and Miami, whose networks will be of the Type 2 variety. That contrasts with the current network plan for the San Francisco Bay Area, whose Type 1 approach relies mostly on HOV-lane conversions and one-lane-per direction MLs. (And also, as of now, would give away half or more of its capacity to HOV-2s.)

HOV policy relates not only to revenue potential, but also to reliability of trip times. To maintain level of service (LOS) C conditions during peak periods requires that all, or nearly all, vehicles be priced and hence deterable from entering the network when demand exceeds capacity. But a system that exempts car-pools, hybrids, electric vehicles, etc. not only foregoes needed revenue but drastically reduces the effectiveness of its pricing.

Those who are seriously planning ML networks are cognizant of these problems, which is why we’ve seen several metro areas (including Atlanta and Miami) switch from HOV-2 to HOV-3 in implementing new MLs. Also positive is the requirement that all ML users register and obtain a transponder in order to use the lanes. In my view, ML network advocates should be pushing hard for person-throughput goals to be met via express buses and vanpools, rather than car-pools and fam-pools. That is the current policy for the MLs under construction on I-95 in Maryland and on the MoPac in Austin, as well as what’s proposed for Bus Toll Lanes in Tampa.

Related to this question is enforcement of HOV occupancy requirements. The newest approach of switchable transponders still relies on manual enforcement by highway patrol officers, a costly and inefficient process. And even the newest camera-based systems can, at best, verify front-seat occupants, which means they are not applicable to MLs that require at least HOV-3 for free passage. I have suggested in a Transportation Research Board paper that if HOVs are to be given free or discounted passage at all, the enforcement should be off-line, via regular audits of the registered, employer-sponsored car-pool agreements.

My bottom line on ML networks is this. A network based on the Type 1 (HOV) business model will be very difficult to fund and will not be very effective at reducing congestion. ML networks should be planned, developed, and operated according to the Type 2 business model, focused on providing fast and reliable trips to paying customers, buses, and vanpools. That’s how they will maximize toll revenue to pay for the network and minimize congestion for their paying customers.

Robert W. Poole Jr. is the Searle Freedom Trust Transportation Fellow and Director of Transportation Policy at the Reason Foundation. This article first appeared in Public Works Financing.

Robert Poole is Searle Freedom Trust Transportation Fellow and Director of Transportation Policy