Price Caps Are Hurting Managed Lane Performance

Share this article on:

Priced managed lanes provide drivers with an option to bypass congestion for a small fee. The key to ensuring uncongested travel is raising or lowering the variable toll to ensure the managed lane operates at a level of service D or better, allowing travel at 45 mph or higher. Toll rates typically range from $0.10 to $1.00 per mile, with higher rates on very busy corridors such as I-66 in Northern Virginia.
The low overall toll rates haven’t stopped misleading news stories and claims by anti-toll advocates that Americans are being forced into poverty by high tolls. Since voters have a negative visceral reaction to tolls, some politicians have pressured managed lane operators to impose price caps. While these people may mean well, a price cap undermines the purpose of pricing, worsening travel conditions in the managed lane without improving conditions in the general purpose lanes.
Traffic throughput versus speed is a sideways U-shaped curve. At 40 mph, a freeway lane has a throughput of 2,000-2,500 vehicles per hour. However, at 10 mph a freeway lane can accommodate only half that many vehicles. Further, when a significant share of managed lane vehicles are buses and vanpools, managed lane travel speeds have an even more dramatic impact on the number of people moving through a corridor. Pricing the managed lanes too low hurts almost every user of the highway.
A recent Texas A&M Transportation Institute study written by Nick Wood quantified the impact of price caps on two managed lane corridors: I-95 in South Florida and I-405 in the Seattle suburbs. Speed, volume and toll rate data were collected. The study focused on weekdays; weekends and holidays were not examined.
Wood found that the price cap limits the effectiveness of the 95 Express lanes in South Florida. In the northbound lanes during the afternoon peak period, the capped toll rate is in effect between 4:15 and 6:45—the peak of the peak. While the cap is in effect, traffic speeds and managed lanes performance nosedive. The speed drops below the federally-required minimum of 45 mph for 3 hours (4:00-7:00). The number of vehicles the managed lane can accommodate drops from 1,267 per hour at 3:30 PM to 1,123 during the peak travel time. Overall PM peak period capacity falls by almost 10 percent.
Despite having the legal ability to do so, FDOT has not increased the cap. Originally, the Florida Legislature limited the maximum toll between mile markers 4 and 12 to $1.50 per mile. However, Florida DOT can increase the toll by up to $0.50 per mile if the managed lanes reach the capped toll rate on at least 45 days in a six-month period. Thus far FDOT has not increased the toll cap. Further, if a customer’s average trip speed falls below 40 mph, a recent Florida law requires that they be charged a minimum toll (although it is unclear what that toll would be).
The South Florida price cap leads to some perverse effects. While traffic moves at the speed limit early in rush hour, as more users flood the lane the speed quickly drops. As a result, commuters are penalized for leaving work later. Those who use the Express Lanes at 3:30 PM pay $7 to travel at 52 mph. But those who use the lanes at 6:00 PM pay $10 to travel at 35 mph. Travel speeds in the managed lanes average only 10-15 mph higher than the GP lanes during rush hour. The entire purpose of managed lanes is to provide a congestion-free option during rush hour when the general purpose lanes are more congested. If the managed lanes are almost as congested as the GP lanes, managed lanes are not operating as intended.
The Washington State price cap is even more problematic than Florida’s. In the one-lane southbound segment during the morning peak, the capped toll rate is $0.68 per mile between 7:20 and 8:35 AM. As with the I-95 lanes, performance suffers. During the peak morning rush, managed lane speeds average 24 mph. GP lane speeds are only 2-6 mph slower. The managed lane speeds are well below the federally-required 45 mph for three hours (6:10-9:10). Managed lane carrying capacity drops from 1,592 vehicles/hour at its peak to 1,063 vehicles at 8:30 AM. Overall AM peak capacity decreases by 27 percent.
The Washington State Transportation Commission is more limited in its ability to adjust the $10.00 cap. The managed lanes must meet the federal performance minimum of at least 45 mph at least 90 percent of the time, yet WSTC cannot raise toll rates to limit overloading the managed lanes. Further, humans are not allowed to override the algorithm, even in the case of an accident. As a result, when a traffic crash occurs in the GP lanes, the price in the managed lane cannot be increased, so it is quickly overloaded, losing most of its effectiveness.
The Washington cap, combined with geometric design shortcomings in the corridor, has led to poor managed lane performance. A University of Minnesota study sensibly recommended that the $10.00 cap be eliminated. It also found that the variable toll rates do not rise fast enough; some drivers were paying $4 less than others in similar conditions.
Building on the University of Minnesota study, I have several recommendations to improve managed lane performance on both I-95 and I-405. First, DOTs and toll road agencies need to eliminate or at least the raise price caps. Second, they should further limit free passage for carpools in managed lanes, so that nearly all vehicles are subject to variable pricing. Finally, USDOT needs to enforce strictly the rule that federally aided managed lanes operate at 45 mph or more at least 90 percent of the time, and require states operating managed lanes that consistently violate this rule to repay federal highway construction funds.
Most managed lane agencies’ goals are to 1) provide a guaranteed uncongested travel option and 2) move more people through the corridor in cars, vanpools and buses. Agencies need to operate the lanes to best serve their customers. The threat of loss of funds should help ensure that political interest does not take precedence over serving managed lane customers.