Timothy T. Taylor Home Page
Journal of Economic Perspectives
Articles and Writing
Economics Textbook
Classroom Teaching
The Teaching Company
High School Pedagogy

Articles and Writing

September 14, 1992
"(Thinking In Traffic)"
San Jose Mercury News

By Timothy Taylor
<< Back to 1992 menu

DOWNS'S LAW of Peak-Hour Traffic Congestion was formulated back in 1962. It reads: "On urban commuter expressways, peak- hour traffic rises to meet maximum capacity."

The logic is remorseless: As long as highways are faster than taking stop-and-go surface roads, drivers will take to the on-ramps. The eventual result will be stop-and-go traffic on the freeway.

If a new highway or mass transit system is constructed, rush hour traffic may disperse for a time. But then commuters who were using other routes, or driving at other times, or taking mass transit will recognize that they no longer need to avoid the rush hour traffic. They will shift to the highways until jams re-occur.

In his just-published book "Stuck in Traffic," Anthony Downs of the Brookings Institution summarizes three decades of thinking about how to tackle his gloomy-sounding law. After offering background about the causes and trends in traffic congestion, the book considers 23 solutions that have been proposed, listed nearby in his order of preference.

Consider them from the bottom up, saving the best for last, and remembering that the focus here is how these policies affect traffic congestion, not the sum total of their virtues and vices.

Growth limits, option 23, are Downs's least-favored option, not because they don't reduce traffic in a given community, but because they tend to move vehicles to the next town, or to spread traffic flow over a larger area, neither of which qualifies as a real solution.

Raising auto fees, option 22, might discourage some folks from owning cars, but fees are unlikely to rise by enough to reduce congestion substantially.

Improving the job-housing balance (21) in a community could make commutes shorter, while clustering jobs together (20) might make mass transit or ride-sharing more practical. But either step affects only traffic related to jobs, which is perhaps half of total rush-hour traffic.

Moreover, these policies will have little effect on existing congestion, and will be difficult to enact. Businesses will complain loudly (and legitimately) if a loud bureaucratic voice dictates the locations to which they can assign their workers. And while Downs readily admits that shifting the jobs-housing imbalance can have a variety of social benefits over time, he writes, "It is a task akin to moving mountains without using either bulldozers or dynamite."

Options 15 through 19, which involve building more roads or larger mass transit systems, can all make a short-term difference, but they all run into two problems: high cost, and the inexorable forces of Downs's Law.

Options 9 through 14 are worth trying, because they don't cost much, but don't expect too much from them, either. For example, working at home (9 and 10) and staggered work hours (12) share the problem that many people like to be near their co-workers most of the day, for reasons of convenience, sociability, and efficiency.

Traffic flow might be improved (11) by coordinating signals, planning one-way streets, using ramp signals, and so on. Putting new high-density housing near existing mass transit (13) has little effect on existing traffic, but can help reduce how much worse the congestion gets over time. Better road maintenance (14) could reduce the need for the sort of major road repairs that back up exits and highways for days.

Options 6 through 8 also have relatively low costs for society, and promise somewhat greater reduction of existing congestion. Higher gasoline prices (6) would discourage driving at all times, with added benefits of reducing pollution and dependence on imported oil. Local Transportation Management Associations (7) can help coordinate and stimulate ride- sharing. By keeping density in new growth areas above a minimum level (8), mass transit and ride sharing remain viable, whereas sprawl commits everyone to the car.

Which brings us to the top five options. Option 3, clearing traffic accidents more rapidly, is undoubtedly the least controversial. The proposed method is to have a network of roving vans already on the streets, ready to dispense gasoline, do minor repairs, or just get a car off to the side of the road. If every stalled vehicle or accident was cleared 5 or 10 minutes sooner, the effect over a year would be substantial.

Choices 2, 4, and 5 go after free parking at work. One might install a parking surcharge for those who parked during rush hour (2). Or treat free parking as income and tax it (4). Or have employers offer a "commuting allowance" (5) that workers could either use to pay their parking fee, or cover other commuting expenses like mass transit or ride-sharing. In each case, the pressure is for commuters to shift to a different time, or travel by a method that doesn't require a parking space.

The top choice on Downs's list is to charge tolls during the rush hour. His preferred method would be an "electronic number plate," which sticks under cars. Electric sensors buried under the road would keep track of cars as they passed over, and send a computerized bill automatically. Enforcement could be through a system of cameras that would photograph the license plate of any car that didn't set off the sensor.

Technology of this sort was tested extensively in Hong Kong during the early 1980s, and worked well. In 1991, Congress passed legislation calling for trial programs of this sort in the United States, but no area has yet volunteered for the role of guinea pig.

The top two options on Downs's list -- tolls for driving or charges for parking during rush hour -- are choices only a mother and an economist could love. They face two main challenges: possible harm to the poor, and public outrage.

I suspect that much of the concern for the poor is disingenuous; you won't see the same tender concern for the poor when someone wants to build affordable high-density housing in an existing neighborhood.

Whatever the motives of some critics, it remains true that a higher price for anything affects the poor disproportionately. But the appropriate way to help the poor is with with increases in food stamps, rent subsidies, or welfare payments. If a fee is otherwise a good idea, arguing that no one should have to pay it because a minority of people are poor is just sloppy thinking.

Charging tolls or parking surcharges during peak driving hours seems outrageous to some people, but I think the main reason is simply that they aren't used to it. After all, most people understand the basic concept that prices are higher in peak times: phone companies charge more between 9 and 5; movies offer afternoon matinees; airlines charge more to heavy business travelers.

In fact, parking is often cheaper already on weekends and evenings than the weekday. The only new step here is to charge extra to those who drive or arrive during the rush hour.

The disbelief and near-hysteria that are sometimes evoked by peak-load tolls and surcharges seems oddly misplaced to me. After all, people already pay for traffic jams in time and aggravation, and claim to hate it. So why not pay the fees that would reduce congestion?

Moreover, some of the same people who claim to be willing to pay higher taxes to build mass transit or more roads, simultaneously oppose paying a rush-hour toll or parking surcharge. If you're willing to pay extra to avoid congestion, why not choose a method that avoids the perils of Downs's law?

There is no single pain-free magic miracle cure for traffic congestion. But taking the plunge on rush hour tolls or parking surcharges, and trying a bit of everything on a regional basis, at least offers the hope of spending more time with friends and family, and less looking at a stranger's rear bumper.

1. Peak-hour tolls on main roads
2. Parking tax on peak-hour arrivals
3. Remove traffic accidents rapidly
4. Tax employee parking benefits
5. A deductible commuting allowance
6. Increasing gasoline taxes
7. Form Transportation Management Associations
8. Keep density in new growth areas above minimum
9. Promote working at home
10. Change laws that discourage working at home
11. Better planning of traffic flow
12. Staggered working hours
13. Cluster high-density housing near public transit
14. Improve highway maintenance
15. Upgrade city streets
16. Build added high-occupancy vehicle lanes
17. Build new roads
18. Improve public transit service and amenities
19. Build or expand off-road transit systems
20. Concentrate jobs in clusters
21. Improve the jobs-housing balance
22. Increase automobile license fees
23. Adopt local growth limits

Source: Anthony Downs, "Stuck in Traffic," 1992

<< Back to 1992 menu