RAND's Paul Sorensen Authors Study Analyzing Short-Term Fixes for Congestion Relief

In October, you and a team of RAND researchers completed a report entitled "Moving Los Angeles: Short-Term Policy Options for Improving Transportation," which details the worsening problem of congestion in Southern California. Who was your client for the study? What was the RAND study's focus?

James Thomas, who is a member of one of RAND's advisory boards and who has a long-standing interest in traffic congestion, kicked off the study. He requested that we look at ways to improve the current traffic situation. The project was also funded in part by LA Metro and the Los Angeles Music Center, and RAND contributed to the study as well. The goals, as they were framed, were to look at what strategies Los Angeles might pursue in the short term to relieve traffic congestion. We defined "short term" as signifying strategies that could be implemented and produce notable effects within a period of about five years or less.

The RAND study begins by characterizing L.A.'s congestion, then diagnosing, and finally offering short-term recommended options for improving mobility. How did the study characterize the congestion in metropolitan L.A.?

We looked at some existing metrics that are commonly cited. For example, the annual report by the Texas Transportation Institute examines congestion characteristics across all large American cities over a span of about 25 years. That gave us a sense of how Los Angeles compares against other cities and how it has evolved over time. We also looked at more detailed data on the road networks in Los Angeles so we could understand spatially where the problems tend to be most severe. For that, we looked at PEMS (Performance Evaluation Monitoring System) data, which is developed and maintained at UC Berkeley and relies on sensor data embedded in the freeway system by Caltrans. We looked at data going back five or six years so we could understand how traffic flows on the freeway network have been changing in recent years. We also took a look at some of the modeling data that Metro has of traffic conditions on the surface streets. Putting all of those together, we were able to get macro interpretations of traffic congestion in L.A. and a detailed look at how that plays out in different areas across the county.

Traffic congestion in Los Angeles is well appreciated. How did the report characterize L.A. traffic?

We found that on the freeway network traffic congestion is fairly ubiquitous. There are certainly some hot spots, for example, on the 101, coming out of Downtown Los Angeles or on the 5 Freeway heading into Orange County. In general, though, you're going to find fairly severe traffic congestion spread throughout much of the freeway network. We also saw that the level of congestion increased considerably between 2001 and 2006, which was the last full year for which we had data. We saw many places on the freeway network where the daily hours of congested travel conditions were growing by one, two, or three hours per day, just over that six year period.

When we examined the surface street network we again saw congestion spread broadly throughout Los Angeles. But, there was a more dominant concentration of traffic congestion on the surface streets between Downtown L.A. and the Westside. We've often heard that traffic congestion is especially bad on the Westside. I would say that we have a "yes" and "no" answer to that proposition. "No" in the sense that it's pretty bad all over on the freeway network, but there does seem to be an even higher level of congestion on the surface streets on the Westside.

The RAND study moved from characterizing L.A. traffic to engaging in diagnosis. What did the report conclude?

There are several points worth mentioning. In general, where you have higher levels of population density, you're going to tend to have higher levels of traffic congestion. With higher levels of population density, per-capita vehicle travel tends to decrease a bit. Often, origins and destinations are closer together, so some car trips will be shorter. Other trips can be made by public transit, biking, etc. So each person may be driving a bit less in denser areas. On the other hand, when you have a higher population density, you also have far more people competing for the same road space within an area, and this exacerbates congestion.

The city of Los Angeles is not especially dense as big cities go. But, when you look beyond the city boundaries and consider the Greater Los Angeles region, we are by far the densest of any urban area in the country. The reason for that is the suburbs that surround Los Angeles have, on average, much higher population density than the suburbs around most other big cities. As an example, if you take the city of New York and the suburbs around it that make up the greater urban New York area, the surrounding suburbs are about one tenth the density of New York City itself. In contrast, if you take the city of Los Angeles and look at the suburbs that make up the Greater Los Angeles region, the surrounding suburbs are about three quarters of the density of the city of L.A.

Angelenos also drive more than one would expect, given how dense the region is. As I mentioned, you tend to see that per-capita vehicle travel decreases as population density increases. But in fact Los Angeles stands as a significant outlier; we drive as much as people in much lower density regions like Atlanta, Houston, and Dallas. The combination of high regional population density and relatively high per capita travel makes for really severe traffic congestion.

The report included a number of short-term options for mitigating Los Angeles' congestion, including high-occupancy toll lanes, variable curb-parking rates, and mandatory transportation demand management programs. Elaborate on these RAND report recommendations?

The idea with the HOT lanes is to use variable tolls to ensure free flowing travel within one or two lanes on a given freeway. Depending on the implementation details, carpools may be able to use those lanes for free, or they might pay a reduced toll. Drivers of single occupant vehicles would have to pay a toll for using those lanes, and the toll will get more expensive during the peak hours to make sure that those lanes remain free flowing. There are several advantages to this type of arrangement. One advantage is that it expands the choices available to travelers. There will still be free general-purpose lanes that you can use if the value of your travel time is not that high. In cases where you need to be somewhere in a hurry, and it's worth it for you to pay more to ensure that you get there on time, that option is available. Another advantage of HOT lanes is that they provide an opportunity for much faster bus transit service on the freeways. When you price those lanes to keep them free flowing, buses that use the lanes are able to travel at free-flowing speeds. This would provide a competitive advantage for transit service against driving alone.

A final point to make about HOT lanes is that when you have free flowing travel you're able to move far more vehicles per lane, per hour, than if you have heavy congestion conditions. By putting HOT lanes in place you're actually getting more capacity out of the existing roads that we have. On State Route 91 there are four lanes in the middle, two in each direction, that have variable tolls to keep them free flowing. On the outside there are four lanes in each direction, which are free. During the peak hours those lanes on the outside move somewhere between 15-20 mph and carry about one thousand cars per lane, per hour. The priced lanes in the middle move at about 65 mph and carry closer to two thousand cars per lane, per hour. You're basically getting double the productivity on a per-lane basis by using prices to keep lanes free flowing.

The motivation for variable curb parking rates is that in a lot of our urban areas we under-price curb parking. In other words, at the price we charge, there are far more people who would like to park than there are spaces available, and as a result the spaces are routinely scarce. This sets up a dynamic where cars often cruise around the neighborhoods, looking for an open space. All of this cruising for parking creates a lot of traffic. One report showed that as much as 30 percent of the traffic in a typical urban commercial district is from people looking for parking. The idea with variable curb parking rates is to raise the rate enough such that on any given block you would typically have one or two spaces available. That way, any car that is coming and looking for a parking space could find one and not need to drive around looking for a space. This gets rid of quite a bit of traffic within commercial districts. It also provides additional revenue to cities, which they can turn around and invest in a variety of ways, including subsidizing better transit in the area. Incidentally, the rate needed to make sure that there will usually be one or two spaces available on a block will typically vary by location and time of day, so the meter rates would be allowed to vary as well.

The final strategy you mentioned is employer programs to provide different commute options. The report didn't recommend mandating that employers put these programs in place. Our recommendation is that organizations such as Metro increase resources devoted to outreach to firms to help them provide programs like vanpools, telecommuting, and options other than driving to work alone everyday. We didn't suggest making this program mandatory, in part, because there is currently a state law that prohibits localities from mandating employers to provide trip reduction programs. We already do voluntary commuter programs, like carpools, throughout Southern California, and we've had some modest success. At the time the report was coming together gas prices were quite high. Lots of people were looking for other commute options. Our thinking is that there is now a much more receptive audience and employees really struggling with the cost of getting to work by driving long distances in their cars. This seems like a good time to revisit and renew our efforts in that area.

RAND's report is being reviewed by reformers and agencies like Metro, the FastLanes program, and the LAEDC. As an author, is the report being used properly as a basis for thinking about actions that could be taken?

The short answer to that question is that it does seem to be being used in a context that we would hope for. The city of Los Angeles has not done a strategic transportation plan for a number of years, but it is something on their agenda. We've received indications that our report and its recommendations are being given serious consideration as Los Angeles thinks about its strategic directions. Many of the recommendations we made, which five years ago might have seemed quite radical, are already being planned or developed by agencies here in Los Angeles. For example, Metro is developing HOT lane demonstration projects on the 110 and the 10 freeways. The city of L.A. has begun to put into place what's known as multi-space parking meter technology, which would allow for variable curb parking rates. In those contexts we also see a potential utility of this report in terms of providing more public education about some of these strategies and hopefully, in some cases, providing additional support and explanation of why these are good directions to pursue.

What's the significance of the timing of this? Are we at an inflection point here in 2008 with a new federal administration coming in? What does the timing have to do with actually getting the results we desire?

I don't know that the timing of the report was planned in particular as being an especially strategic moment. But the report does come at a very interesting time in transportation. We have recently seen very high fuel prices that have created some demand for better alternatives here in the region. The report is coming out at a time when Metro has received federal funding to plan certain types of congestion pricing. We also have a new administration coming in and a federal reauthorization of the transportation bill coming within the next year. With Proposition R having passed we're now going to have a lot more transportation funding to work with in the coming years. Some of the strategies we recommend will raise money, but others will cost a significant amount. The report may provide some ideas as to how we could productively spend some of that Measure R revenue. Through fortunate circumstances, the report has been released while many interesting things are happening in transportation.

Has RAND done similar studies around urban transportation and congestion in other parts of the country or world, so we could compare and contrast these findings?

I wouldn't say that RAND has done a comparable study looking at a strategic set of strategies for another urban area in recent years. RAND, as an organization, and some of the individuals here at RAND, have been fairly involved in looking at various forms of transportation finance, electronic tolling, congestion pricing, and these kinds of things. In fact, within the next few months, RAND, in conjunction with RAND Europe, located in Cambridge in the U.K., is going to be putting out what we're calling a "Handbook on Pricing for Decision Makers," which reviews evidence from some of these programs-where they've been implemented, whether they would make sense in a given setting, and some of the potential advantages and pitfalls. From that perspective, this project fits into an existing and ongoing body of work. But, this is the first time, at least recently, that we've taken such a broad look at what the options are for a specific urban area and tried to understand what would make the most sense.

The voters of Santa Monica, where RAND's headquarters are located, took a slightly different approach regarding congestion by putting the RIFT initiative on the ballot in November. It was defeated, but nevertheless speaks to the frustration residents have over the problem of traffic. Talk a bit about the RIFT initiative, and whether that would have been a reasonable short-term way of dealing with congestion.

RIFT represented an effort to stem the growth in traffic congestion by placing restrictions on large commercial development projects in Santa Monica. We did not examine this idea within our report. As I mentioned earlier, our goal was to look at short-term strategies, options that might be implemented and produce notable effects within a five-year timeframe. While land use strategies such as RIFT could be implemented quickly, the effects on traffic would evolve much more slowly, over a period of decades, as land use patterns gradually change. So strategies like RIFT fell outside of the scope of our study.

That said, the broader analysis presented in the report suggests that the effectiveness of RIFT, had it passed, would have been limited. There are many factors – population growth, increased income levels, and economic activity – that contribute to increased vehicle travel. Even though we're currently experiencing a significant economic downturn, most experts expect that population, income levels, and economic activity within the region will grow over the coming decades. Though RIFT might reduce the pace of economic development in Santa Monica, it would have little effect on population growth or income levels. As a result, it would likely fail to stem, let alone reverse, the growth in local traffic congestion.

Another important point is that urban traffic congestion is a regional phenomenon, and Santa Monica does not exist in isolation. If there is pressure for increased commercial development on the Westside, and if that development is not accommodated in Santa Monica, then the development could certainly occur just outside of Santa Monica's borders. This could generate additional traffic that flows through Santa Monica, even if the developments are not located within the city's borders.

Traffic congestion is a complex problem, and land use strategies will certainly play an important role in addressing this challenge over the longer term. But an effective approach will require both land use and transportation strategies that are thoughtfully integrated – for example, concentrating new development in areas with high quality transit options. Because traffic flows are regional, it will also be necessary for cities to work more closely with one another to coordinate their land use and transportation strategies.