It surely isn't getting better. Nonetheless, it is still far from intolerable for residents in many regions, according to the Texas Transportation Institute. The institute knows this because it has developed a measure of congestion for each region called the travel time index. The index calculates the number of additional minutes needed to make a trip during peak travel periods rather than other times of the day, when traffic is presumably flowing freely. The higher the travel time index, the greater the congestion.
In the 68 metropolitan areas studied by the Texas Transportation Institute in its 2001 Urban Mobility Report, the average travel time index moved up from 1.07 in 1982 to 1.58 in 1999. Thus, a trip that should take 20 minutes without undue traffic was taking 21.4 minutes in peak hours during 1982 and 31.6 minutes during 1999. The 10.2- minute difference represents a gain of 47.7 percent.
The next step in using TTI's data was to discover what affects this index--and thus congestion. I set up a regression analysis using the index as the dependent variable and factoring in several relevant independent variables. What I found is that there are three independent variables with statistically significant impacts: One is the absolute change in a region's population between 1990 and 2000, with larger increases causing higher index scores. The second is the 1990 population density of the region or its central city, with higher densities corresponding to higher index scores. The third is the location of the region, with a location in the Northeast corresponding to a lower index score.
Surprisingly, the percentage gain in a region's population between 1990 and 2000 has no significant effect. It's absolute gains that matter, and these are likely to be greatest in the largest metropolitan areas, even if those areas are growing slowly. Another surprise is that a region's total population is statistically significant only when coupled with the density of the overall urbanized area, not with the density of the central city.
What these findings indicate is that congestion is likely to be worst in very large metropolitan areas--areas that normally have the highest central city densities and the largest absolute population gains--and in somewhat large areas with fast growth that produces absolutely large population gains. Las Vegas would be an example of the latter. Similarly, congestion is likely to be least intense in all small metropolitan areas and in all non-metropolitan areas. In between these two groups are medium-sized metropolitan areas and large ones that did not have big absolute population gains between 1990 and 2000.
If an absolutely large population gain is defined as an increase of 250,000 or more, then the total populations of these groups can be determined. As of 2000, the metropolitan areas in the most congestion- prone group contained 101.6 million residents, or 36.1 percent of the nation's total. The least congestion-prone group contained 105.3 million residents, or 37.4 percent. The intermediate-congestion group contained 74.5 million residents, or 26.5 percent.
Of course, many other factors besides the ones tested in these regressions affect the intensity of congestion in each region. But this analysis shows that a large fraction of Americans are not experiencing severe traffic congestion. This conclusion is borne out by the fact that the 1999 travel time index for the smallest metropolitan regions in the TTI's 2001 study was 57 percent less than in large regions, or an index of 1.13 in small areas compared with 1.77 in very large regions.
The fact that many Americans are not actually experiencing bad traffic congestion is confirmed by a Gallup telephone survey exploring commuting behavior and experiences with traffic congestion. The survey, taken in May 2000 of 1,032 randomly chosen American adults, found that about 58 percent drove to work each day and that the average commute was 26 minutes each way, compared with 23.7 minutes in 1998. Among the 601 respondents who drove, 19 percent thought the traffic they encountered was "a major inconvenience and problem," and 31 percent thought it was a "minor inconvenience and problem." But 48 percent thought it was "not a significant problem," and 2 percent had no opinion. So just half of the drive-to-work set regarded congestion as a day-to-day annoyance.
There were further breakdowns: 11 percent said they got stuck in traffic jams every day; 14 percent, several times a week. But 22 percent said they were stuck only a few times a month and 31 percent said it happened only a few times a year; 20 percent said they never experienced a traffic jam. Among all 1,032 respondents, 69 percent said they had not changed their lives or schedules in any way because of traffic in recent years.
Thus, for a large percentage of Americans, traffic congestion isn't so bad after all--at least not yet!
