Skip to contentUnited States Department of Transportation - Federal Highway Administration FHWA HomeFeedback
Office of Transportation Policy Studies

FHWA > Policy > Office of Transportation Policy Studies > Transportation Futures


< Previous | Table of Contents | Next >

Chapter IV. Variables of Interest

In addition to the standard explanatory variables of travel behavior, we hypothesize that the travel of youth will be uniquely affected by four different factors—high youth unemployment rates, changes in household structure (in part related to high youth unemployment), increased use of information and communications technologies, and the adoption of state graduated driver’s licensing regulations. We discuss each of these factors below and present descriptive data for these variables from the three national travel surveys.

A. Unemployment Rates

The recession that began in 2008 hit teens and young adults particularly hard. As Figure 4 shows, youth unemployment rates have always been higher than non-youth unemployment rates; however, the gap between the two has widened over time (U.S. Department of Labor, various years). In 2008, one in five young workers (16-24) was unemployed, almost three times the unemployment rate of older individuals (25+).

Figure 4: Unemployment Rates by Age Group (1948-2012)

A line graph plots values in percent over years for two age groups. The plot for the age group 16 to 24 shows an oscillating pattern along the value of 4 percent for the years 1948 through the mid 1960s, then a drop to nearly 2 percent in the late 1960s, a swing upward to a value of 7.5 percent in the early 1980s, a drop to 3 percent by the late 1990s, and then an upward trend ending at 7 percent by 2012. The plot for the age group 25 and above has sharper fluctuations, from an initial value of 7 percent in 1948, dropping to 6 percent in the early 1950s, with a peak at 13 percent in the early 1960s, a valley at just over 8 percent in the late 1960s,  another peak at nearly 18 percent in the mid 1980s, a valley at nearly 9 percent by the year 2000, a peak at just over 18 percent approaching the year 2010, and dropping to just above 16 percent by the year 2012.

Similarly, data from the national travel surveys show that the share of youth employed dropped dramatically between 2001 and 2009, particularly for the youngest respondents in our sample. As Figure 5 shows, fewer than 20 percent of 16-year olds in 2009 had a job, while the figure was more than double that in 2001.

Figure 5: Percent Working By Age and Year (1990, 2001, and 2009)

A line graph plots values for three age groups in percentage working over respondent age. The plot for the year 1990 is a curve from zero percent for age 15 swinging up to a value of about 79 percent for age 20 and flattening to a value of 75 percent for age 26. The plot for the year 2001 is a curve from 10 percent for age 15, swinging up to a value of 82 percent for age 19 and age 20, and flattening to a value just above 75 percent for age 26. The plot for the year 2009 is a curve from zero percent for age 15 swinging up to a value of 75 percent for age 20, trending down to a value just below 70 percent for age 23, and then oscillates along and above this value to end at 80 percent for age 26.

High youth unemployment rates may well explain changes in the travel of youth. The current recession directly contributes to a decline in work travel and, consequently limits the resources that youth have available to travel for other purposes (such as shopping and recreation). The recession may also make it more difficult for youth to afford owning and operating automobiles thereby increasing their travel by alternative modes such as public transit and walking.

B. Household Structure

The downturn in the economy may also influence youth travel patterns through household structure. Studies show that young adults tend to live at home for their own benefit, particularly when they are young, unmarried, and have low-incomes (Di et al., 2002; Kreider, 2007; Messineao and Wojkiewicz, 2004). Further, as discussed above, young adults who face a difficult labor market often “boomerang” back to live with parents. Wiemers (2011) finds that individuals who become unemployed are twice as likely to move in with others, with young adults the most likely to move into shared living arrangements. Kaplan (2009) finds that youth (16-23) who do not attend college frequently move back home and that moves from employment to non-employment substantially increase the likelihood of moving home. Youth who live at home can take advantage of their parents’ resources including, perhaps, their automobiles.

Figure 6 illustrates the dramatic increase (58%) in the proportion of young adults who live with their parents in 2009. Figure 7 illustrates the percentage of young people living at home by age for each of the three decades. The data indicate that a higher percentage of young adults lived with their parents in 2009 compared to previous survey years. In 1991 and 2001, there was a sharp decline in living with parents between the ages of 18 and 19. The decline is far more gradual in 2009 and it takes until age 23 to achieve the same percentage of youth living with parents than we observe for 19 year olds in the earlier decades. The graph also shows that in 2009 almost 10 percent of young teens (15-17) did not live at home with their parents.

Figure 6: Percent of Young People (19–26) Living At Home with Parents (1990, 2001, and 2009)

A vertical bar graph plots values in percentage for each year. In the year 1990, the breakdown is 67 percent not living with parents versus 33 percent living with parents. In the year 2001, the breakdown is 69 percent not living with parents versus 31 percent living with parents. In the year 2009, the breakdown is 42 percent not living with parents versus 58 percent living with parents.

Figure 7: Percent of Youth Who Live at Home with Parents by Age and Year (1990, 2001, and 2009)

A line graph plots values in percentage over respondent age for each year. The plot for the year 1990 shows a drop from a value of 100 percent living at home with parents for age 18 to a value of 60 percent for age 19, a value of 30 percent for age 22 and on to a value of about 12 percent for age 26. The plot for the year 2001 shows a drop from a value of 100 percent living at home with parents for age 18 to a value of 60 percent for age 19, a value of 30 percent for age 22, and trailing off to about 12 percent for age 26. The plot for the year 2009 shows a drop from an initial value of 90 percent living at home with parents for age 17 to a value of 70 percent for age 20, swinging down to a value of about 32 percent for age 26.

C. Use of Technology

Youth tend to be early and frequent adopters of communication technologies (Mans et al., 2011; Lenhart et al., 2005; Pew Research Center, 2010b). Figure 8 shows the use of technology (daily web use) by age in 2001 and 2009. Even in this relatively short period of less than a decade, internet usage increased substantially across all ages. While web use is high among all ages, the data suggest that youth and young adults are slightly more likely to use the web daily compared to older adults.

Figure 8: Use of Technology by Age (2001 and 2009)

A line graph plots values for percentage that uses the web daily over age of respondent for each year. The plot for the year 2001 has an initial value of about 42 percent for age 16 years, a jump to a value of 50 percent for age 17 followed by a drop to a value of about 38 percent for age 19, oscillating to a peak at a value of nearly 55 percent for age 27, then trailing off to end at a value of 27 percent for age 62. The plot for the year 2009 has an initial value of 70 percent for age 16 years, a jump to a value of 75 percent for age 18, oscillating to a valley at a value of about 62 percent for age 32, a peak at a value of about 75 percent at age 37, then trailing off to end at a value of 55 percent for age 62.

The rapid profusion and adoption of new communication technologies may influence how people use their time and, therefore, how much they travel (Kwan, 2002). The literature on information and communications technology (ICT) and travel among adults is inconclusive, although a majority of studies find that it is complementary to travel by potentially generating additional travel (Ferrell, 2004; Mokhtarian, 2000, 2002). Viswanathan and Goulias (2001) find high levels of person-to-person variance in the relationship between mobile technology and travel suggesting that individuals use technology in different ways; some individuals use technology to consolidate trips and others use it to reach activity opportunities that require longer travel times. Thus far, studies on this topic have not focused on youth in particular; therefore, the effect of ICT on youth travel behavior remains unknown.

D. Graduated Driver’s Licensing Regulations

Stringent licensing regimes may suppress travel if young people can no longer become drivers. On the other hand, it is possible that young people find alternative modes to meet their daily travel needs, such as getting a ride with a parent or older friend, using public transit, walking, or using a bicycle.

Table 9. Number of States per License Rating (1990, 2000, and 2008)
Ranking 1990 2000 2008
License Stringency: Lowest 42 13 -
License Stringency: Low 9 19 9
License Stringency: Medium - 14 12
License Stringency: High - 3 30

Table 9 presents the number of states in each ranking category in each year. As discussed above, we use licensing information for the year preceding our year of interest. Variation in licensing regulations by state was highest in 2000. In 1990, by contrast, the majority of states still had not begun to introduce graduated driver’s licensing regulations (GDL) and by 2008, most states achieved the highest ranking from the IIHS. The licensing regimes do not follow a strict urban-rural pattern. The states with the lowest ranking (Marginal) in 2008 were mostly rural states: Arkansas, Idaho, Kansas, Mississippi, Montana, New Mexico, North Dakota, North Carolina, and South Dakota. However, many other rural states also had the highest ranking in 2008, including West Virginia, Nevada, Oklahoma, and Utah.

The data show that driver status varies by age and licensing stringency. We can see from Figure 9 that, compared to young adults, a smaller proportion of teens are drivers because many are not legally able to acquire a license. The data also show that 16-year-olds in 2009 were less likely to drive compared to 16-year-olds in 2001; however, they were more likely to drive compared to 16-year-olds in 1990, a period where there were relatively few regulatory restrictions on driving. Unfortunately, the 1990 survey did not ask 15-year-old respondents about their driver status and, therefore, we must rely on our best estimate of driver status, which, as we discuss in Chapter III, substantially underestimates the proportion of 15-year-old drivers.

Figure 9: Percent Drivers by Age, 15–22 (1990, 2001, and 2009)

A vertical bar graph plots values in percentage driver over respondent age for each year. The plot for the year 1990 shows an initial value of about 12 percent for age 15, a sharp increase to a value of 70 percent for age 17, slow steady increases to a value of 89 percent for age 21, and trailing slightly lower for age 22. The plot for the year 2001 has an initial value of 20 percent for age 15 and a sharp increase to a value of 55 percent for age 16, followed by steady increases to a value of 80 percent for age 19, and trending along this value to end at 85 percent for age 22. The plot for the year 2009 has an initial value of 22 percent for age 15, followed by a steady increase to a value of 81 percent for age 19, and trending slowly upward to end at a value of 90 percent for age 22.

In Figure 10, we see there are fewer young drivers in states with tough licensing regulations, but by age 22, young people in states with the strictest licensing regulations are just as likely as young people in states with less strict regulations to have a license. This finding suggests that the introduction of more stringent licensing regulations has only a short-term effect on driver status.

Figure 10: Percent Youth with Drivers License by Age when License Restrictions are Lifted (2009)

A line graph plots values in percentage licensed drivers over age for three driver categories. For the category lifted age 16 to 16.99, the curve has an initial value of 40 percent for age 15, increases steadily to a value of 85 percent for age 17, trends upward to a value of 91 percent for age 19, and trails off to end at a value of 90 percent for age 22. For the category lifted age 17 to 17.99, the curve has an initial value of 20 percent for age 15 and swings upward to a value of 86 percent for age 19, and oscillates slightly to end at a value of 90 percent for age 22. For the category lifted age 18, the curve has an initial value of 20 percent for age 15 and swings upward to a value of 79 percent for age 17, and trends upward to end at a value of 90 percent for age 22.

Finally, Table 10 presents the proportion of people in each age group who are drivers disaggregated by license regulations ranking. For teens, more stringent driver’s license regulations are, in general, associated with a lower proportion of respondents who are drivers in all three years. This finding conforms to expectations because we might expect stricter standards to act as a barrier to driver status or as an incentive to delay licensing. Contrary to expectations, however, the relationship between license stringency and proportion of drivers also materializes, albeit to a lesser extent, for adults and young adults in some years. Yet we do not expect graduated licensing regulations to affect the licensing decisions of young adults or adults.

In Figure 11, we can more clearly see how the proportion of drivers has changed over time within each age group. Again, we caution the reader regarding the interpretation of the proportion of young drivers in 1990 because the measure is undoubtedly an underestimate.13 For this reason, we restrict our comparison of teen drivers to 2001 and 2009 where we see an increase in the proportion of teen drivers within each licensing regime over time.

Table 10. Percent of Drivers by Age Group and Strength of Licensing Regulation (1990, 2001, and 2009)
  1990 2001 2009
Teen (15–18)
License Stringency: Lowest 59% 58%  
License Stringency: Low 41% 61% 70%
License Stringency: Med   54% 55%
License Stringency: High   40% 54%
Difference
(lowest observed-highest observed)
18 18 16
Young Adult (19–26)
License Stringency: Lowest 92% 91%  
License Stringency: Low 82% 93% 83%
License Stringency: Med   85% 86%
License Stringency: High   85% 87%
Difference
(lowest observed-highest observed)
10 6 -4
Adult (27–61)
License Stringency: Lowest 95% 96%  
License Stringency: Low 91% 96% 94%
License Stringency: Med   94% 94%
License Stringency: High   93% 93%
Difference
(lowest observed-highest observed)
4 3 1

Figure 11: Percent Drivers by Licensing Regime and Year, Teens 15–18 (2001 and 2009)

A horizontal bar chart plots values for teenage drivers in four categories of license stringency. For the year 2001 the share of teen drivers in the lowest stringency category was 58 percent; in the low stringency category, 61 percent; in the medium stringency category, 54 percent, and in the high stringency category, 40 percent. For the year 2009, the share of teen drivers in the low stringency category was 70 percent; in the medium stringency category, 55 percent, and in the high stringency category, 54 percent.


13Among respondents aged 16-18 in 1990, 68 percent of respondents in the lowest stringency category and 54 percent in the low stringency category report being drivers.
< Previous | Table of Contents | Next >


FHWA Home | Feedback
FHWA