Intercity Passenger Rail in the Context of Dynamic Travel Markets (2016)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

85 8.1 Introduction Any study of the role of rail in a dynamic, competitive marketplace must include an analysis of how the recent evolution of intercity bus has changed the nature of the competition. And yet, the intercity bus, despite regaining prominence in many major U.S. corridors over the past decade, remains only sporadically researched and poorly understood by many transportation analysts. Much of the analysis conducted by public agencies to assess improved rail service ignores the sector’s changing characteristics and market penetration. The absence of com- prehensive data on prices and passenger volumes remains a major impediment to analysis of bus–rail competition. 8.2 Previous Research on Intercity Bus Service This NCRRP project provides a variety of new forms of information about the market charac- teristics of the intercity bus industry. To some extent it builds upon an existing mass of research, but, in reality, research on the causes and effects of the industry’s ups and downs has been com- paratively thin, in part due to a lack of data. The federal government does not maintain numbers on ridership or fares, as are available for air and train travel. Carriers are not required to report ridership or passenger-mile data to a centralized authority, and Greyhound ceased systematically reporting its ridership several years ago. In the last American Travel Survey, conducted in 1995, it was estimated that scheduled inter- city buses accounted for around 40 million passenger trips annually. How much the sector has grown since then, however, remains unknown. Estimates by Schwieterman, Schulz et al. (2015) put the number close to 55 or 60 million (This estimate is based on the Chaddick Institute’s Intercity Bus Data Set, which applies load factors to estimates of the number of daily bus opera- tions on approximately 100 intercity bus lines). While scarce compared to other modes, some research does exist on this industry, including an important study by the US General Accounting Office in 1992. A systematic analysis by the University of Delaware evaluates issues posed by the expansion of curbside service and safety issues (Scott et al. 2013). A much wider body of research explores rural and state-supported services (KFH Group Inc. 2002, Higgins et al. 2011). State governments have also been quite active in studying service levels within their jurisdictions. Comprehensive passenger surveys were undertaken in Florida, Michigan, and Minnesota (Florida DOT 2009, Grengs 2009, SRF Consulting Group 2010). This research documents the extent to which bus service, particularly service on rural and quasi-rural routes, caters to travelers without access to private vehicles. It also clearly shows that passengers using buses tend to be younger and have less income than the general population. C H A P T E R 8 Competition to Rail from Intercity Bus

86 Intercity Passenger Rail in the Context of Dynamic Travel Markets 8.2.1 Bus Diversions from Rail? Research by Schwieterman, Schulz et al. (2015) shows that the expansion of city-to-city service has resulted in an industry—dominated by BoltBus and Megabus—which now accounts for about 1,000 daily bus operations. Passenger surveys administered as part of this NCRRP research show that the expansion of this service has drawn passengers away from rail travel in both the East and Midwest, but to a much higher extent in the former (Table 30). Although 34% of travelers in the East reported they would have taken the train if the services of these bus lines were not available, only 21.9% indicated this in the Midwest, where the diversion from air service was much higher. Passengers were surveyed in 2011 in Chicago, Indianapolis, New York, Philadelphia, and St. Louis (Schwieterman and Fischer 2012). This same research showed that only about 17% of travelers indicated “business” as their trip purpose, with the rest indicating the purpose as “pleasure” (58%) or “personal business” (26%). 8.2.2 Understanding the Role of Price Schwieterman, Kohl et al. (2014) also documented the price differences between modes in 2013 by evaluating fares in 55 city-pairs with travel distances between 90 and 400 miles. [Thirty-five of these city-pairs were (and continue to be) served by both Amtrak and city-to- city express bus lines. The analysis considers fares in three advance purchase scenarios—1 day, 7 days, and 28 days—on two different travel dates, with the lowest fare in the 3-hour period of the day being recorded.] The results show that the lowest bus fares over a designated 3-hour period were generally about 33% to 60% less than train tickets. Rail fares, in turn, undercut air fares (even with the assumption that travel is roundtrip with a Saturday minimum stay) by more than half. The difference between rail and bus fares is much larger in the NEC than in other parts of the country, particularly in high-frequency Amtrak corridors. To better understand how recent changes have affected fares, fresh data were collected for this project in late June and July 2015. The results show that price differentials remained relatively con- stant between 2011 and 2015. Examples of fare and scheduled travel time differences are shown in Table 31. On most routes, buses are less expensive but markedly slower than trains. The difference in travel time would likely be greater if delays induced by traffic congestion were taken into account. For the 35 corridors served by both bus and rail, the analysis showed: • Average bus fares purchased 7 or 28 days in advance are about $30 less than train fares. In percentage terms, advance purchase bus fares are, generally, 30% to 60% less expensive, although the difference is less in markets without discount city-to-city express carriers. Rail fares remained less than half of airline fares. How would you have traveled had this service not been available? East Midwest Total Air 6.1% 15.2% 10.6% Drive 23.3% 31.9% 27.5% Greyhound/other bus lines 17.9% 10.2% 14.1% Rail 34.0% 21.9% 28.0% Other 1.1% 1.9% 1.5% Would not have traveled 22.0% 21.9% 22.0% Sample size 258 393 651 Source: Schwieterman and Fischer (2012). Table 30. Traffic generation and diversion of city-to-city express bus service.

Competition to Rail from Intercity Bus 87 • The differences between bus and trains fares are greatest in the NEC, where they are often in the $60 to $80 range. The differences are smaller, about $20, in the Midwest, and smaller still on the Portland–Seattle corridor, where the gap is around $10 to $17, or about 40% to 33% less. • In some instances, however, bus fares are higher than train fares. This is often the case, for example, on the Chicago–Indianapolis corridor. 8.3 NCRRP Attitudinal Model for Intercity Bus Travel This project has created a new structural equation model which helps fill the void in under- standing the dynamics of bus–rail competition. Of particular value are the STE coefficients, which show how the attitudes and preferences of a traveler affect her/his propensity to travel on the intercity bus. This section reviews these effects and illustrates the implications for policy makers and service providers. 8.3.1 Content of the Model A diagram of the full bus model is shown in Figure 47, with the latent factors representing the four longer-term values on the left-hand side of the diagram. These four latent factors reflect the use of 11 observed variables, as originally shown in Figure 36 in Chapter 4. The latent factors concerning short-term attitudes and the outcome factor are shown in Figure 48; they are different from the earlier rail model, as they concern the propensity to take intercity bus. 8.3.2 Explanatory Factors for Taking Bus The factors most important for bus travel, ranked with respect to their STE, are as follows: • A belief that bus travel is inconvenient • The extent to which a person has a strong auto orientation • A belief that bus travel is less stressful than driving • The value one places on privacy in travel • Concern that the bus trip has issues of personal safety A comparison of the STEs on the propensity to travel by bus and train appears in Table 32. Corridor Travel Time Advance Purchase Fare Train Bus* Bus Train 1 day 28 day 1 day 28 day Boston, MA – New York, NY 3:50 4:05 $107 $52 $25 $19 Chicago, IL – Indianapolis, IN 3:25 5:05 $24 $24 $29 $25 Chicago, IL – St. Louis, MO 5:30 5:30 $39 $39 $19 $17 Dallas, TX – Oklahoma City 5:10 3:58 $47 $37 $34** $24** New York, NY – Rochester, NY 7:35 6:38 $106 $62 $57 $38 New York, NY – Washington, DC 4:25 3:13 $123 $86 $25 $17 Portland, OR – SeaŽle, WA 3:50 3:40 $35 $27 $19 $18 *All bus fares shown are for either BoltBus or Megabus, whichever is lower, except for the Dallas–Oklahoma City corridor. Megabus and BoltBus fares are increased by $1.50 to reflect the booking fee they impose on one way or round trips. **Bus fares shown are for Greyhound Express and have been adjusted upward by $3.10, and rounded to the nearest dollar, to reflect the facility fees that carriers charge. (This was determined to be the average fee imposed for bus travel in the markets shown on the table.) Table 31. Travel times and minimum fares in selected corridors, June 29, 2015.

88 Intercity Passenger Rail in the Context of Dynamic Travel Markets Figure 47. Attitudinal Model for Bus based on results of the NCRRP 03-02 survey (2014, n  5,625). Figure 48. Four short-term attitudes and outcome factor based on observed variables for bus model.

Competition to Rail from Intercity Bus 89 8.4 Interpreting Results from the NCRRP Attitudinal Model for Intercity Bus Travel 8.4.1 Convenience and Cost The Attitudinal Model for Bus provides some new evidence about how perceptions of conve- nience and cost affect bus travel. Convenience On the one hand, perceived convenience of taking the bus ranks as the highest predictor of the propensity to take the bus. However, even though perceptions about how bus schedules are inconvenient have the greatest STE in the bus model, it is markedly lower than that for rail travel (STE of -0.46 and -0.73 for the bus and rail model, respectively). Young travelers, who, on average, earn less than older travelers and are especially apt to travel by bus, may consider the opportunity cost of a couple extra hours onboard to be far less consequential than the typical Amtrak rider. The extent to which the train is inconvenient is a far more powerful predictor of mode choice than is the extent to which the bus is inconvenient. While buses are often slower than trains, they go more places more directly, and more often. As discussed, the lower level of correlation between variation in perceived convenience with variation in choice with bus may reflect a lesser array of alternatives available for those who choose bus. But, for both modes, a perception that the service is inconvenient is a powerful factor in the prediction of mode selection. Perceived Cost The bus trip is seen as more expensive than the car trip by 33% of the NCRRP sample, while the train trip is seen as more expensive by 65%. For those who actually took the bus on their last trip, only 19% of them thought the bus was more expensive than the car. The propensity to travel by bus is less closely linked to perceptions about cost than is train travel, although the effect for both modes is small (-0.09 for buses and -0.15 for trains). Consis- tent with the observations about convenience for the bus user, the small standardized effect on Bus Rail STE Predictor* STE Predictor* 0.46 Inconvenient 0.73 Inconvenient 0.42 Less Stressful 0.34 Auto 0.39 Auto 0.33 Less Stressful 0.13 Privacy 0.29 Privacy 0.10 Unsafe 0.22 Unsafe 0.10 Educaon 0.19 ICT 0.09 Expensive 0.15 Expensive 0.06 ICT 0.09 Educaon 0.06 Urbanism 0.03 Employed 0.05 Density 0.03 Urbanism 0.01 Employed N.S.** Density *The four basic values are shown in italic bold; the four short term atudes are shown in roman; and demographics are shown in italic. **N.S. = not significant Table 32. Comparison of predictors between bus and rail model.

90 Intercity Passenger Rail in the Context of Dynamic Travel Markets bus travel may reflect a pervasive belief that bus travel is generally the lowest-cost option when compared to trains and planes. Further, it may also reflect the fact that those who do choose bus are more constrained in their options. Consequently, few passengers are avoiding bus travel based on the perception that it is expensive. Implications. The considerations about convenience and cost mentioned so far may have several notable implications: • Amtrak and other rail-passenger operators can invest in service improvements to attract passengers valuing convenience to a much greater extent than bus operators. Whereas most rail routes hold the prospect of improved speed, bus companies have fewer opportunities to improve service in this way. Most major bus routes already operate either with nonstop or limited-stop service at high frequencies. The maximum speed of travel is determined by federal law. Amtrak, however, can make investments to allow gradual improvements to convenience with respect to both speed and frequency. • To the extent that Amtrak has been vulnerable to a shift to buses in the largest markets, the shift has probably already largely occurred. Buses operate at least hourly service in most major NEC markets. For example, at least 60 buses operate daily in each direction between New York and Boston. Intercity bus frequencies are at least twice those of Amtrak in most routes in other parts of country. Between Portland (Oregon) and Seattle, BoltBus and Grey- hound together operate 14 trips daily, whereas Amtrak operates five. Further improvements to the frequency of bus service will likely have only marginal effects on the perceived convenience of bus service. • Rail operators may be vulnerable to the establishment of bus service to and from new urban and suburban locations that are closer to a traveler’s origin or destination. New bus services to and from “neighborhood” locations can attract convenience-oriented travelers by offering pickup and drop-off locations that are easier than more centralized stations. As a result, pas- senger rail faces a greater traffic loss as a result of the expansion of bus services on NEC routes from lightly served locations, such as Bethesda and College Park, Maryland, and Lorton, Virginia, rather than from central Washington, DC, which is already saturated with service. • The advent of premium bus services can make bus travel more attractive to passengers, although the market potential of these services remains unclear. Premium bus services are more costly to travelers but offer more spacious seating arrangements and make stops at luxury hotels and air terminals. The recent expansion by Vamoose Gold Bus, as well as Royal Sprinter’s luxury service in the New York–Washington market, fall into this category. The marketing efforts of these companies, which offer first-class style seating and have onboard attendants, tend to focus squarely on trying to win the loyalties of existing airline and rail pas- sengers. These carriers quite explicitly “sell” privacy to passengers who are wary of bus travel but can do little to eradicate the disadvantages of buses with respect to reliability. At present, these luxury carriers have only a small market niche, operating a handful of trips per day. 8.4.2 Privacy and Safety The results of the NCRRP 03-02 survey should be of concern for advocates of all public modes in long-distance travel: nearly half of those in the total sample reported that they worried about personal safety or disturbing behavior on the bus trip. This is considerably higher than that reported for either the train or air, both at around 30%—but even these are worthy of further exploration. By contrast, most people do not worry about disturbing behavior when they take a trip by car (Figure 49). Further, the overall level of concern is not strongly related to the propensity to use the bus. Table 32 shows that the effect of perceiving the bus as unsafe on the propensity to take the bus is

Competition to Rail from Intercity Bus 91 relatively small (-0.10), ranking fifth in the explanation of bus use. Perceptions of personal safety therefore have a much smaller effect on bus travel than rail travel (-10 and -0.22, respectively). Bus travelers are more likely to face severe budget constraints and may thus have few alternatives to bus travel. In the same way that bus riders, compared to train riders, are less likely to act on their concerns regarding perceived inconvenience, they are also less likely to act on their concerns regarding personal safety. Perhaps more encouraging is that the high overall level of concern for personal safety stems mainly from those who do not use the bus. For those who did report using the bus, their pro- pensity to worry about personal safety (Figure 49) was very similar to the general population’s level of worry about the train and bus (approximately 30%). Looking at the value of privacy in travel (derived from survey questions not specifically designed to elicit different responses between bus and rail), bus users have a somewhat lower need for privacy. They also report not minding traveling with people they do not know and report less discomfort when being with people they do not know (Figure 50). For the public, concerns about personal safety in bus travel can loom large. These differences might be explained in part by concerns over the loss of privacy created by the need to “rub elbows” with less-affluent demographic segments when traveling by bus. Anecdotal reports also suggest 0% 10% 20% 30% 40% 50% 60% Bus Train Air Car Worry about personal safety/disturbing behavior Figure 49. Concern about personal safety is highest for bus. Safety Concerns by Prior Mode Choice 0% 10% 20% 30% 40% 50% 60% 70% Worry about personal safety/disturbing behavior Bus The idea of being on a train or a bus with people I do not know is uncomfortable I don't mind traveling with people I do not know Bus on Last Trip Other Modes Figure 50. Individuals traveling by bus report less concern about personal safety.

92 Intercity Passenger Rail in the Context of Dynamic Travel Markets that many older travelers harbor unpleasant memories of their interactions with fellow travelers on Greyhound and Trailways trips years ago. These travelers may remember invasions of pri- vacy as one of the greatest perils of choosing this mode. However, as noted in Figure 50, these negative perceptions are in contrast to the evaluation of bus travel by those who actually took the bus recently. The higher density of seating on buses may similarly add to public concerns about the lack of privacy with bus travel (Seat width on Amtrak is generally about 23 inches compared to about 18 inches on major bus lines). Crowding at curbside locations and conditions at bus terminals may further contribute to the negative perception of bus travel. Implications. The high degree of concern about personal safety in the bus trip, and the more general desire for privacy in travel, could have several implications: • Bus ridership is negatively affected by the views—largely by those who have not experienced the bus recently—that the bus trip is unsafe. While the reported standardized effects are lower than might be expected, they are negative and statistically significant. • As reported earlier, the millennial generation is more likely to hold these views about privacy and personal safety than the older generations. Chapter 6 has explored alternative possible futures looking only at rail; the available results from the attitudinal modeling process suggest that such futures might also be relevant for the future of intercity bus. • Arguments that the quality of the hedonic experience at rail stations are not important are not supported by this analysis. Years of debate over the expansion of passenger service facilities at the Moynihan Terminal may have contributed to prolonging an unacceptable level of passenger experience at the prime New York City rail terminal, as was reported in the UVM TRC rural sample, where New York City was the only city explicitly considered in that study. 8.4.3 Stress of Driving and Auto Orientation Attitudes about cars and driving seem to affect decisions to travel by bus. Having a high level of auto orientation affects the propensity to travel by bus and train to a similar degree, but the perceived stress of driving has a slightly greater effect on a traveler’s propensity to take the bus than the train. (The standardized effect concerning auto trip stress for bus travelers is 0.42, compared to rail’s 0.33.) Understanding why the perceived stress of driving has a greater effect on bus travel requires taking into account the different market segments that those modes serve. As previously noted, prior studies have documented that bus travelers are younger and have less disposable incomes than the population as a whole. For these passengers, the stress associated with driving may include (1) the difficulty of finding an available car (or renting one), (2) driving on an unfamiliar route, and (3) dealing with the complexities of parking the car in environments such as campuses and downtown metropolitan areas that are less conducive to this than rural and suburban areas. The appeal of bus travel to passengers who consider driving stressful may explain why Megabus and other bus lines place more emphasis on promoting their service as an alternative to the car than as a substitute for air and train travel. Relatively few promotion efforts focus on the difference in fare or the added frequencies associated with bus travel. Implications. These perceptions about autos and perceived stress from auto travel are important for bus–rail competition for several reasons: • Both intercity bus and train travel are poised to benefit from the continual shift in public attitudes about automobiles and driving. Many factors contributing to the perceived stress of

Competition to Rail from Intercity Bus 93 driving—congestion, tolls, and parking costs—appear to be on the rise. As a result, the portion of the population with higher levels of auto orientation do not appear likely to rise anytime soon, given the evident preferences of millennials. • Rising highway congestion is clearly beneficial for rail operators but will remain a double- edged sword for bus companies, benefiting them by increasing driver stress but also hurting them by making their rides longer and less predictable. An increase in highway congestion will no doubt further add to the perceptions of bus travel as “more inconvenient” than rail. • Issues such as a society’s view of auto ownership remain almost entirely beyond the control of rail and bus companies. How quickly these factors might change remain difficult to predict. One might reasonably expect, however, that the trends will be most favorable in densely populated areas with severe congestion. 8.4.4 Urbanism and Urban Location Neither a traveler’s preference for urbanism/sociability nor the density of their residential setting has discernable effects on the propensity to travel by bus—a finding with several notable implications for bus–rail competition. The small standardized effect of urbanism/sociability for bus travel (0.06) and train travel (0.03) suggests instead that these considerations are largely inconsequential for the bus industry’s—or Amtrak’s—recent growth in ridership. This result seems to run counter to the idea that many travelers choose buses and trains because these modes allow them to experience the full range of cultures, places, and experiences that urban living offers. Whereas their parents may have been pessimistic about bus and train travel due to their aversion to urban living, this logic goes, the new generation of bus/train riders would be attracted to it for this reason. The small coefficients, while being positive, suggest that this conclusion needs to be further explored with some caution and skepticism. Implications. The issues concerning urbanism may shed light on several aspects of bus–rail competition: • The absence of a clear link between attitudes about urbanism and train and bus travel helps explain why Amtrak has experienced stronger traffic growth between smaller cities and more rural intermediate points than between certain major cities. Much attention has been given, for example, in the large share of ridership on prominent Amtrak “string of pearls” networks that move between smaller cities, where an affinity toward urbanism may be less strong, than that of large terminals. • The results might shed light on why Megabus’s expansion to cities with lower population density (such as those in Florida and Texas) has apparently been successful, in spite of the fact that these cities lack the robust transit systems and pedestrian-oriented neigh- borhoods of Northeast and Midwest cities. Passengers in these less densely populated areas appear to have less aversion to the idea of bus travel than might have been anticipated sev- eral years ago, when major hubs for express city-to-city services were concentrated in dense transit-oriented cities. 8.4.5 Role of Information Communications Technology Attitudes about the importance of using information technology have a greater effect on the propensity to ride trains (0.19) than to ride buses (0.06). This result may reflect the higher pro- pensity for business travelers (who place a higher value on their time) to go by train than bus. On most routes, trains offer three tech-friendly amenities—tray tables, power outlets, and gener- ous seat pitches—that allow for passengers to, in effect, create mobile offices while they travel. Research from 2011 shows that the Acela Express is effectively in a class of its own with respect to

94 Intercity Passenger Rail in the Context of Dynamic Travel Markets technology use, with more than 60% of passengers engaged with devices at randomly observed points in time, much higher than on buses (Schwieterman and Fischer 2011a). The environment onboard standard intercity buses is generally less conducive to this level of technology engagement, although the free Wi-Fi is a “plus.” Even the small effect of this variable on bus travel is noteworthy. Research by the Chaddick Institute shows that the share of time that bus and train travelers spend on portable electronic devices has grown dramatically in the past 5 years and that usage rates are similar between modes (Schwieterman and Fischer 2011b). Overall, technology appears to be a more important mode-influencing factor for train riders than for bus riders. Implications. These observations suggest that several notable trends may be under way: • The small effect observed for both modes may reflect the sheer ubiquity of ICT in people’s lives, particularly among the younger demographic segments. This small coefficient for bus travel is largely consistent with the experiences of information technology enhancements on other modes, particularly air travel. Despite the push to install Wi-Fi on planes, relatively few passengers purchase this service, suggesting that few would change their mode choice decision if free Wi-Fi were offered. • The value of Wi-Fi appears to be falling as more passengers subscribe to 3G/4G services. Several years ago, there was arguably a wider gulf in the ability to engage in digital communica- tion on cars and buses due to the prevalence of Wi-Fi on the latter. The differences, however, appear to have diminished as more and more travelers subscribe to plans giving them access to the Internet on their smartphones. 8.4.6 Interpretation of Results from the Bus Attitudinal Model Interpreted broadly, the results support the idea that Amtrak and intercity bus lines both are poised to attract a bigger piece of the market as attitudes about driving change. Rather than viewing themselves as being locked in a battle for an existing market, the modes should recognize that they have much to gain by sending a common message to prospective riders who are attracted to alternatives to the private automobile. Moreover, the relatively low standardized effect of the cost (“expense”) variable suggests that changing perceptions about cost may not result in dramatic shifts in travel behavior. Amtrak may be better off targeting its discounts and advertising to specific demographic segments that have chosen to travel by bus—especially young and “pleasure” travelers—than to lower fares across the board. The carrier should also take advantage of the enhanced privacy and conveniences that it offers; benefits that will likely persist even as the intercity bus industry grows. Because rail travelers are more responsive than bus travelers to the service-quality variables used to forecast travel demand (e.g., convenience), Amtrak has a greater opportunity to gain ridership by improving its image in these areas than bus lines. Widening the advantage that trains have with respect to travel time would likely compel many bus riders to make the switch because convenience has the greatest effect on the propensity to choose either mode. 8.4.7 Lessons from the ICLV Model Observations relevant to the study of intercity bus from the full ICLV model presented in Chapter 5 can be made concerning the modal constants. The modal constants in the ICLV model in Chapter 5 provide important perspective on how the perceived desirability of traveling by bus differs from the other three modes when attitudinal

Competition to Rail from Intercity Bus 95 factors and service characteristics (such as travel time and cost) are held constant. The most notable findings include the following: • Regardless of gender, business travelers rank train travel first and bus travel last among the four modes. The fact that these passengers rank bus travel at the bottom is consistent with the common observation that briefcase-carrying business travelers are still relatively scarce on most intercity bus routes. • For female business travelers, air and bus are essentially tied for last. Travelers who are highly educated (i.e., have a graduate degree) are also less likely to go by bus than those who are not. Travelers who are employed are less likely to go by bus than by the other modes, regardless of the trip purpose. These results suggest that bus service generally behaves as an “inferior good,” especially for business travel, diminishing in prevalence as income status rises when all other factors are held equal. High income or highly educated travelers will generally opt for bus travel only when latent factors (attitudes) render them predisposed to do so or when some aspect of that mode, such as cost or travel time, offsets the mode’s perceived disadvantages. On vacation trips, buses fare better, ranking second—behind only trains—for males and third—behind rail and air—for females. Both males and females nevertheless rank bus travel last for trips involving visits to family and friends. Similarly, travelers are more apt to avoid buses for visits that require an overnight stay than those that do not. Travelers in groups, and especially those with three or more travelers, also rank bus travel much lower than other modes. Knowing a passenger’s age can also help predict whether he/she will be inclined to travel by bus when, again, the attitudinal factors are included in the equation. Older travelers (those over 55) tend to reject bus travel for trips to visit friends and relatives. Young travelers, however, have a stronger affinity for bus travel than for rail. Respondents under 35 have a significantly higher aptitude to go by bus than older respondents. Attitudinal factors appear to magnify the propensity for youthful travelers to choose buses, illustrating why this sector has been a core constituency for bus lines in recent years.