«Technical Report Documentation Page 1. Project No. 2. Government Accession No. 3. Recipient's Catalog No. SWUTC/11/161127-1 4. Title and Subtitle 5. ...»
LDC/TRC involve traveling a long distance to go to work, more often than not, across the boundary of a metropolitan region, either from one metropolitan to another or from/to non metro areas to/from metro areas. While some LDC/TRC commuters choose to commute daily, others may opt to live in a secondary home near work place and commute back to primary home weekly. This section first presents two economic theories explaining the residential or work location selections. Next it assembles the results of studies on intra- and trans-regional long distance commute.
3.1. Commuting and Job/Home Locations
Commuting is a necessity as long as there is a separation between residence and work place.
Hence commuting behavior is highly related to residential and job locations. According to the neoclassical model of urban residential location (Alonso, 1964; Mills, 1972; Muth, 1969) households try to maximize their utility subject to their income constraints, and the decision of their residential locations is a tradeoff between housing cost and transportation cost. This residential location model was developed based on the monocentric urban setting, in which land density and housing prices are lower far away from a central business district (CBD) than those closer in. Thus, households may choose to live further from the CBD, the workplace, and commute longer in exchange of lower housing cost and better living condition. When a technological innovation decreases the transportation costs, or household income increases, the urban periphery will be expanded since households have more desire for land with higher income and are able to commute longer. The residential location theory provides a basic explanation for commuting behaviors.
However, the residential location model assumed a perfect market condition under which workers were fully informed and were always able to choose the optimal amount of commuting distance. On the contrary, another body of economic theory, the search theory, suggests that workers have to search for jobs and dwellings continuously in order to improve their current position (J. van Ommeren, Rietveld, & Nijkamp, 1997, 2000). From the point view of search theory, commuting behavior is "determined by chance - the probability of receiving a job or residential offer at a certain distance - and a decision-making process - the decision to accept the offer" (J. van Ommeren, et al., 1997, p. 404). Based on the search theory, long distance commuting is often compensated by higher wage. Yet, the current highly specialized workforce has created a situation in which labor markets cannot provide rich job options within a moderate distance thus forcing workers to commute a longer distance even when it is not fully offset by wages in order to avoid costly job or residence moving (Sandow & Westin, 2010; J. Van Ommeren & Rietveld, 2007).
3.2. Long Distance Commuting within One Metropolitan Area
Research examining long distance across-region commuting has been limited. Long distance commuting has often been studied as excess commuting within one metropolitan region. Excess commuting represents the deviation of the actual average commute and the theoretical smallest average commute given the spatial setting of residential and workplace sites in a metropolitan area (Horner, 2002; Ma & Banister, 2006). It is often used as an indicator of the overall geographical imbalance between jobs and housing in a city and is associated with the issue of unsustainable urban land use pattern.
One subset of literature in excess commuting is concerned with factors that contribute to the long distance commute, which can provide useful insight for this researc. Ma & Banister (2006) listed many factors examined by researchers since 1980s that could prevent workers from finding jobs near residential location in a comprehensive review of excess commuting research. According to this review, the most studied and proved factor was the increased number of two worker households since there exist more obstacles to an optimized commute for both workers in a household as mentioned in the earlier section; home owners tend to have higher level of excess commuting than renters; people who have relatively unstable jobs also tend to accept longer commutes; occupation variation and pay variation in a job market could be another factor induce excess commuting; transport subsidies such as parking allowance may encourage people drive more; in addition, the costs of moving and rapid job turnover, neighborhood amenities and family life, and imperfect labor market information have all been examined as possible influences on excess commuting (Ma & Banister, 2006).
Excess commuting is commonly deemed as inefficient and unnecessary, so most of the literature has been concentrated on finding solutions to reduce excess commuting. The redistribution of workers and encouraging mixed land use to achieve job and housing balance are the mainly suggested policy implications (Ma & Banister, 2006). However, the intricate causes of long distance commuting challenges the effectiveness of these policies; the continuing complex travel behavior of households and more advanced ICT applications could also weaken the power of policies that only focus on shortening journey-to work trips. Thus, it is not surprising to see more excess commuting emerging. Furthermore, Some researchers have argued from the psychological point view that humans prefer the switch buffer between home and work activities as well as the distinct territories of home and work created by commuting (Lyons & Chatterjee, 2008; Ma & Banister, 2006). Humans also have the natural desire for travel (Mokhtarian & Salomon, 2001). These psychological factors could contribute to the decision to commute long distances.
In terms of regional economic strength, the land use and transportation policies that attempt to balance job and work and reduce long distance travel pose another question, as described by Pisarski (2006, p. 150) "if most workers actually work in their residence county[or other region unit], that would clearly be better for the transportation system in terms of congestion, but would it be better for the region as an entity?...Isn't the strength-the hallmark- of a region based on its ability to provide a market in the millions? For example, an employer in a very specialized sphere locating in a large region has a market of prospective employees measuring in the millions. This is also true of an exotic restaurant, great art gallery, or any specialty store. This suggests that transportation policies that would suppress longer distance travel and encourage short-distance trips are destroying part of what makes a big region a greater region."
Except for generating problems such as congestion and a waste of energy, excess commuting might deliver a message to us that workers are demanding access to larger job markets or larger housing markets. When the needs exceed the boundary of one metropolitan area, interregional commuting is inevitable. Thus while on the one hand, it is important to provide better balance in job and housing in a particular region so that households have options to locate homes near work, on the other hand, long distance interregional commuting should be further examined, and solutions should be offered to workers who choose to take such commuting to do so in a more sustainable way thus creating a society in which people have more freedom to choose live and work where they want.
3.3. Long Distance Commuting across Regions
One relatively new study focused on the interregional long distance commuting in the US was conducted by Lee (Lee, 1995, 1996) in 1990s as his dissertation research. This research investigated the motivations of the long distance commuters who traveled from homes in California's San Joaquin Valley to work in the San Francisco Bay Area. The long distance commuting in this study was defined as journeys that were over 45 minutes one way and involved crossing a metropolitan area boundary. Based on the Census PUMS 5% data, Lee concluded that a white married male with a medium education level (but relatively higher level than other commuters in the San Joaquin valley) and worked in construction, communications or other public utilities field represented a typical long distance commuter between the Central Valley and the Bay area.
Since the Census data did not provide detailed answers for behavior questions, Lee further carried out four focus group discussions and in-depth interviews with 40 interregional commuters. The respondents were contacted through the Bay Area regional ridesharing agency.
He discovered that the expensive housing price in the Bay area was an important reason why these people chose to reside in the San Joaquin valley where they could afford larger houses.
Additionally, maintaining a better living environment was a common reason for households with small kids. There were also commuters who preferred rural lifestyle, and who did not mind driving. Although these commuters expressed some dislike to the driving, they all became used to it after a while and treated the long distance commuting as a long-term goal. Lee concluded that the "pull" power that attracted people to live in the valley had more influence on the decision to long-distance commuter than the "push" factors that repelled people from wanting to live in the city, such as high housing costs or crime.
Lee's research is quite similar to this study. Therefore it is noteworthy to compare the two cases.
First, in the California case, the interregional commuting is between the San Francisco Bay area, a highly job concentrated place with high living cost and the San Joaquin Valley, a "bedroom" community. So the commuting has a fixed direction and more resembles the "suburb to CBD" commuting pattern. While in the Texas Triangle case, there is no clearly presumed commuting flow direction since the living cost in the four metro areas are not as dramatically uneven as the California case. Second, the average commuting time in the California case was 90 minutes, and commuting was all done by workers on a daily bases, yet in the Triangle Area, the distance among the metropolitans are much longer, and commuting between them may become weekly.
Third, Lee performed this research more than 10 years ago. Since then, the ICT has been rapidly advancing. This improvement could have large impact on the interregional commuting.
More recent studies on long distance interregional commuting have been performed by researchers in Europe. According to the literature, the number of long distance commuters crossing municipal regions or even national boundaries has been steadily increasing in some European countries. Such commuting includes daily travel between home and work and weekly commute (Lyons & Chatterjee, 2008; Sandow & Westin, 2010). The literature that is concerned with this type of commuting can be grouped into two directions. One group studied the long distance commuting from travel behavior point of view and attempted to explain people's commuting decision using quantitative methods. However, research in this group did not specifically distinguish between daily and weekly commuting but rather typically defined long distance commuting as a work trip longer than 30 to 45 minutes. Another group paid a special attention to the weekly commuting phenomenon. Studies in this group were typically done by social/family life researchers and geographic researchers who devoted their effort to inspect the impact of weekly commuting on family life and the connection between migration and weekly commuting. In this section, the first group of literature will be reviewed, and the second group will be discussed in the later section.
A common method for studying long distance commuting is the utilization of secondary data developing regression models to scrutinize the causal effects of various factors on long distance commuting (Ohman, 2010; Sandow & Westin, 2010; Titheridge & Hall, 2006). One study also explored people's preferences and options of commuting through a survey (Sandow & Westin, 2008).
Titheridge & Hall (2006) inspected long distance commuting in the Greater South East Region, the "global mega-city region" (Hall & Pain, 2006) in England. The study focused on the East Corridor and the North Corridor radiating from London to the periphery of the region where rail service is available. Six models with dependent variables of commuting distance and different commuting modes were developed for each corridor for years of 1981 and 1991 based on the Census data. Titheridge & Hall concluded that a lack of job opportunities near one’s residence was a significant reason for workers to choose a longer commute. In addition, they found that higher social class people were the ones who travelled the longest distance.
Titheridge & Hall's finding about the association between long distance commuting with social classes was confirmed by Sandow & Westin (2008) who conducted a survey of 2,500 samples in 2004 in four municipalities in northern Sweden where the population density is only 2-15 inhabitants/km2. Based on this survey, Sandow & Westin examined people's inclination and opportunities to commute in this relatively sparsely populated environment. Their results showed that people with a high level of education, especially males with a higher level of education working in the private sector, were more willing to accept longer commutes. The survey also revealed that 45 minutes one way was a common maximum that people deem feasible and tolerable for daily commuting. The 45-minute time limit has also been verified by some other researchers (Levinson & Wu, 2005; Ohman, 2010; Sandow & Westin, 2010; Van Ham & Hooimeijer, 2009; J. van Ommeren, 1998), and when commuting is longer than 45 minutes, weekly commuting or migration may be preferred (Sandow & Westin, 2008).
Sandow & Westin (2010) later conducted another study using the 1995-2005 register data of Sweden to analyze the duration of long distance commutes. They found that most long distance commuters had been commuting long distances for many years, and economic incentive was important for them to sustain such long work trips. Sandow & Westin concluded that since longer commuting provided more opportunities for all members in a household and was often associated with higher income, it tended to be a long range mobility strategy rather than a temporary solution for households.