Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Persistence of commuting habits: context effects in Germany

Persistence of commuting habits: context effects in Germany In this study, I investigate the commuting behavior of workers in Germany. Using comprehensive geo-referenced administrative employee and firm data, I can cal- culate the exact commuting time and the distance between workers’ residence and workplace locations. Based on a behavioral economic approach (Simonson and Tveresky in J Mark Res 29:281–295, 1992), I show that individual commuting deci- sions are influenced by wages and individual heterogeneity as well as depending on the context individuals observed in the past. In particular, my results show that previously observed commutes have an impact on subsequent commuting behavior: workers choose longer commuting times in the region they recently moved to when the average commute in the region they left was longer. The results indicate that while selectivity and sorting do not influence the effect of the context, the inclusion of individual fixed effects is crucial. JEL Classification J60 · R10 · R19 · R23 1 Introduction The importance of commuting is growing rapidly—both the number of commuters and the distance they commute are growing steadily (Gimenez-Nadal et  al. 2020). From an economic perspective, commuting is essential for a well-functioning labor market as it is an important measure to overcome spatial separations (Lux and Sunega 2012; Zabel 2012). At the individual level, commuting implies better labor accessibility and subsequently improves job and career opportunities, leading to bet- ter outcomes and improved individual utility. However, commuting also has negative impacts on both the environment and the infrastructure (Brueckner 2000; Rouwen- dal and Rietveld 1994), as well as on individuals’ well-being as it is associated with congestion and high costs (Frey and Stutzer 2007). Understanding the determinants * Ramona Jost ramona.jost@iab.de Institute for Employment Research (IAB), Regensburger Str. 104, 90478 Nuremberg, Germany Vol.:(0123456789) 1 3 R. Jost of and the reasons for commuting is thus an important topic for policymakers deal- ing with economic and labor market issues. Studies on commuting find different factors and effects that influence individu- als’ commuting behavior, for example commuting is more common among males and among workers with higher incomes as well as among homeowners. The same applies to workers who are older and work in specific occupations and have specific skill levels (Gimenez-Nadal et  al. 2018, 2020; Ross and Zenou 2008; Hanson and Johnston 1985; Dargay and Clark 2012; McQuaid and Chen 2012). However, individuals’ commuting behavior might also be explained from a behavioral economic perspective. In particular, previous research shows that previ- ously observed options can influence individuals’ perceptions and therefore their subsequent decision-making behavior (Simonson and Tveresky 1992). Applied to individuals’ commuting behavior this means that previously observed commuting options influence their preferences for commuting and consequently their own com- muting decisions. This approach can explain, for example, why individuals who move to Munich commute 30 percent less than the average in Munich if they come from regions with shorter average commuting times, while individuals commute 35 percent more than the average in Munich if they previously lived in regions with longer commuting times than those typical in Munich. This might indicate that com- muting decisions are influenced by the context of commuting options observed in the past, such as other individuals’ commutes. This study analyzes such commuting behavior, based on the study conducted by Simonsohn (2006) for the US, and contributes to the literature in at least four ways: first, it contributes to the literature on commuting behavior and the factors that are important for explaining commuting (Gimenez-Nadal et al. 2020; Dargay and Clark 2012; McQuaid and Chen 2012). In particular, I show that the context of commuting options observed in the past is crucial for analyzing individuals’ commuting behav- ior. In this context I show that the results obtained by Simonsohn (2006) are biased due to the omission of individual fixed effects and the consideration only of migrants between two metropolitan areas. Second, I reveal effects for different groups, dis- cussing effect heterogeneity for age, gender, skill level, as well as rural and urban areas, for an entire country. Third, I use geo-referenced employer-employee data. These administrative registry data possesses higher validity than survey data and provides precise information about individuals’ residence and workplace locations with a high number of observations. This makes it possible to calculate the exact commuting distance and time for German workers. Fourth, the study contributes to the migration literature (van Ham and Hooimeijer 2009; Brueckner and Stastna 2020; Shuai 2012). In particular, I show that the greater the difference between a worker’s individual commuting time and the average commuting time at their place of residence, the more likely they are to move again. When individuals choose where to live, they face the difficult decision of how far they are willing to commute, weighing up the benefits and costs of commut- ing. Advantages of commuting may include cheaper rents and housing prices out- side the city center, resulting in a higher disposable income. Furthermore, com- muting can provide more job opportunities for individuals who live in rural areas where there may be no or no adequate employment offers. However, commuting 1 3 Persistence of commuting habits: context effects in Germany also has disadvantages; it takes up time, causes stress, and impacts the reconcili- ation of work and family. It can therefore have a negative effect on individuals’ well-being (Frey and Stutzer 2007). When deciding how far they wish to com- mute, individuals have to trade off the benefits with the disutility of commuting. Indeed, costs and benefits do not have the same effect on utility: the response to losses is stronger than the response to the corresponding benefits (loss aversion, Kahneman and Tveresky 1979). In the context of commuting decisions, however, Dauth and Haller (2020) find no sign of loss aversion, which contradicts previous experimental evidence (Tveresky and Kahneman 1991). Empirical evidence from urban economics reveals the disutility of commuting for which individuals wish to be compensated. For the Netherlands, van Omme- ren et al. (2000) and van Ommeren (2005) find a marginal willingness to pay for an additional kilometer of commuting of 0.15 euros per day or 17 euros for one additional hour of commuting (van Ommeren and Fosgerau 2009). With regard to compensation by the employer, Heuermann et  al. (2016) find that employers compensate only few employees directly for additional commuting costs. Hence, the decision to commute is mainly an individual one, which can be strongly influ- enced by prior experiences. However, individuals are often unable to assess correctly the disutility of com- muting and are frequently uncertain about their preferences, which contradicts the standard economic theory (Kahneman and Tveresky 1979). Instead, they form their preferences as and when they are needed, for instance when making choices (Bett- man et  al. 1998). For example, in the context of commuting decisions, individuals rely on a wide range of possible cues, such as other individuals’ commutes. Moreo- ver, in the literature on decision-making (Bettman et al. 1998; Huber et al. 1982) it becomes fundamental that an individual’s decision can be influenced by the context: individuals interpret information by comparing it not only to other available options, but also to what was recently observed. According to Hartzmark and Shue (2017), these context effects have the potential to affect a variety of important real-world decisions. They not only distort judicial perceptions of the severity of crimes, lead- ing to unfair sentencing, but also affect employee hiring, medical diagnoses as well as housing and commuting decisions. The context effect that is relevant for this study is the background context effect, according to which choices depend on options encountered in the past—preferences can change with the history of choices. The intuition behind this is that the same product may seem more attractive against the background of less attractive alter- natives and unattractive compared to more attractive alternatives (Simonson and Tveresky 1992). Simonson and Tveresky (1992) document this effect in an experi- ment comprising two stages in which subjects have to make choices in sequence. In the first stage, half of the subjects are confronted with two options that have a rela- tive high cost for one attribute, and the other half should make a choice with a rela- tively low cost for the same attribute. In the second stage, all subjects are confronted with the same choice. In line with the background context effect, subjects who are confronted with a relatively high cost for an attribute in the first stage are more likely to choose the more expensive option in the second stage because it appears cheaper to them. 1 3 R. Jost There is ample evidence of the background context effect. Bhargava and Fisman (2014) demonstrate this effect in the context of speed dating. They show that the attractiveness of previous partners reduces the probability of finding a date. Moreo- ver, Hartzmark and Shue (2017) demonstrate that today’s earnings impress investors more when previous earnings were poor. Furthermore, Simonsohn and Loewenstein (2006) present the effect with regard to housing choices: individuals who move from cities with relatively high housing costs are more likely to pay higher prices in the new city compared to individuals coming from cities with cheaper markets. Applied to commuting behavior, this means that commuting options encountered by indi- viduals in the past affect their current commuting decisions. However, relatively lit- tle research has been conducted into when and why the background context effect influences commuting decisions. The only such study was conducted by Simonsohn (2006). He considers individuals relocating between two metropolitan areas in the US and takes the average commuting time in the previous city as a proxy for com- muting options encountered in the past to examine how previously observed com- mutes influence commuting decisions when moving to a new city. He finds that individuals choose longer commutes in the new city, the longer the average com- mute was in the city they came from. Commuting decisions are thus influenced by commuting options encountered by individuals in the past, which is in line with the background context effect. In this study I consider workers who relocate between NUTS-3 regions in Ger- many and examine the context effect for workers of an entire country, which is why I deviate from the approach of Simonsohn (2006) and use the average commuting time at the NUTS-3 level for the proxy of commuting options encountered in the past. The results show that individuals coming from backgrounds with longer aver- age commuting times initially choose longer individual commutes in the destination region compared to individuals from regions with shorter average commutes. In contrast to Simonsohn (2006), I additionally differentiate between individuals moving between different region types of rural and urban regions and thus I show that the context effect is strongest for workers who move from rural to urban areas. Further, the robustness checks show that selectivity of a relocation does not influ- ence the effect of the context and I find no evidence of workers selecting themselves into regions because of their taste for commuting. However, my results do indicate that it is very important to control for individual fixed effects. Moreover, I find no sign of stable taste difference as traditional economic theory would suggest. The remaining paper is structured as follows. Section  2 provides the theoretical motivation for the background context effects. Section  3 discusses the data and the sample. The identification strategy used is shown in Sect.  4. The empirical results are presented in Sect. 5, and Sect. 6 concludes. 2 Theoretical motivation for the background context effects As empirical evidence shows, decisions are preference-dependent (Bettman et  al. 1998; Huber et  al. 1982; Hartzmark and Shue 2017; Bhargava and Fisman 2014; Simonsohn and Loewenstein 2006). However, these preferences change with 1 3 Persistence of commuting habits: context effects in Germany previously observed options. As Tveresky and Simonson (1992) demonstrate in their background contrast experiment, individuals’ previous experiences influence their perceptions and therefore their subsequent decision-making behavior. For commut- ing decisions, this implies that commuting options encountered previously affect current commuting preferences and thus individuals’ commuting behavior. The fol- lowing approach is based on this concept, which is also used by Simonsohn (2006). The idea is that the disutility of commuting decreases when a person was only con- fronted with longer commuting options in the past, whereas, the disutility increases when individuals were only exposed to short commutes. To investigate this approach and to measure the effect of the context, I use relo- cations involving individuals moving between two NUTS-3 regions in Germany. According to the background contrast experiment conducted by Tveresky and Simonson (1992), the commuting behavior after the move should be affected by pre- viously observed commuting options. This concept is formally represented as: = (1 − ) + (1) t−1 t with β∈ [0, 1]. Abstracting all other influences, such as sociodemographic factors, represents a person’s individually chosen commuting time as a weighted sum of the observed commuting options in the present  and the past  , with the weights t t−1 decreasing exponentially into the past (Ryder and Heal 1973). More precisely, under the assumption of β = 1 there is no impact of commutes observed in the past on the current commuting time, since  =  and thus no impact of the context. In contrast, if β = 0 the current commuting preferences are determined only by the previously observed commuting times, corresponding to  =  . In the following, I expect β t−1 to take values between 0 and 1 (0 < β < 1), such that two otherwise identical indi- viduals with different numbers of previously observed commuting options will have different levels of  when moving to the same region. Moreover, I use the average commuting time in the region of residence before the move as a proxy for previously observed commuting options (Simonsohn 2006) . According to Eq.  (1), individu- als moving from regions with longer average commutes accept a longer commuting time  when choosing places of work and residence in the destination region com- pared to individuals coming from regions with shorter average commuting times. This is the first prediction I investigate in this study. The average commuting time in the region a person leaves has a positive influ- ence on the individually selected commuting time in the destination region However, if individuals stay in the new region and observe the commut- ing options in the new region, their preferences for commuting change due to the new observed commutes in the new region. This leads to a change in the desired commuting duration. For example, movers who relocate from regions with longer In contrast, Simonsohn (2006) uses the average commuting time on the city level, as he only analyzes movers between two metropolitan areas. Thus, while the predictions are quite similar to those of Simon- sohn (2006), the objects of investigation differ due to the different target group of movers. 1 3 R. Jost commutes to regions with shorter ones initially have a greater tolerance for long commutes and prefer cheaper and larger living space outside the city center. There- fore, they initially commute longer than the average commute in the new region. If they remain in this region and observe shorter commutes, however, their preferences for shorter commutes grow and the disutility for commuting increases. They thus become dissatisfied with the commutes they chose initially and might move again within the new region to reduce their commuting time, thereby correcting an origi- nally excessive amount of commuting. This relationship is illustrated by the second prediction. Individuals readjust their commuting times and move again when remaining in the new region The second prediction is therefore useful for ruling out explanations based on stable unobserved differences across individuals who move from different regions. Because if individuals who come from regions with longer average commutes travel more after relocating because they are different from those coming from regions with shorter average commutes, I would not expect them to revise their commutes by moving again. 3 Data and sample selection 3.1 Data For the analysis, I use the employment biographies of a 6-percent random sample of all German workers subject to social security contributions. The administrative registry data does not include self-employed persons or civil servants; however, it covers more than 80 percent of the German labor force. The Employment History (BeH – Beschäftigenhistorik V10.01.00, 2016) collated by the Institute for Employ- ment Research (IAB) provides exact information about periods of employment based on the status reports submitted to the pension insurance. Besides the sociode- mographic characteristics, information at the firm level are included, which comes from the Establishment History Panel (BHP). This dataset contains information about the branch of industry, the establishment location, number of employees and marginal part-time employees. As daily wages are top-coded at the social security contribution ceiling, I use the imputation procedure developed by Card et al. (2013) to recover wages above this threshold. A unique feature of this dataset is the supplement IEB GEO, which provides anonymized address information in the form of geocodes for the locations of an individual’s residence and place of work for the years 2000–2014 (Ostermann et al. 2022). Combining this address information with road network data from Open- StreetMap, I calculate door-to-door commuting distances (Huber and Rust 2016; Dauth and Haller 2020; Duan et al. 2022). It is only possible to determine distances for individuals traveling by car in this way; those for users of public transport may differ. However, the car is the most important mode of transport. Almost 70 percent 1 3 Persistence of commuting habits: context effects in Germany of workers commute to work by car (Destatis 2017), whereas only 14 percent of commuters use the public transport system. In addition, to calculate the commuting time I take average values for highways, primary, and residential roads. By using geocodes, the commuting time is not limited by administrative units, which reduces measurement error for individuals close to administrative borders and mitigates the problem of spatial sorting within areas. Yet, using driving time can cause issues regarding the experienced commuting time: for example, the algorithm cannot rec- ognize dense traffic in the daily rush hours. Nevertheless, as the time is measured before and after the regional move, the change in the duration might be affected less by this measurement problem. 3.2 Sample In this study, I investigate the commuting behavior of German workers, excluding persons in marginal and part-time employment as well as workers older than 57 and younger than 18 years of age. Regarding the commuting time, I restrict the sample to workers with a commuting time between 1 and 90  min. I choose 1  min as the minimum because this represents the first percentile of the data and hence ensures that outliers who do not commute are not considered. The restriction to 90  min is because the data does not provide any information about the number of commuting trips. Thus, the data could also include workers who commute weekly and have a second place of residence. To exclude those workers, I restrict the data to workers with commuting times of up to 90 min. This is comparable to other German studies that restrict the commuting distance to 100 km (Dauth and Haller 2020; Duan et al. 2022) and ensures that commuting is conducted on a daily basis. To test prediction 1, whether the average commuting time in the region a per- son leaves has a positive influence on the individually selected commuting time in the destination region, several restrictions have to be considered. First, to be able to analyze commuting decisions, I have to consider only those individuals who face such a decision. This group comprises individuals who are required to make a new commuting decision due to moving home or changing their job. For my study, how- ever, I consider individuals who simultaneously change both their place of residence and their place of work. The reason for this is, first, that for individuals who only change their place of work it is not possible to examine the influence of the con- text of commutes observed in the past, because for job changers the region of the place of residence does not change. Second, if individuals only change their place of residence they might, for example, be relocating due to dissatisfaction with com- muting and I would therefore not be able to identify the influence of the context However, the results obtained by Simonsohn (2006) show that the context has almost the same effect for people who use public transport. For the analysis in this study I consider commuting time. However, all the results are very similar when commuting distance is used. In a robustness check, I investigate the effect of the context for this group, then also provide evidence of a context effect for this group of movers. 1 3 R. Jost correctly. To avoid this, I restrict the sample to workers who change both residence and workplace locations, which further guarantees a relocation of the entire center of their lives. In addition, I restrict the sample to those movers who relocate between two of the 402 German NUTS-3 regions. I also keep the NUTS-3 region of the place of work and the place of residence constant for two years before and after the move. This guarantees that movers are able to adopt the commuting options as well as the commuting behavior of the region they lived in. In addition, this assumption means that it is possible for movers to relocate again within the target region to read- just their initially chosen commuting time. After these restrictions I identify 15,671 workers who move between two NUTS-3 regions. Furthermore, the time periods are categorized to t − 1 for the year before the move, t = 0 for the year of the relocation and t + 1 for the year after the move. To test prediction 2, I look at workers who relocate again within the new region in period t + 1 (one year after the move), keeping the place of work constant. The number of second-time movers is 4267. 4 Identification strategy To test the first prediction, I estimate how the average commuting time in the region of residence before the relocation C influences the individually chosen commut- i,t−1 ing time in the target region C , I consider a dynamic fixed effects model, where i,t=0 the lag of the dependent variable C is used as an explanatory variable : i,t−1 ln(C )=  ln(C )+  ln(C )+  X +  + (2) i,t=0 1 i,t−1 2 i,t−1 3 i i,t i,t where ln(C ) represents the dependent variable, the logarithm of the individual i,t=0 chosen commute in minutes after the relocation t = 0, while ln(C )—the lag of i,t−1 the dependent variable—is added as an independent variable. The variable of inter- est ln(C ) shows the logarithm of the average commuting time in the region of i,t−1 residence before the relocation t − 1. The average commuting time is calculated for each NUTS-3 region and represents the context of previously observed commutes. Further, I include X as a vector of control variables. This vector includes the i,t=0 log wage, calendar years, occupational status and indicator variables for firm size (number of employees, 4 categories), age group (4 categories), occupation (12 cat- egories), industry (9 categories) and region type of the place of residence as well as of the place of work (according to the classification of the Federal Institute for Research on Building, Urban Affairs and Spatial Development BBSR). These region types represent whether individuals live and work in a metropolitan city, city, large Estimating the model for the group of movers who only change their place of residence also reveals an effect of the context. The results can be provided additionally on request. However, investigating movers between German labor market regions generates almost the same results (see Web Appendix G). In this sample, I include all workers who relocate between two NUTS-3 regions. For all workers I have 5 observations, two observations before the move, the period of the relocation, and two after. 1 3 Persistence of commuting habits: context effects in Germany town, small town or in a rural area (5 categories). Moreover, X incorporates sev- i,t=0 eral dummies indicating whether a worker is a supervisor, has a leading position, is a trained/professional, specialist/expert or has an auxiliary job. In addition, X i,t=0 incorporates a dummy for women, migrants, western Germany and for being low- skilled (without vocational training) medium-skilled (with vocational training) or high-skilled (academic degree). And  shows the time invariant individual-specific effects. According to prediction 1,  should be positive because individuals with stronger observed commuting backgrounds have a lower disutility of commuting and thus prefer to live outside the city center, thereby facing longer commutes. However, in the case of unobserved heterogeneity, omitted variable bias and selectivity which can influence the estimates of C or sorting—meaning that mov- i,t−1 ers relocate to certain regions because of their taste for commuting—my results would not be valid. First, to address the issue of unobserved heterogeneity regard- ing, for example, commuting preferences, the estimates control for individual fixed effects  (Eq.  2). Thus, unobserved heterogeneity regarding individual commuting should not impact my results. Second, to deal with the issue of omitted variable bias, I conduct several robust- ness checks excluding observable individual and firm characteristics in my analysis. The results are presented in the robustness checks in Sect.  5 (Table  8) and confirm my presented results, as the results barely change. Third, workers might endogenously choose whether or not to move. To control for this selectivity, I use a two-stage Heckman selection method (Heckman 1979) where I first account for the decision to move, which can be estimated as a latent variable model: P =  S + (3) 1 i i With the decision to move: 1 if P > 0 P = (4) 0 otherwise P represents the latent variable for the propensity to move between two NUTS-3 regions and S is a vector of sociodemographic characteristics and information on industry and firm size, which influence individual i. To estimate whether or not a worker moves, I use a probit estimation. These results are then taken to construct an inverse Mills ratio. This inverse Mills ratio is then included in the second step equa- tion to correct for selection bias (Eq. 2). The third issue is sorting: For example, individuals who dislike (like) commut- ing choose regions with shorter (longer) commuting times. To face this selectivity issue, I include the individual’s own commuting time in the region before the reloca- tion C (see Eq. 2), and perform a robustness check. In line with selectivity, indi- i,t−1 viduals select themselves into a region because of their commuting taste. If people select themselves into regions with longer average commutes because of their taste for long commuting, they should also have commuted longer in the region before the move. To exploit this fact, I perform a reversed regression in which I regress the 1 3 R. Jost individual commute in the previous region on the average commuting time in the target region—after the relocation. ln(C )=  +  ln(C )+  X + (5) i,t−1 0 1 i,t=0 2 i i,t=0 In line with the above argument indicating selectivity, I should find a positive effect of the average commutes in the destination region C on the individuals’ i,t=0 commuting time in the region before the movement C . The results are presented i,t−1 in the robustness checks in Sect. 5 (Table 9). Another neglected effect could be due to imperfect information: when moving to a new region worker have no information about the commuting situation there. Therefore, they might commute longer initially and then change their commutes by relocating again within the new region—thereby explaining the second prediction. However, information about commuting and the local housing market is relatively cheap. Nevertheless, the commuting costs are high: commuting takes time, causes stress, and is very expensive. I would thus expect workers to obtain information about the commuting situation in the new region before they move. In addition, the decision regarding accommodation might be made under time pressure, thus representing a random event. For example, when individuals have found a new job but then have little time left to find a new apartment. In this case, they might be willing to take any accommodation, wherever it is located, as long as it seems to be acceptable. However, if it appears to be the case that the new commut- ing time is a random event, first I would not expect the individual’s own previous commuting time as well as the average commuting time in the region before the move to have a significant influence on the selected commuting time in the target region. And second, I would not expect those workers to move again within the new region and adjust their commuting time to the average commuting time in the new region. The travel time budget—and thus the commuting decision—might also be influ- enced by trip chaining or by the fraction of remote work. In particular, with the Covid-19-shock remote work has increased and there is some consistency in remote work. Due to the possibility of working from home the travel-time budget becomes more relaxed and thus longer commuting distances might be expected and accepted. However, as my observation period is restricted (2000–2014) and the data does not include the fraction of remote work, I cannot analyze how the results might be affected by the Covid-19-shock. In addition, Brunow and Gründer (2013) found that the daily allocation of time in Germany is affected by trip chaining, such that unob- served factors may influence the time budget. In particular, after migration not just the trip “home-to-work” influences the persistence of habits but also other factors such as shop accessibility or child care institutions leading to a potential bias in esti- mates. However, I suspect that this bias is negligible in this study, because people living in the destination area still form the daily activity chains. To test prediction 2, I restrict the sample to workers who move again within the new region, one period after the first move t + 1. I use the following identification strategy, in which only changes are analyzed. Because of these differences, individ- ual fixed effects are canceled out: 1 3 Persistence of commuting habits: context effects in Germany ln(C − C )=  ln(C − C )+  ln(W − W )+ (6) i,t+1 i,t=0 1 i,t=0 i,t−1 2 i,t+1 i,t=0 i The dependent variable (C − C ) is the change in the individual chosen i,t+1 i,t=0 commuting time after the second and the first move within the new region. The con- trol variable is the change in wages (W − W ) between the second and the first i,t+1 i,t=0 move. And the key predictor is represented by the difference between the observed commuting time in the new region t = 0 and in the region before the move t − 1, cor- responding to (C − C ) . This classification of the reference point presupposes i,t=0 i,t−1 that the workers’ perceptions have fully adjusted after one period. However, this might still not be a correct estimate of the change in the commut- ing time as workers might endogenously choose whether to move a second time. Therefore, I again use a two-step Heckman selection method (see Eqs. 3, 4). If work- ers decide to move a second time within the new region, in line with prediction 2, the coefficient  (Eq.  6) should be positive: individuals moving from regions with observed long commutes to a new region (with shorter average commutes) commute too long at first. This leads to a change in the desired commuting durations. There- fore, if they move again within this new region, they reduce their commutes and adopt the commuting behavior prevalent in the new region. 5 Empirical analysis of the commuting behavior 5.1 Descriptive statistics Figure  1 presents the distribution of the average commuting times for the place of residence for each NUTS-3 region in Germany. Workers living in metropolitan cit- ies, like Munich, Berlin, Frankfurt or Bremen, have shorter average commuting times than those in the surrounding regions. Specifically, the average commuting time in metropolitan cities is 16.8 min, while workers in rural areas commute almost 20  min to work on average. This implies that workers who live in large cities are most likely to work there as well, while workers living in the suburbs travel from the surrounding regions into the city center to work. This may be because job oppor- tunities are better in the city center and housing costs are cheaper in the suburbs (Alonso 1964). 5.1.1 Comparison of movers and non‑movers To demonstrate how the characteristics of workers who relocate differ from those who do not, I compare the two groups. The results are represented in the Web Appendix A. They show that movers and non-movers differ especially in terms of their productivity-related characteristics: employees who relocate are more highly qualified (academic degree) than non-movers. Differences also become obvious with regard to industries, occupations, and age groups. While the share of mov- ers is much larger between 18 and 34, non-movers are mainly between 35 and 1 3 R. Jost Bremen Berlin Frankfurt Munich Notes: The map shows the mean commuting time in workers’ place of residence by NUTS-3 regions in manually chosen time categories. Source: Own calculation and presentation. Fig. 1 Regional distribution of commuting time in the year 2014 56  years old. Moreover, movers tend to drive an average of 1.2  min longer to Table 1 Summary statistics of the daily wage and commuting time Variable Mean Std. dev 25th perc 50th perc 75th perc Commuting time t = − 1 in minutes 18.8 16.6 6.9 13.9 25.1 Δ Commuting time t = 0 in minutes + 3.9 23.8 − 8.2 2.7 14.9 Wage t = − 1 (euros/day) 85.9 55.9 49.7 74.2 106.5 Δ Wage t = 0 (euros/day) + 12.8 41.0 −2 .7 8.5 27.1 N 15,671 Means, standard deviation, 25th, 50th, 75th percentiles of commuting time and the wage. Comparison of movers before and after the relocation work than non-movers. This comparison therefore shows considerable heteroge- neity between movers and non-movers. 1 3 Persistence of commuting habits: context effects in Germany Table 2 Summary statistics of Variable Occupation Industry Promotion changes in occupation, industry, and promotion Change as a 33.0 33.9 12.7 percentage N 5238 5372 2009 Percentage of workers who change occupation or industry, or are promoted after the relocation t = 0 5.1.2 Comparison of movers before and after the relocation In the following, I examine summary statistics of workers who move. Table  1 shows the difference between movers’ driving times and wages before t = − 1 and after the relocation t = 0. The average mover experiences an increase in wages (+ 12.8 euros per day), which supports the idea that workers are more likely to move if they can achieve a wage increase, as non-movers on the other hand only experience an average wage increase of about 3.4 euros per day between two periods. Not only wages rise due to the relocation, the commuting time does so too. On average, the commuting time among movers increases by 3.9 min. 5.1.3 Motivation of movers As already mentioned, when workers move to a new region, they achieve an increase in wages, which could be an important motivation to move. Further- more, Table 2 shows that 33 percent of workers change their occupation after the move. In addition, almost 34 percent of movers work in a different industry after relocating. Workers might therefore move in particular for job-related reasons. Simonsohn (2006) obtains a similar finding. He reports that more than 36 percent of indi- viduals in the US move for job-related reasons. Moreover, in many cases (12.7 percent) the move is associated with a promotion, for example from trained/pro- fessional assistant to specialist/expert (see Table 2). 5.1.4 Comparison of movers and second‑time movers In the following, I take a closer look at second-time movers. These are workers who relocate a second time within the new region. Table  3 compares these sec- ond-time movers with the share of regular movers (workers who move once) after the first and before the second move. Of the 15,671 movers in t = 0 4,267 relocate a second time in t = 1. Especially medium-skilled workers tend to move again within the new region. In addition, the shares of men, migrants, and workers in western Germany are higher for sec- ond-time movers, and they are younger on average (between 18 and 24 years old). 1 3 R. Jost Table 3 Summary statistics of Variable Movers (%) Second- main variables time mov- ers (%) Woman 50.6 47.6 Migrant 3.9 4.3 West Germany 86.4 89.0 Age groups  18–24 14.5 18.4  25–34 47.9 46.8  35–44 26.8 24.9  45–56 10.8 10.0 Skill level  Low-skilled 6.5 7.9  Medium-skilled 63.1 68.5  High-skilled 30.4 23.6 N 11,597 4267 Means of main variables. Comparison of movers and second-time movers after the first move t = 0 Table 4 Summary statistics of commuting time and wage Variable Mean Std. dev 25th perc 50th perc 75th perc N Movers Commuting time 18.7 15.5 7.7 14.6 25.0 11,597 (min) Second-time movers Commuting time 33.4 22.6 15.2 28.1 48.1 4,267 (min) Movers Wage (euros/day) 99.7 58.6 61.6 84.1 122.4 11,597 Second-time movers Wage (euros/day) 96.0 55.0 62.7 82.0 113.8 4,267 Means, standard deviation, 25th, 50th, 75th percentiles of driving time and the wage. Comparison of movers and second-time movers after the first move t = 0 Table  4 shows the difference between the daily wages and the commuting times of movers and second-time movers after the first relocation t = 0. Compared to movers, second-time movers have much longer commuting times after the first move in t = 0. Workers who move only once have a commuting time of 18.7 min in t = 0, while those who move a second time drive over 14 min longer to work after the first relocation. This results not only from the fact that second-time movers come from regions with longer commutes compared to movers, but also that they are more likely to move from rural regions with longer average commut- ing times. According to the background context effect, this leads to a higher toler - ance for commuting and thus to a longer chosen individual commuting time after the move. This could explain why especially these workers move again within the new region and reduce their commuting time by more than 13 min (see Table 5). 1 3 Persistence of commuting habits: context effects in Germany Table 5 Summary statistics of commuting time and wage Variable Mean Std. dev 25th perc 50th perc 75th perc Commuting time t = − 1 in minutes 19.9 17.1 7.6 15.1 26.4 Δ Commuting time t = 0 in minutes + 13.4 27.7 − 2.7 10.7 29.7 Δ Commuting time t = + 1 in minutes − 13.5 24.8 − 28.0 − 6.6 2.3 Wage t = − 1 (euros/day) 81.8 51.7 47.4 72.1 101.7 Δ Wage t = 0 (euros/day) + 14.3 36.4 − 0.6 9.2 27.5 Δ Wage t = + 1 (euros/day) + 4.9 29.9 0.2 3.6 9.09 N 4267 Means, standard deviation, 25th, 50th, 75th percentiles of commuting time and the wage. Comparison of second-time movers before and after the first move and after the second move Table 5 shows the difference between wages and commuting times before the first move t = − 1 after the first move t = 0 and after the second move t = + 1 for individu- als who moved a second time. As explained above, the increase in the commuting time after the first move is far higher for individuals moving twice than for those moving only once. Second-time movers increase their commuting time by over 13  min in t = 0. However, they shorten their commuting time by the same amount after the second relocation in t = + 1. This corrects the originally excessive commut- ing time, and confirms prediction 2. 5.2 Empirical analysis Prediction 1: the average commuting time in the region a person leaves has a positive influence on the individually selected commuting time in the destina- tion region In the following, I test the first prediction, in which I investigate how the aver - age commuting time in the region before the relocation influences the individually selected commuting time in the target region (Eq. 2). As workers may endogenously choose to move, I use a two-step regression (Heckman 1979). In the first step I esti- mate a probit regression for the decision to relocate (Eq.  3). The results for this probit regression are provided in the Web Appendix B and show, for example, that workers with higher wages, high-skilled workers and workers in western Germany are more likely to relocate. In the second step, I use the inverse Mill’s ratio from the first step as an additional control variable and analyze how the average commuting time in the region before the relocation influences the commuting time in the new region (Eq. 2). Table 6 shows the results of 4 specifications. According to model 1, which includes the lag of the individual commuting time t  −  1, the longer the commuting time was in the region before the relocation, the longer the individually selected commuting time is in the target region. In addition, the wage has a positive significant effect, which might be the result of compensatory wages for longer commutes as shown by Mulalic et al. (2014). In the second model I include the average commuting time in the region in which the previous place of 1 3 R. Jost Table 6 Individually selected commuting time after relocation NUTS-3 region Dependent variable: ln(C ) i,t=0 Model 1 Model 2 Model 3 Model 4 Ln(C ) 0.228*** 0.225*** 0.225*** 0.225*** i,t−1 (0.006) (0.006) (0.006) (0.006) 0.216*** 0.222*** 0.212*** Ln(C ) i,t=−1 (0.030) (0.029) (0.029) Inverse of Mill’s ratio* 0.620*** 0.560*** 0.143 0.531*** (0.205) (0.206) (0.144) (0.206) Ln(wage) 0.107*** 0.100*** 0.117*** (0.034) (0.034) (0.034) Ln(wage ) − 0.103*** t−1 (0.015) Medium-skilled 0.172*** 0.163*** 0.103** 0.165*** (0.047) (0.047) (0.042) (0.047) High-skilled 0.231*** 0.211*** 0.093 0.206*** (0.079) (0.079) (0.066) (0.079) Migrant − 0.106 − 0.098 − 0.044 − 0.092 (0.066) (0.066) (0.063) (0.066) Specialist/expert 0.036 0.035 0.032 0.037 (0.044) (0.044) (0.044) (0.044) Trained/professional assistant 0.003 0.002 0.004 0.005 (0.038) (0.038) (0.038) (0.038) Age groups Yes Yes Yes Yes Occupation dummies Yes Yes Yes Yes Industry dummies Yes Yes Yes Yes Occupational status Yes Yes Yes Yes Firm size (Number of workers) Yes Yes Yes Yes Year dummies Yes Yes Yes Yes Place of residence type Yes Yes Yes Yes Place ofwork type Yes Yes Yes Yes Constant − 0.667 − 1.030 0.841* − 0.560 (0.830) (0.827) (0.508) (0.830) N 45,232 45,232 45,232 45,232 N (cluster) 15,262 15,262 15,262 15,262 R 0.5773 0.5777 0.5775 0.5783 Adj. R 0.3607 0.3614 0.3611 0.3622 The table reports regressions of the individually selected log commuting times after the first relocation on the average log commuting time in the region before the relocation and control variables. Standard errors clustered by individuals, below parameter estimates. Levels of significance: *1%, **5%, ***10% *Inverse of Mill’s ratio is obtained from the first stage probit estimation of the move residence was located C as a proxy for commuting options observed in the past. i,t−1 Consistent with the first prediction, model 2 shows a positive significant effect on the individual commuting time. Moreover, the effect can be interpreted as causal, as I control for selectivity and unobserved heterogeneity, and can rule out the issue 1 3 Persistence of commuting habits: context effects in Germany Table 7 Adjustment of the NUTS-3 region Depend- commuting time in t + 1 ent variable: ln(C − C ) i,t+1 i,t=0 0.100* Ln(C − C ) i,t=0 i,t−1 (0.046) Change in ln(wage) 0.049 (0.108) Inverse of Mill’s ratio* 1.971*** (0.084) Constant − 2.729*** (0.084) N 4,135 0.3531 Adj. R 0.3526 The table reports the regression of the adjustment of the individually selected commuting time after the second move on the difference between the average commutes in the new and the old region. Stand- ard errors clustered by NUTS-3 regions, below parameter estimates. Levels of significance: *1%, **5%, ***10% *Inverse of Mill’s ratio is obtained from the first stage probit estima- tion of moving again within the new region of omitted variable bias and sorting (see Sect.  5.3). Hence, mobile workers com- ing from NUTS-3 regions with longer observed commutes have a greater tolerance for commuting and choose longer individual commutes in the target region. This indicates the presence of a context effect and is therefore consistent with the result obtained by Simonsohn (2006). However, a comparison of the effects with those found by Simonsohn (2006) shows that he overestimates the effect of the context (see Sect.  5.3 Table  8). This is because he does not include individual unobserved fixed effects. In addition, comparing R reveals that the model I consider performs much better than that of Simonsohn (2006) (0.36 vs. 0.15). Since commuting may be endogenous with respect to wages, model 4 excludes daily wages, which has little impact on the size of the coefficient of C . In addi- i,t−1 tion, in model 5 I include time-lagged wages t − 1. In this estimation, too, the result shows no change for the variable of interest C . i,t−1 Thus, the results indicate that workers’ current commuting behavior is affected not only by their own previous commuting time but also by the average commuting time in the region they moved from. Prediction 2: Individuals readjust their commuting times and move again when remaining in the new region If workers relocate from regions with longer commutes to regions with shorter average commuting times ( C > C ), they initially commute longer than the i,t−1 i,t=0 average in the target region. The reason for this is that they have a greater toler- ance for commuting as they come from regions where long commutes are common. Nevertheless, if they remain in the new region and observe fewer commutes, they 1 3 R. Jost Table 8 Robustness check: individually selected commuting time after the move NUTS-3 region Dependent variable: ln(C ) i,t=0 Model 1 Model 2 Model 3 Model 4 Model 5 Ln(C ) 0.531*** 0.226*** 0.226*** 0.227*** 0.225*** i,t−1 (0.004) (0.006) (0.006) (0.006) (0.006) 0.154*** 0.224*** 0.220*** 0.223*** 0.224*** Ln(C ) i,t=−1 (0.023) (0.029) (0.029) (0.029) (0.029) Inverse of Mill’s ratio* 0.092 − 0.009 0.292** − 0.021 (0.078) (0.032) (0.132) (0.032) Ln(wage) 0.061*** 0.088*** 0.034 (0.013) (0.027) (0.024) Medium-skilled 0.039** 0.119*** 0.078** (0.018) (0.040) (0.035) High-skilled 0.035 0.124** 0.040 (0.028) (0.062) (0.048) Migrant − 0.037* − 0.063 − 0.027 (0.019) (0.063) (0.060) Specialist/expert 0.026 0.016 0.027 (0.024) (0.044) (0.044) Trained/professional assistant 0.001 − 0.012 0.001 (0.021) (0.038) (0.038) Age groups Yes Yes Yes Occupation dummies Yes Yes Yes Industry dummies Yes Yes Yes Occupational status Yes Yes Yes Firm size (Number of workers) Yes Yes Yes Year dummies Yes Yes Yes Yes Yes Place of residence type Yes Yes Yes Yes Yes Place of work type Yes Yes Yes Yes Yes Constant 0.190 1.438*** − 0.042 1.476*** 1.194*** (0.310) (0.134) (0.533) (0.132) (0.139) N 45,232 45,232 45,232 45,232 45,232 N (cluster) 15,262 15,262 15,262 15,262 15,262 R 0.3415 0.5768 0.5763 0.5753 0.5776 Adj. R 0.3407 0.3606 0.3595 0.3586 0.3612 The table reports regressions of the individually selected log commuting times after the first relocation on the average log commuting time in the region before the move and control variables. Standard errors clustered by individuals, below parameter estimates. Levels of significance: *1%, **5%, ***10% *Inverse of Mill’s ratio is obtained from the first stage probit estimation of the move become dissatisfied with their initially chosen commutes and their desired commut- ing time changes. Therefore, I expect them to reduce their commutes by relocating again within the new region. To analyze the adjustment of the commuting time after a second move, I consider only individuals who move again within one year after 1 3 Persistence of commuting habits: context effects in Germany relocating to the new region. A total of 4,135 individuals move again within the new NUTS-3 region in t = 1. The regression estimates of Eq. 6 are presented in Table 7, where (C − C ) , i,t+1 i,t=0 the dependent variable, measures the change in the individual commuting time after the second and the first relocation. Therefore, it represents the adjustment of the individual commuting time between t = 0 and t = + 1. The key predictor is the differ - ence between the average commuting time in the new region and that in the previ- ous region (C − C ) . Moreover, as workers may endogenously choose whether i,t=0 i,t−1 to move a second time, I use a two-step regression (Heckman 1979): in the first step, I estimate a probit regression for the decision to relocate a second time in the new region (Eq.  3). The results of this probit regression can be found in the Web Appendix C. They show, for example, that the greater the difference between the average commuting time and the individual’s own selected commuting time in the target region, the more likely a second move is. In the second step, I use the inverse Mill’s ratio from the first step as an additional control variable. The results are pre- sented in Table 7 and are seen to be in line with prediction 2, the greater the differ - ence between the new and the old region (C − C ) the stronger the adjustment i,t=0 i,t−1 of the individually chosen commuting time after the second move is. Comparing the estimated effect of  (Table 7) with the estimation of  in prediction 1 (Table 6 2 2 model 2) it can be seen that the coefficient  of the first prediction is twice as large as  in the second prediction. Thus, second-time movers do not fully reverse the original impact of C , but it is moving in that direction. i,t−1 With this result, I can therefore rule out an explanation for the commuting behav- ior that is based on stable unobserved differences across movers from different regions, as individuals readjust their commuting time by moving again within the new region—they adopt the commuting behavior of the new region. 5.3 Robustness checks Although the presence of stable unobserved differences can be ruled out by con- firming prediction 2, there could be other explanations for the presented results and several issues that might influence the outcome, such as unobserved heterogeneity, omitted variable bias, selectivity, and sorting. However, in the following, I am able not only to reject other explanations, but also to confirm my results by means of several robustness checks. Therefore, the effect of C on C can be interpreted i,t−1 i,t=0 as causal. 5.3.1 Unobserved heterogeneity In fact, unobserved heterogeneity can have an influence on the estimates of C , i,t−1 thereby driving the effect of the context (see Sect.  4). To deal with this issue, I include individual fixed effects in my analysis (see Eq.  2). This is especially impor- tant, and failure to do so generates a bias. This can be observed in Table 8 (model 1). Excluding individual fixed effects overestimates the effect of the individual previous commuting time C , and underestimates the influence of the context of previously i,t−1 1 3 R. Jost observed commutes C . It is therefore important to include individual fixed i,t−1 effects. Failure to do so leads to a bias, as in the study by Simonsohn (2006) which does not include individual fixed effects in the analysis and therefore underestimates the effect of the context. 5.3.2 Omitted variable bias In addition, I conduct several robustness checks excluding individual and firm char - acteristics. In model 2 (Table 8) I exclude firm characteristics, which yields similar results for the context of previously observed commutes to those in Table 6 (model 2), which included all control variables. Also, almost the same results are obtained when firm characteristics are excluded and when both individual and firm character - istics are excluded (models 3 and 4). Thus, the results on the previous average com- muting time are very robust and do not seem to be influenced by observed individual or firm characteristics. This leads me to conclude that there is no evidence of omit- ted variable bias. 5.3.3 Selectivity To control for the selectivity of a relocation—as workers may endogenously choose to relocate—I use a two-step Heckman selection model (Heckman 1979), in which I control for the selectivity of a relocation (Eq. 3). To gain an impression of whether selectivity is important I estimate the model without controlling for selectivity. The results are provided in Table 8 (model 5) and show almost the same effects for pre- viously observed commutes as those in Table 6 (model 2). Only the coefficients for wages and the skill-level variables change. Thus, controlling for the selectivity of the relocation is not important for interpreting the variable of interest but influences other control variables. 5.3.4 Sorting Another issue might be sorting, as workers select themselves into certain regions because of their taste for commuting. To address this issue, I run a reversed regres- sion of Eq. 5. In line with the definition of sorting, I should find a positive correla- tion between the average commuting time in the destination region and the indi- vidual commuting time in the region before the move. However, my results show no significant effect of the average commuting time in the destination regions (Table  9). Thus, there is no sign of a sorting process—individuals do not select themselves into regions because of their taste for commuting—but this once again shows the presence of the context effect. Moreover, workers might also move for job-related reasons, such as higher wages. As wages are highly correlated with commuting in theory, I consider only work- ers who earn almost the same wage before and after the first relocation. Table  10 Wages are rounded to the nearest ten. 1 3 Persistence of commuting habits: context effects in Germany Table 9 Robustness check: NUTS-3 region Depend- individuals select themselves ent variable into regions because of their ln(C ) i,t−1 taste for commuting Ln(C ) 0.082*** i,t=0 (0.008) − 0.109 Ln(C ) i,t=0 (0.077) 0.951*** Ln(C ) i,t−1 (0.059) Ln(wage) 0.086*** (0.021) Medium-skilled 0.049* (0.029) High-skilled 0.078** (0.034) Migrant − 0.109** (0.043) Specialist/expert 0.053 (0.047) Trained/professional assistant − 0.005 (0.042) Woman − 0.070*** (0.019) Age groups Yes Occupation dummies Yes Industry dummies Yes Occupational status Yes Firm size (Number of workers) Yes Year dummies Yes Place of residence type Yes Place of work type Yes Constant − 0.658** (0.297) N 15,262 R 0.056 0.0520 Adj.R The table reports the regression of the individual commuting time in the previous region on the average commuting time in the target region (after the relocation). Standard errors clustered by NUTS-3 regions, below parameter estimates. Levels of significance: *1%, **5%, ***10% shows that the average commuting time in the region before the move has a positive and significant influence on the commuting time of workers who do not achieve an increase in wages after the relocation. This indicates that endogeneity issues with respect to wages do not drive the results. In addition to restricting the sample to per- sons earning the same wage before and after relocating, I also restrict it to workers 1 3 R. Jost Table 10 Robustness check: NUTS-3 region (1) Depend- (2) Depend- movers, who earn almost the ent variable ent variable same wage before and after ln(C ) ln(C ) i,t=0 i,t=0 relocating (1) and who have the same wage as well as the same Ln(C ) 0.202*** 0.199*** i,t−1 task level (2) before and after (0.013) (0.015) relocating 0.381*** 0.361*** Ln(C ) i,t=−1 (0.071) (0.081) Ln(wage) − 0.164* − 0.125 (0.088) (0.109) Medium-skilled − 0.089 − 0.073 (0.102) (0.117) High-skilled − 0.033 − 0.042 (0.151) (0.196) Migrant 0.161 0.060 (0.108) (0.138) Specialist/expert 0.066 0.014 (0.103) (0.197) Trained/professional assistant 0.063 0.018 (0.081) (0.155) Age groups Yes Yes Occupation dummies Yes Yes Industry dummies Yes Yes Occupational status Yes Yes Firm size (Number of workers) Yes Yes Year dummies Yes Yes Place of residence type Yes Yes Place of work type Yes Yes Constant 1.745*** 1.687*** (0.449) (0.559) N 9,193 7,473 N (cluster) 3,094 2,514 R 0.5797 0.5833 Adj. R 0.3603 0.3645 The table reports regressions of the individually selected log com- muting time after the first relocation on the average log commuting time in the region before the move and control variables. Standard errors clustered by individuals, below parameter estimates. Levels of significance: *1%, **5%, ***10% who do not change their task level. Once again, the coefficient of the average com- muting time in the previous region does not change. To sum up, the robustness checks show that it is crucial to include the individual fixed effects when investigating the individual commuting behaviors. In addition, the robustness checks indicate that my results on the average commuting time are not driven by omitted variable bias—as the coefficient is very robust when indi- vidual and firm-specific characteristics are excluded. Furthermore, sorting does not 1 3 Persistence of commuting habits: context effects in Germany seem to influence my results, either. Therefore, the investigated influence of the pre- viously observed commutes on the individually chosen commuting time (Table  6) can be interpreted as causal. 5.4 Eecfft heterogeneity In the following, I investigate the heterogeneous effects of the context on the indi- vidual selected commuting time. I differentiate movers by different age groups, skill levels, and gender. In addition, I consider movers between different types of regions—urban and rural areas—as well as movers between labor market regions. 5.4.1 Age groups, gender, and skill level Since it is possible that individuals differ in their behavior due to their age, gender, or skill level, I take up this point by performing the estimation for different interac- tions (Web Appendix D). In particular, I interact the average commuting time in the region before the relocation C with age, gender, and skill level. The results i,t=−1 show no significant group differences in terms of age and skill level. Nor can any significant differences be observed between women and men. Thus, there is no effect heterogeneity for different groups regarding age categories, skill level, or gender. 5.4.2 Movers between different types of rural and urban regions Considering movers between different types of place of residence, I interact the average commuting time in the previous location C with the different types of i,t=−1 rural and urban regions before and after the move. The results are shown in the Web Appendix E and indicate that the effect of the context of previously observed commutes is strongest for those moving to urban areas, especially for the group moving from a rural to an urban area. This is related to the fact that workers who previously lived in a rural area with long average commutes are used to commuting long distances. Therefore, when moving to urban regions such workers have a higher tolerance for commuting and choose longer than average commutes in the urban region. However, for movers to rural areas the results indicate a smaller or insignifi- cant effect of the context. The reason could be that the majority of workers moving from urban to rural areas do not only relocate their place of residence but also take up a new job in the rural area. Thus, other conditions, such as job availability, are more important than commuting preferences for this group of movers. Hence, the results indicate that the size of the effect of the context depends par - ticularly on the region type of the place of residence before and after the reloca- tion. Considering only movers between metropolitan areas (Simonsohn 2006) might therefore lead to a bias in the estimated effect. I consider only the location of the place of residence. However, as I cannot take into account dense traffic or congestion when calculating the commuting time, the results might underestimate the commuting costs especially in dense urban areas. 1 3 R. Jost 5.4.3 Labor market regions Next, I show the results for individuals moving between German labor market regions (Kosfeld and Werner 2012). The restrictions are the same as for movers between NUTS-3 regions, i.e., workers have to relocate both their place of work and their place of residence to a different German labor market region. Moreover, the labor market region of the place of work and the place of residence must be constant for two years before and after the move. In contrast to the consideration of indi- viduals moving between NUTS-3 regions, I calculate the average commuting time at the level of labor market regions (as a proxy for previously observed commuting options). The results are shown in Web Appendix F and are comparable with the effect of the context for persons moving between NUTS-3 regions (Table 6). 6 Conclusion This study investigates for the first time commuting behavior in terms of a behav - ioral economic concept based on geo-referenced data for Germany. The basis of this investigation is the approach developed by Simonsohn (2006), who examines commuting behavior for the US. However, I can show that his estimated effects are biased due to the absence of individual fixed effects and the consideration only of individuals moving between metropolitan areas. The presented results show that workers’ commuting decisions are influenced by commuting options observed in the past. This explains why individuals who move from different regions to one and the same region initially commute differently: indi- viduals moving from areas with long average commutes have a greater tolerance for commuting and therefore commute more than individuals coming from regions with shorter commutes. However, if they remain in the new region, they adjust their initially chosen commuting times to the average commutes in the new region. This refutes the assumption of stable unobserved differences across individuals. Instead, individuals change their marginal utility of commuting when moving to a new region, as they adjust their commuting time by means of a second relocation within the new region. The reason for this behavior is the change in the context: the origi- nal context was seen as the average commuting time in the previous region, but the context changes with the relocation to a new region. Thus, commuting preferences change. In addition, the results indicate that selectivity and sorting do not influence the effect of the context, but it is crucial to include individual fixed effects. Moreo- ver, the context has different effects depending on the region type of the place of residence: the context effect is greatest for those moving from rural to urban areas. However, the travel time budget can be influenced by remote work that increased during the Covid-19-shock and might increase the expected and acceptable commut- ing distance. Future research could examine whether such increase in remote work influences the effect of the context. Additionally, for future investigation that exam- ine consumer preferences and other labor market decisions, the study highlights the importance of identifying the context of previously observed options and including 1 3 Persistence of commuting habits: context effects in Germany them in the analysis. Finally, the results indicate the essentiality of including indi- vidual fixed effects, as they influence the outcome of commuting decisions. Supplementary Information The online version contains supplementary material available at https:// doi. org/ 10. 1007/ s00168- 023- 01223-4. Acknowledgements I thank Peter Haller, Wolfgang Dauth, Joachim Möller, the ERSA Congress 2019 and the BGPE Workshop in Regensburg (2019) for many helpful comments on this project. Further I appreciate very detailed and helpful comments by Janet Kohlhase and two anonymous reviewers. Funding Open Access funding enabled and organized by Projekt DEAL. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Com- mons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. References Alonso W (1964) Location and land use. Toward a general theory of land rent. Harvard University Press, Cambridge Bettman J, Luce FL, Payne J (1998) Constructive consumer choice processes. J Consumer Res 25:187–217 Bhargava S, Fisman R (2014) Contrast effects in sequential decisions: evidence from speed dating. Rev Econ Stat 96:444–457 Brueckner JK (2000) Urban sprawl: diagnosis and remedies. Int Reg Sci Rev 23:160–171. https:// doi. org/ 10. 1177/ 01600 17007 61012 710 Brueckner JK, Stastna L (2020) Commuting and migration. J Reg Sci 60:853–877. https:// doi. org/ 10. 1111/ jors. 12498 Brunow S, Gründer M (2013) The impact of activity chaining on the duration of daily activities. Trans- portation 40:981–1001 Card D, Heining J, Kline P (2013) Workplace heterogeneity and the rise of West German wage inequality. Q J Econ 128:967–1015 Dargay JM, Clark S (2012) The determinants of long distance travel in Great Britain. Transp Res Part A Policy Pract 46:576–587. https:// doi. org/ 10. 1016/j. tra. 2011. 11. 016 Dauth W, Haller P (2020) Is there loss aversion in the trade-off between wages and commuting distances? Reg Sci Urban Econ. https:// doi. org/ 10. 1016/j. regsc iurbe co. 2020. 103527 Destatis (2017) Commuting in germany: 68% use a car to commute to work. https:// www. desta tis. de/ DE/ Zahle nFakt en/ ImFok us/ Arbei tsmar kt/ Pendl erArb eitsw eg. html Duan Y, Jost O, Jost R (2022) Beyond lost earnings: the long-term impact of job displacement on work- ers’ commuting behavior. IAB-Discussion Paper, 15/2022 Frey B, Stutzer A (2007) Commuting and life satisfaction in Germany. Informationen zur Raumentwick- lung. Heft 2/3 Gimenez-Nadal JI, Molina JA, Velilla J (2018) Spatial distribution of US employment in an urban effi- ciency wage setting. J Reg Sci 58:141–158. https:// doi. org/ 10. 1111/ jors. 12351 Gimenez-Nadal JI, Molina JA, Velilla J (2020) Trends in commuting time of European workers: a cross- country analysis. IZA Discussion Paper No 12916 Hanson S, Johnston I (1985) Gender differences in work trip lengths: Implications and explanations. Urban Geogr 6:193–219. https:// doi. org/ 10. 2747/ 0272- 3638.6. 3. 193 1 3 R. Jost Hartzmark S, Shue K (2017) A tough act to follow: contrast effects in financial markets. NBER Working Paper No. 23883 Heckman J (1979) Sample selection bias as a specification error. Econometrica 47:153–161. https:// doi. org/ 10. 2307/ 19123 52 Heuermann D, Assmann F, Freund F, vom Berge P (2016) The distributional effect of commuting subsi- dies-evidence from geo-referenced data and large-scale policy reform. Reg Sci Urban Econ 67:11– 24. https:// doi. org/ 10. 1016/j. regsc iurbe co. 2017. 08. 001 Huber S, Rust C (2016) osrmtime: calculate travel time and distance with OpenStreetMap data using the open source routing machine (OSRM). Stand Genomic Sci 16:416–423. https:// doi. org/ 10. 1177/ 15368 67X16 01600 209 Huber J, John P, Puto C (1982) Adding asymmetrically dominated alternatives: violations of regularity and the similarity hypothesis. J Consumer Res 9:90–98. https:// doi. org/ 10. 1086/ 208899 Kahneman D, Tveresky A (1979) Prospect theory: an analysis of decision under risk. Econometrica 4:263–287 Kosfeld R, Werner A (2012) German labour markets—new delineation after the reforms of Ger- man district boundaries 2007–2011. Raumforsch Raumordn 70:49–64. https:// doi. org/ 10. 1007/ s13147- 011- 0137-8 Lux M, Sunega O (2012) Labor mobility and housing: the impact of housing tenure and housing afford- ability on labor migration in the Czech Republic. Urban Stud 4:489–504. https:// doi. org/ 10. 1177/ 00420 98011 405693 McQuaid R, Chen T (2012) Commuting times—the role of gender, children and part-time work. Transp Econ 34:66–73. https:// doi. org/ 10. 1016/j. retrec. 2011. 12. 001 Mulalic I, van Ommeren J, Pilegaard N (2014) Wages and commuting: quasi-natural experiments´ evi- dence from firms that relocate. Econ J 124:1086–1105. https:// doi. org/ 10. 1111/ ecoj. 12074 Ostermann K, Eppelsheimer J, Gläser N, Haller P, Oertel M (2022) Geodata in labor market research: trends, potentials and perspective. J Labor Mark Res 56:1–17 Ross SL, Zenou Y (2008) Are shirking and leisure substitutable? An empirical test of efficiency wages based on urban economic theory. Reg Sci Urban Econ 38:498–517 Rouwendal J, Rietveld P (1994) Changes in commuting distances of Dutch households. Urban Studies 31:1545–1557 Ryder H, Heal G (1973) Optimal growth with intertemporally dependent preference. Rev Econ Stud 40:1–33 Shuai X (2012) Does commuting lead to migration? J Reg Anal Policy 42:237–250 Simonsohn U (2006) New Yorkers commute more everywhere: contrast effects in the field. Rev Econ Stat 88:1–9. https:// doi. org/ 10. 1162/ rest. 2006. 88.1.1 Simonsohn U, Loewenstein G (2006) Mistake #37: the effect of previously encountered prices on current housing demand*. Econ J 116:175–199. https:// doi. org/ 10. 1111/j. 1468- 0297. 2006. 01052.x Simonson I, Tveresky A (1992) Choice in context: tradeoff contrast and extremeness aversion. J Mark Res 29:281–295 Tveresky A, Kahneman D (1991) Loss aversion in riskless choice: a reference-dependent model. Q J Econ 106:1039–1061 Van Ham M, Hooimeijer P (2009) Regional differences in spatial flexibility: long commutes and job related migration intentions in the Netherlands. Appl Spat Anal Policy 2:129–146. https:// doi. org/ 10. 1007/ s12061- 008- 9016-2 Van Ommeren J (2005) Commuting: the contribution of search theory. Emerald Group Publishing Lim- ited, Bingley, pp 347–380. https:// doi. org/ 10. 1108/ S0573- 8555(2005) 00002 66013 Van Ommeren J, Fosgerau M (2009) Workers’ marginal costs of commuting. J Urban Econ 65:38–47. https:// doi. org/ 10. 1016/j. jue. 2008. 08. 001 Van Ommeren J, van den Berg G, Gorter C (2000) Estimating the marginal willingness to pay for com- muting. J Reg Sci 40:541–563. https:// doi. org/ 10. 1111/ 0022- 4146. 00187 Zabel J (2012) Migration, housing market, and labor market responses to employment shocks. J Urban Econ 72:267–284. https:// doi. org/ 10. 1016/j. jue. 2012. 05. 006 Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. 1 3 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The Annals of Regional Science Springer Journals

Persistence of commuting habits: context effects in Germany

Jost, Ramona
The Annals of Regional Science , Volume 72 (3) – Mar 1, 2024
Download PDF
26 pages

Loading next page...
 
/lp/springer-journals/persistence-of-commuting-habits-context-effects-in-germany-7KjgotSK7v

References (52)

Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2023
ISSN
0570-1864
eISSN
1432-0592
DOI
10.1007/s00168-023-01223-4
Publisher site
See Article on Publisher Site

Abstract

In this study, I investigate the commuting behavior of workers in Germany. Using comprehensive geo-referenced administrative employee and firm data, I can cal- culate the exact commuting time and the distance between workers’ residence and workplace locations. Based on a behavioral economic approach (Simonson and Tveresky in J Mark Res 29:281–295, 1992), I show that individual commuting deci- sions are influenced by wages and individual heterogeneity as well as depending on the context individuals observed in the past. In particular, my results show that previously observed commutes have an impact on subsequent commuting behavior: workers choose longer commuting times in the region they recently moved to when the average commute in the region they left was longer. The results indicate that while selectivity and sorting do not influence the effect of the context, the inclusion of individual fixed effects is crucial. JEL Classification J60 · R10 · R19 · R23 1 Introduction The importance of commuting is growing rapidly—both the number of commuters and the distance they commute are growing steadily (Gimenez-Nadal et  al. 2020). From an economic perspective, commuting is essential for a well-functioning labor market as it is an important measure to overcome spatial separations (Lux and Sunega 2012; Zabel 2012). At the individual level, commuting implies better labor accessibility and subsequently improves job and career opportunities, leading to bet- ter outcomes and improved individual utility. However, commuting also has negative impacts on both the environment and the infrastructure (Brueckner 2000; Rouwen- dal and Rietveld 1994), as well as on individuals’ well-being as it is associated with congestion and high costs (Frey and Stutzer 2007). Understanding the determinants * Ramona Jost ramona.jost@iab.de Institute for Employment Research (IAB), Regensburger Str. 104, 90478 Nuremberg, Germany Vol.:(0123456789) 1 3 R. Jost of and the reasons for commuting is thus an important topic for policymakers deal- ing with economic and labor market issues. Studies on commuting find different factors and effects that influence individu- als’ commuting behavior, for example commuting is more common among males and among workers with higher incomes as well as among homeowners. The same applies to workers who are older and work in specific occupations and have specific skill levels (Gimenez-Nadal et  al. 2018, 2020; Ross and Zenou 2008; Hanson and Johnston 1985; Dargay and Clark 2012; McQuaid and Chen 2012). However, individuals’ commuting behavior might also be explained from a behavioral economic perspective. In particular, previous research shows that previ- ously observed options can influence individuals’ perceptions and therefore their subsequent decision-making behavior (Simonson and Tveresky 1992). Applied to individuals’ commuting behavior this means that previously observed commuting options influence their preferences for commuting and consequently their own com- muting decisions. This approach can explain, for example, why individuals who move to Munich commute 30 percent less than the average in Munich if they come from regions with shorter average commuting times, while individuals commute 35 percent more than the average in Munich if they previously lived in regions with longer commuting times than those typical in Munich. This might indicate that com- muting decisions are influenced by the context of commuting options observed in the past, such as other individuals’ commutes. This study analyzes such commuting behavior, based on the study conducted by Simonsohn (2006) for the US, and contributes to the literature in at least four ways: first, it contributes to the literature on commuting behavior and the factors that are important for explaining commuting (Gimenez-Nadal et al. 2020; Dargay and Clark 2012; McQuaid and Chen 2012). In particular, I show that the context of commuting options observed in the past is crucial for analyzing individuals’ commuting behav- ior. In this context I show that the results obtained by Simonsohn (2006) are biased due to the omission of individual fixed effects and the consideration only of migrants between two metropolitan areas. Second, I reveal effects for different groups, dis- cussing effect heterogeneity for age, gender, skill level, as well as rural and urban areas, for an entire country. Third, I use geo-referenced employer-employee data. These administrative registry data possesses higher validity than survey data and provides precise information about individuals’ residence and workplace locations with a high number of observations. This makes it possible to calculate the exact commuting distance and time for German workers. Fourth, the study contributes to the migration literature (van Ham and Hooimeijer 2009; Brueckner and Stastna 2020; Shuai 2012). In particular, I show that the greater the difference between a worker’s individual commuting time and the average commuting time at their place of residence, the more likely they are to move again. When individuals choose where to live, they face the difficult decision of how far they are willing to commute, weighing up the benefits and costs of commut- ing. Advantages of commuting may include cheaper rents and housing prices out- side the city center, resulting in a higher disposable income. Furthermore, com- muting can provide more job opportunities for individuals who live in rural areas where there may be no or no adequate employment offers. However, commuting 1 3 Persistence of commuting habits: context effects in Germany also has disadvantages; it takes up time, causes stress, and impacts the reconcili- ation of work and family. It can therefore have a negative effect on individuals’ well-being (Frey and Stutzer 2007). When deciding how far they wish to com- mute, individuals have to trade off the benefits with the disutility of commuting. Indeed, costs and benefits do not have the same effect on utility: the response to losses is stronger than the response to the corresponding benefits (loss aversion, Kahneman and Tveresky 1979). In the context of commuting decisions, however, Dauth and Haller (2020) find no sign of loss aversion, which contradicts previous experimental evidence (Tveresky and Kahneman 1991). Empirical evidence from urban economics reveals the disutility of commuting for which individuals wish to be compensated. For the Netherlands, van Omme- ren et al. (2000) and van Ommeren (2005) find a marginal willingness to pay for an additional kilometer of commuting of 0.15 euros per day or 17 euros for one additional hour of commuting (van Ommeren and Fosgerau 2009). With regard to compensation by the employer, Heuermann et  al. (2016) find that employers compensate only few employees directly for additional commuting costs. Hence, the decision to commute is mainly an individual one, which can be strongly influ- enced by prior experiences. However, individuals are often unable to assess correctly the disutility of com- muting and are frequently uncertain about their preferences, which contradicts the standard economic theory (Kahneman and Tveresky 1979). Instead, they form their preferences as and when they are needed, for instance when making choices (Bett- man et  al. 1998). For example, in the context of commuting decisions, individuals rely on a wide range of possible cues, such as other individuals’ commutes. Moreo- ver, in the literature on decision-making (Bettman et al. 1998; Huber et al. 1982) it becomes fundamental that an individual’s decision can be influenced by the context: individuals interpret information by comparing it not only to other available options, but also to what was recently observed. According to Hartzmark and Shue (2017), these context effects have the potential to affect a variety of important real-world decisions. They not only distort judicial perceptions of the severity of crimes, lead- ing to unfair sentencing, but also affect employee hiring, medical diagnoses as well as housing and commuting decisions. The context effect that is relevant for this study is the background context effect, according to which choices depend on options encountered in the past—preferences can change with the history of choices. The intuition behind this is that the same product may seem more attractive against the background of less attractive alter- natives and unattractive compared to more attractive alternatives (Simonson and Tveresky 1992). Simonson and Tveresky (1992) document this effect in an experi- ment comprising two stages in which subjects have to make choices in sequence. In the first stage, half of the subjects are confronted with two options that have a rela- tive high cost for one attribute, and the other half should make a choice with a rela- tively low cost for the same attribute. In the second stage, all subjects are confronted with the same choice. In line with the background context effect, subjects who are confronted with a relatively high cost for an attribute in the first stage are more likely to choose the more expensive option in the second stage because it appears cheaper to them. 1 3 R. Jost There is ample evidence of the background context effect. Bhargava and Fisman (2014) demonstrate this effect in the context of speed dating. They show that the attractiveness of previous partners reduces the probability of finding a date. Moreo- ver, Hartzmark and Shue (2017) demonstrate that today’s earnings impress investors more when previous earnings were poor. Furthermore, Simonsohn and Loewenstein (2006) present the effect with regard to housing choices: individuals who move from cities with relatively high housing costs are more likely to pay higher prices in the new city compared to individuals coming from cities with cheaper markets. Applied to commuting behavior, this means that commuting options encountered by indi- viduals in the past affect their current commuting decisions. However, relatively lit- tle research has been conducted into when and why the background context effect influences commuting decisions. The only such study was conducted by Simonsohn (2006). He considers individuals relocating between two metropolitan areas in the US and takes the average commuting time in the previous city as a proxy for com- muting options encountered in the past to examine how previously observed com- mutes influence commuting decisions when moving to a new city. He finds that individuals choose longer commutes in the new city, the longer the average com- mute was in the city they came from. Commuting decisions are thus influenced by commuting options encountered by individuals in the past, which is in line with the background context effect. In this study I consider workers who relocate between NUTS-3 regions in Ger- many and examine the context effect for workers of an entire country, which is why I deviate from the approach of Simonsohn (2006) and use the average commuting time at the NUTS-3 level for the proxy of commuting options encountered in the past. The results show that individuals coming from backgrounds with longer aver- age commuting times initially choose longer individual commutes in the destination region compared to individuals from regions with shorter average commutes. In contrast to Simonsohn (2006), I additionally differentiate between individuals moving between different region types of rural and urban regions and thus I show that the context effect is strongest for workers who move from rural to urban areas. Further, the robustness checks show that selectivity of a relocation does not influ- ence the effect of the context and I find no evidence of workers selecting themselves into regions because of their taste for commuting. However, my results do indicate that it is very important to control for individual fixed effects. Moreover, I find no sign of stable taste difference as traditional economic theory would suggest. The remaining paper is structured as follows. Section  2 provides the theoretical motivation for the background context effects. Section  3 discusses the data and the sample. The identification strategy used is shown in Sect.  4. The empirical results are presented in Sect. 5, and Sect. 6 concludes. 2 Theoretical motivation for the background context effects As empirical evidence shows, decisions are preference-dependent (Bettman et  al. 1998; Huber et  al. 1982; Hartzmark and Shue 2017; Bhargava and Fisman 2014; Simonsohn and Loewenstein 2006). However, these preferences change with 1 3 Persistence of commuting habits: context effects in Germany previously observed options. As Tveresky and Simonson (1992) demonstrate in their background contrast experiment, individuals’ previous experiences influence their perceptions and therefore their subsequent decision-making behavior. For commut- ing decisions, this implies that commuting options encountered previously affect current commuting preferences and thus individuals’ commuting behavior. The fol- lowing approach is based on this concept, which is also used by Simonsohn (2006). The idea is that the disutility of commuting decreases when a person was only con- fronted with longer commuting options in the past, whereas, the disutility increases when individuals were only exposed to short commutes. To investigate this approach and to measure the effect of the context, I use relo- cations involving individuals moving between two NUTS-3 regions in Germany. According to the background contrast experiment conducted by Tveresky and Simonson (1992), the commuting behavior after the move should be affected by pre- viously observed commuting options. This concept is formally represented as: = (1 − ) + (1) t−1 t with β∈ [0, 1]. Abstracting all other influences, such as sociodemographic factors, represents a person’s individually chosen commuting time as a weighted sum of the observed commuting options in the present  and the past  , with the weights t t−1 decreasing exponentially into the past (Ryder and Heal 1973). More precisely, under the assumption of β = 1 there is no impact of commutes observed in the past on the current commuting time, since  =  and thus no impact of the context. In contrast, if β = 0 the current commuting preferences are determined only by the previously observed commuting times, corresponding to  =  . In the following, I expect β t−1 to take values between 0 and 1 (0 < β < 1), such that two otherwise identical indi- viduals with different numbers of previously observed commuting options will have different levels of  when moving to the same region. Moreover, I use the average commuting time in the region of residence before the move as a proxy for previously observed commuting options (Simonsohn 2006) . According to Eq.  (1), individu- als moving from regions with longer average commutes accept a longer commuting time  when choosing places of work and residence in the destination region com- pared to individuals coming from regions with shorter average commuting times. This is the first prediction I investigate in this study. The average commuting time in the region a person leaves has a positive influ- ence on the individually selected commuting time in the destination region However, if individuals stay in the new region and observe the commut- ing options in the new region, their preferences for commuting change due to the new observed commutes in the new region. This leads to a change in the desired commuting duration. For example, movers who relocate from regions with longer In contrast, Simonsohn (2006) uses the average commuting time on the city level, as he only analyzes movers between two metropolitan areas. Thus, while the predictions are quite similar to those of Simon- sohn (2006), the objects of investigation differ due to the different target group of movers. 1 3 R. Jost commutes to regions with shorter ones initially have a greater tolerance for long commutes and prefer cheaper and larger living space outside the city center. There- fore, they initially commute longer than the average commute in the new region. If they remain in this region and observe shorter commutes, however, their preferences for shorter commutes grow and the disutility for commuting increases. They thus become dissatisfied with the commutes they chose initially and might move again within the new region to reduce their commuting time, thereby correcting an origi- nally excessive amount of commuting. This relationship is illustrated by the second prediction. Individuals readjust their commuting times and move again when remaining in the new region The second prediction is therefore useful for ruling out explanations based on stable unobserved differences across individuals who move from different regions. Because if individuals who come from regions with longer average commutes travel more after relocating because they are different from those coming from regions with shorter average commutes, I would not expect them to revise their commutes by moving again. 3 Data and sample selection 3.1 Data For the analysis, I use the employment biographies of a 6-percent random sample of all German workers subject to social security contributions. The administrative registry data does not include self-employed persons or civil servants; however, it covers more than 80 percent of the German labor force. The Employment History (BeH – Beschäftigenhistorik V10.01.00, 2016) collated by the Institute for Employ- ment Research (IAB) provides exact information about periods of employment based on the status reports submitted to the pension insurance. Besides the sociode- mographic characteristics, information at the firm level are included, which comes from the Establishment History Panel (BHP). This dataset contains information about the branch of industry, the establishment location, number of employees and marginal part-time employees. As daily wages are top-coded at the social security contribution ceiling, I use the imputation procedure developed by Card et al. (2013) to recover wages above this threshold. A unique feature of this dataset is the supplement IEB GEO, which provides anonymized address information in the form of geocodes for the locations of an individual’s residence and place of work for the years 2000–2014 (Ostermann et al. 2022). Combining this address information with road network data from Open- StreetMap, I calculate door-to-door commuting distances (Huber and Rust 2016; Dauth and Haller 2020; Duan et al. 2022). It is only possible to determine distances for individuals traveling by car in this way; those for users of public transport may differ. However, the car is the most important mode of transport. Almost 70 percent 1 3 Persistence of commuting habits: context effects in Germany of workers commute to work by car (Destatis 2017), whereas only 14 percent of commuters use the public transport system. In addition, to calculate the commuting time I take average values for highways, primary, and residential roads. By using geocodes, the commuting time is not limited by administrative units, which reduces measurement error for individuals close to administrative borders and mitigates the problem of spatial sorting within areas. Yet, using driving time can cause issues regarding the experienced commuting time: for example, the algorithm cannot rec- ognize dense traffic in the daily rush hours. Nevertheless, as the time is measured before and after the regional move, the change in the duration might be affected less by this measurement problem. 3.2 Sample In this study, I investigate the commuting behavior of German workers, excluding persons in marginal and part-time employment as well as workers older than 57 and younger than 18 years of age. Regarding the commuting time, I restrict the sample to workers with a commuting time between 1 and 90  min. I choose 1  min as the minimum because this represents the first percentile of the data and hence ensures that outliers who do not commute are not considered. The restriction to 90  min is because the data does not provide any information about the number of commuting trips. Thus, the data could also include workers who commute weekly and have a second place of residence. To exclude those workers, I restrict the data to workers with commuting times of up to 90 min. This is comparable to other German studies that restrict the commuting distance to 100 km (Dauth and Haller 2020; Duan et al. 2022) and ensures that commuting is conducted on a daily basis. To test prediction 1, whether the average commuting time in the region a per- son leaves has a positive influence on the individually selected commuting time in the destination region, several restrictions have to be considered. First, to be able to analyze commuting decisions, I have to consider only those individuals who face such a decision. This group comprises individuals who are required to make a new commuting decision due to moving home or changing their job. For my study, how- ever, I consider individuals who simultaneously change both their place of residence and their place of work. The reason for this is, first, that for individuals who only change their place of work it is not possible to examine the influence of the con- text of commutes observed in the past, because for job changers the region of the place of residence does not change. Second, if individuals only change their place of residence they might, for example, be relocating due to dissatisfaction with com- muting and I would therefore not be able to identify the influence of the context However, the results obtained by Simonsohn (2006) show that the context has almost the same effect for people who use public transport. For the analysis in this study I consider commuting time. However, all the results are very similar when commuting distance is used. In a robustness check, I investigate the effect of the context for this group, then also provide evidence of a context effect for this group of movers. 1 3 R. Jost correctly. To avoid this, I restrict the sample to workers who change both residence and workplace locations, which further guarantees a relocation of the entire center of their lives. In addition, I restrict the sample to those movers who relocate between two of the 402 German NUTS-3 regions. I also keep the NUTS-3 region of the place of work and the place of residence constant for two years before and after the move. This guarantees that movers are able to adopt the commuting options as well as the commuting behavior of the region they lived in. In addition, this assumption means that it is possible for movers to relocate again within the target region to read- just their initially chosen commuting time. After these restrictions I identify 15,671 workers who move between two NUTS-3 regions. Furthermore, the time periods are categorized to t − 1 for the year before the move, t = 0 for the year of the relocation and t + 1 for the year after the move. To test prediction 2, I look at workers who relocate again within the new region in period t + 1 (one year after the move), keeping the place of work constant. The number of second-time movers is 4267. 4 Identification strategy To test the first prediction, I estimate how the average commuting time in the region of residence before the relocation C influences the individually chosen commut- i,t−1 ing time in the target region C , I consider a dynamic fixed effects model, where i,t=0 the lag of the dependent variable C is used as an explanatory variable : i,t−1 ln(C )=  ln(C )+  ln(C )+  X +  + (2) i,t=0 1 i,t−1 2 i,t−1 3 i i,t i,t where ln(C ) represents the dependent variable, the logarithm of the individual i,t=0 chosen commute in minutes after the relocation t = 0, while ln(C )—the lag of i,t−1 the dependent variable—is added as an independent variable. The variable of inter- est ln(C ) shows the logarithm of the average commuting time in the region of i,t−1 residence before the relocation t − 1. The average commuting time is calculated for each NUTS-3 region and represents the context of previously observed commutes. Further, I include X as a vector of control variables. This vector includes the i,t=0 log wage, calendar years, occupational status and indicator variables for firm size (number of employees, 4 categories), age group (4 categories), occupation (12 cat- egories), industry (9 categories) and region type of the place of residence as well as of the place of work (according to the classification of the Federal Institute for Research on Building, Urban Affairs and Spatial Development BBSR). These region types represent whether individuals live and work in a metropolitan city, city, large Estimating the model for the group of movers who only change their place of residence also reveals an effect of the context. The results can be provided additionally on request. However, investigating movers between German labor market regions generates almost the same results (see Web Appendix G). In this sample, I include all workers who relocate between two NUTS-3 regions. For all workers I have 5 observations, two observations before the move, the period of the relocation, and two after. 1 3 Persistence of commuting habits: context effects in Germany town, small town or in a rural area (5 categories). Moreover, X incorporates sev- i,t=0 eral dummies indicating whether a worker is a supervisor, has a leading position, is a trained/professional, specialist/expert or has an auxiliary job. In addition, X i,t=0 incorporates a dummy for women, migrants, western Germany and for being low- skilled (without vocational training) medium-skilled (with vocational training) or high-skilled (academic degree). And  shows the time invariant individual-specific effects. According to prediction 1,  should be positive because individuals with stronger observed commuting backgrounds have a lower disutility of commuting and thus prefer to live outside the city center, thereby facing longer commutes. However, in the case of unobserved heterogeneity, omitted variable bias and selectivity which can influence the estimates of C or sorting—meaning that mov- i,t−1 ers relocate to certain regions because of their taste for commuting—my results would not be valid. First, to address the issue of unobserved heterogeneity regard- ing, for example, commuting preferences, the estimates control for individual fixed effects  (Eq.  2). Thus, unobserved heterogeneity regarding individual commuting should not impact my results. Second, to deal with the issue of omitted variable bias, I conduct several robust- ness checks excluding observable individual and firm characteristics in my analysis. The results are presented in the robustness checks in Sect.  5 (Table  8) and confirm my presented results, as the results barely change. Third, workers might endogenously choose whether or not to move. To control for this selectivity, I use a two-stage Heckman selection method (Heckman 1979) where I first account for the decision to move, which can be estimated as a latent variable model: P =  S + (3) 1 i i With the decision to move: 1 if P > 0 P = (4) 0 otherwise P represents the latent variable for the propensity to move between two NUTS-3 regions and S is a vector of sociodemographic characteristics and information on industry and firm size, which influence individual i. To estimate whether or not a worker moves, I use a probit estimation. These results are then taken to construct an inverse Mills ratio. This inverse Mills ratio is then included in the second step equa- tion to correct for selection bias (Eq. 2). The third issue is sorting: For example, individuals who dislike (like) commut- ing choose regions with shorter (longer) commuting times. To face this selectivity issue, I include the individual’s own commuting time in the region before the reloca- tion C (see Eq. 2), and perform a robustness check. In line with selectivity, indi- i,t−1 viduals select themselves into a region because of their commuting taste. If people select themselves into regions with longer average commutes because of their taste for long commuting, they should also have commuted longer in the region before the move. To exploit this fact, I perform a reversed regression in which I regress the 1 3 R. Jost individual commute in the previous region on the average commuting time in the target region—after the relocation. ln(C )=  +  ln(C )+  X + (5) i,t−1 0 1 i,t=0 2 i i,t=0 In line with the above argument indicating selectivity, I should find a positive effect of the average commutes in the destination region C on the individuals’ i,t=0 commuting time in the region before the movement C . The results are presented i,t−1 in the robustness checks in Sect. 5 (Table 9). Another neglected effect could be due to imperfect information: when moving to a new region worker have no information about the commuting situation there. Therefore, they might commute longer initially and then change their commutes by relocating again within the new region—thereby explaining the second prediction. However, information about commuting and the local housing market is relatively cheap. Nevertheless, the commuting costs are high: commuting takes time, causes stress, and is very expensive. I would thus expect workers to obtain information about the commuting situation in the new region before they move. In addition, the decision regarding accommodation might be made under time pressure, thus representing a random event. For example, when individuals have found a new job but then have little time left to find a new apartment. In this case, they might be willing to take any accommodation, wherever it is located, as long as it seems to be acceptable. However, if it appears to be the case that the new commut- ing time is a random event, first I would not expect the individual’s own previous commuting time as well as the average commuting time in the region before the move to have a significant influence on the selected commuting time in the target region. And second, I would not expect those workers to move again within the new region and adjust their commuting time to the average commuting time in the new region. The travel time budget—and thus the commuting decision—might also be influ- enced by trip chaining or by the fraction of remote work. In particular, with the Covid-19-shock remote work has increased and there is some consistency in remote work. Due to the possibility of working from home the travel-time budget becomes more relaxed and thus longer commuting distances might be expected and accepted. However, as my observation period is restricted (2000–2014) and the data does not include the fraction of remote work, I cannot analyze how the results might be affected by the Covid-19-shock. In addition, Brunow and Gründer (2013) found that the daily allocation of time in Germany is affected by trip chaining, such that unob- served factors may influence the time budget. In particular, after migration not just the trip “home-to-work” influences the persistence of habits but also other factors such as shop accessibility or child care institutions leading to a potential bias in esti- mates. However, I suspect that this bias is negligible in this study, because people living in the destination area still form the daily activity chains. To test prediction 2, I restrict the sample to workers who move again within the new region, one period after the first move t + 1. I use the following identification strategy, in which only changes are analyzed. Because of these differences, individ- ual fixed effects are canceled out: 1 3 Persistence of commuting habits: context effects in Germany ln(C − C )=  ln(C − C )+  ln(W − W )+ (6) i,t+1 i,t=0 1 i,t=0 i,t−1 2 i,t+1 i,t=0 i The dependent variable (C − C ) is the change in the individual chosen i,t+1 i,t=0 commuting time after the second and the first move within the new region. The con- trol variable is the change in wages (W − W ) between the second and the first i,t+1 i,t=0 move. And the key predictor is represented by the difference between the observed commuting time in the new region t = 0 and in the region before the move t − 1, cor- responding to (C − C ) . This classification of the reference point presupposes i,t=0 i,t−1 that the workers’ perceptions have fully adjusted after one period. However, this might still not be a correct estimate of the change in the commut- ing time as workers might endogenously choose whether to move a second time. Therefore, I again use a two-step Heckman selection method (see Eqs. 3, 4). If work- ers decide to move a second time within the new region, in line with prediction 2, the coefficient  (Eq.  6) should be positive: individuals moving from regions with observed long commutes to a new region (with shorter average commutes) commute too long at first. This leads to a change in the desired commuting durations. There- fore, if they move again within this new region, they reduce their commutes and adopt the commuting behavior prevalent in the new region. 5 Empirical analysis of the commuting behavior 5.1 Descriptive statistics Figure  1 presents the distribution of the average commuting times for the place of residence for each NUTS-3 region in Germany. Workers living in metropolitan cit- ies, like Munich, Berlin, Frankfurt or Bremen, have shorter average commuting times than those in the surrounding regions. Specifically, the average commuting time in metropolitan cities is 16.8 min, while workers in rural areas commute almost 20  min to work on average. This implies that workers who live in large cities are most likely to work there as well, while workers living in the suburbs travel from the surrounding regions into the city center to work. This may be because job oppor- tunities are better in the city center and housing costs are cheaper in the suburbs (Alonso 1964). 5.1.1 Comparison of movers and non‑movers To demonstrate how the characteristics of workers who relocate differ from those who do not, I compare the two groups. The results are represented in the Web Appendix A. They show that movers and non-movers differ especially in terms of their productivity-related characteristics: employees who relocate are more highly qualified (academic degree) than non-movers. Differences also become obvious with regard to industries, occupations, and age groups. While the share of mov- ers is much larger between 18 and 34, non-movers are mainly between 35 and 1 3 R. Jost Bremen Berlin Frankfurt Munich Notes: The map shows the mean commuting time in workers’ place of residence by NUTS-3 regions in manually chosen time categories. Source: Own calculation and presentation. Fig. 1 Regional distribution of commuting time in the year 2014 56  years old. Moreover, movers tend to drive an average of 1.2  min longer to Table 1 Summary statistics of the daily wage and commuting time Variable Mean Std. dev 25th perc 50th perc 75th perc Commuting time t = − 1 in minutes 18.8 16.6 6.9 13.9 25.1 Δ Commuting time t = 0 in minutes + 3.9 23.8 − 8.2 2.7 14.9 Wage t = − 1 (euros/day) 85.9 55.9 49.7 74.2 106.5 Δ Wage t = 0 (euros/day) + 12.8 41.0 −2 .7 8.5 27.1 N 15,671 Means, standard deviation, 25th, 50th, 75th percentiles of commuting time and the wage. Comparison of movers before and after the relocation work than non-movers. This comparison therefore shows considerable heteroge- neity between movers and non-movers. 1 3 Persistence of commuting habits: context effects in Germany Table 2 Summary statistics of Variable Occupation Industry Promotion changes in occupation, industry, and promotion Change as a 33.0 33.9 12.7 percentage N 5238 5372 2009 Percentage of workers who change occupation or industry, or are promoted after the relocation t = 0 5.1.2 Comparison of movers before and after the relocation In the following, I examine summary statistics of workers who move. Table  1 shows the difference between movers’ driving times and wages before t = − 1 and after the relocation t = 0. The average mover experiences an increase in wages (+ 12.8 euros per day), which supports the idea that workers are more likely to move if they can achieve a wage increase, as non-movers on the other hand only experience an average wage increase of about 3.4 euros per day between two periods. Not only wages rise due to the relocation, the commuting time does so too. On average, the commuting time among movers increases by 3.9 min. 5.1.3 Motivation of movers As already mentioned, when workers move to a new region, they achieve an increase in wages, which could be an important motivation to move. Further- more, Table 2 shows that 33 percent of workers change their occupation after the move. In addition, almost 34 percent of movers work in a different industry after relocating. Workers might therefore move in particular for job-related reasons. Simonsohn (2006) obtains a similar finding. He reports that more than 36 percent of indi- viduals in the US move for job-related reasons. Moreover, in many cases (12.7 percent) the move is associated with a promotion, for example from trained/pro- fessional assistant to specialist/expert (see Table 2). 5.1.4 Comparison of movers and second‑time movers In the following, I take a closer look at second-time movers. These are workers who relocate a second time within the new region. Table  3 compares these sec- ond-time movers with the share of regular movers (workers who move once) after the first and before the second move. Of the 15,671 movers in t = 0 4,267 relocate a second time in t = 1. Especially medium-skilled workers tend to move again within the new region. In addition, the shares of men, migrants, and workers in western Germany are higher for sec- ond-time movers, and they are younger on average (between 18 and 24 years old). 1 3 R. Jost Table 3 Summary statistics of Variable Movers (%) Second- main variables time mov- ers (%) Woman 50.6 47.6 Migrant 3.9 4.3 West Germany 86.4 89.0 Age groups  18–24 14.5 18.4  25–34 47.9 46.8  35–44 26.8 24.9  45–56 10.8 10.0 Skill level  Low-skilled 6.5 7.9  Medium-skilled 63.1 68.5  High-skilled 30.4 23.6 N 11,597 4267 Means of main variables. Comparison of movers and second-time movers after the first move t = 0 Table 4 Summary statistics of commuting time and wage Variable Mean Std. dev 25th perc 50th perc 75th perc N Movers Commuting time 18.7 15.5 7.7 14.6 25.0 11,597 (min) Second-time movers Commuting time 33.4 22.6 15.2 28.1 48.1 4,267 (min) Movers Wage (euros/day) 99.7 58.6 61.6 84.1 122.4 11,597 Second-time movers Wage (euros/day) 96.0 55.0 62.7 82.0 113.8 4,267 Means, standard deviation, 25th, 50th, 75th percentiles of driving time and the wage. Comparison of movers and second-time movers after the first move t = 0 Table  4 shows the difference between the daily wages and the commuting times of movers and second-time movers after the first relocation t = 0. Compared to movers, second-time movers have much longer commuting times after the first move in t = 0. Workers who move only once have a commuting time of 18.7 min in t = 0, while those who move a second time drive over 14 min longer to work after the first relocation. This results not only from the fact that second-time movers come from regions with longer commutes compared to movers, but also that they are more likely to move from rural regions with longer average commut- ing times. According to the background context effect, this leads to a higher toler - ance for commuting and thus to a longer chosen individual commuting time after the move. This could explain why especially these workers move again within the new region and reduce their commuting time by more than 13 min (see Table 5). 1 3 Persistence of commuting habits: context effects in Germany Table 5 Summary statistics of commuting time and wage Variable Mean Std. dev 25th perc 50th perc 75th perc Commuting time t = − 1 in minutes 19.9 17.1 7.6 15.1 26.4 Δ Commuting time t = 0 in minutes + 13.4 27.7 − 2.7 10.7 29.7 Δ Commuting time t = + 1 in minutes − 13.5 24.8 − 28.0 − 6.6 2.3 Wage t = − 1 (euros/day) 81.8 51.7 47.4 72.1 101.7 Δ Wage t = 0 (euros/day) + 14.3 36.4 − 0.6 9.2 27.5 Δ Wage t = + 1 (euros/day) + 4.9 29.9 0.2 3.6 9.09 N 4267 Means, standard deviation, 25th, 50th, 75th percentiles of commuting time and the wage. Comparison of second-time movers before and after the first move and after the second move Table 5 shows the difference between wages and commuting times before the first move t = − 1 after the first move t = 0 and after the second move t = + 1 for individu- als who moved a second time. As explained above, the increase in the commuting time after the first move is far higher for individuals moving twice than for those moving only once. Second-time movers increase their commuting time by over 13  min in t = 0. However, they shorten their commuting time by the same amount after the second relocation in t = + 1. This corrects the originally excessive commut- ing time, and confirms prediction 2. 5.2 Empirical analysis Prediction 1: the average commuting time in the region a person leaves has a positive influence on the individually selected commuting time in the destina- tion region In the following, I test the first prediction, in which I investigate how the aver - age commuting time in the region before the relocation influences the individually selected commuting time in the target region (Eq. 2). As workers may endogenously choose to move, I use a two-step regression (Heckman 1979). In the first step I esti- mate a probit regression for the decision to relocate (Eq.  3). The results for this probit regression are provided in the Web Appendix B and show, for example, that workers with higher wages, high-skilled workers and workers in western Germany are more likely to relocate. In the second step, I use the inverse Mill’s ratio from the first step as an additional control variable and analyze how the average commuting time in the region before the relocation influences the commuting time in the new region (Eq. 2). Table 6 shows the results of 4 specifications. According to model 1, which includes the lag of the individual commuting time t  −  1, the longer the commuting time was in the region before the relocation, the longer the individually selected commuting time is in the target region. In addition, the wage has a positive significant effect, which might be the result of compensatory wages for longer commutes as shown by Mulalic et al. (2014). In the second model I include the average commuting time in the region in which the previous place of 1 3 R. Jost Table 6 Individually selected commuting time after relocation NUTS-3 region Dependent variable: ln(C ) i,t=0 Model 1 Model 2 Model 3 Model 4 Ln(C ) 0.228*** 0.225*** 0.225*** 0.225*** i,t−1 (0.006) (0.006) (0.006) (0.006) 0.216*** 0.222*** 0.212*** Ln(C ) i,t=−1 (0.030) (0.029) (0.029) Inverse of Mill’s ratio* 0.620*** 0.560*** 0.143 0.531*** (0.205) (0.206) (0.144) (0.206) Ln(wage) 0.107*** 0.100*** 0.117*** (0.034) (0.034) (0.034) Ln(wage ) − 0.103*** t−1 (0.015) Medium-skilled 0.172*** 0.163*** 0.103** 0.165*** (0.047) (0.047) (0.042) (0.047) High-skilled 0.231*** 0.211*** 0.093 0.206*** (0.079) (0.079) (0.066) (0.079) Migrant − 0.106 − 0.098 − 0.044 − 0.092 (0.066) (0.066) (0.063) (0.066) Specialist/expert 0.036 0.035 0.032 0.037 (0.044) (0.044) (0.044) (0.044) Trained/professional assistant 0.003 0.002 0.004 0.005 (0.038) (0.038) (0.038) (0.038) Age groups Yes Yes Yes Yes Occupation dummies Yes Yes Yes Yes Industry dummies Yes Yes Yes Yes Occupational status Yes Yes Yes Yes Firm size (Number of workers) Yes Yes Yes Yes Year dummies Yes Yes Yes Yes Place of residence type Yes Yes Yes Yes Place ofwork type Yes Yes Yes Yes Constant − 0.667 − 1.030 0.841* − 0.560 (0.830) (0.827) (0.508) (0.830) N 45,232 45,232 45,232 45,232 N (cluster) 15,262 15,262 15,262 15,262 R 0.5773 0.5777 0.5775 0.5783 Adj. R 0.3607 0.3614 0.3611 0.3622 The table reports regressions of the individually selected log commuting times after the first relocation on the average log commuting time in the region before the relocation and control variables. Standard errors clustered by individuals, below parameter estimates. Levels of significance: *1%, **5%, ***10% *Inverse of Mill’s ratio is obtained from the first stage probit estimation of the move residence was located C as a proxy for commuting options observed in the past. i,t−1 Consistent with the first prediction, model 2 shows a positive significant effect on the individual commuting time. Moreover, the effect can be interpreted as causal, as I control for selectivity and unobserved heterogeneity, and can rule out the issue 1 3 Persistence of commuting habits: context effects in Germany Table 7 Adjustment of the NUTS-3 region Depend- commuting time in t + 1 ent variable: ln(C − C ) i,t+1 i,t=0 0.100* Ln(C − C ) i,t=0 i,t−1 (0.046) Change in ln(wage) 0.049 (0.108) Inverse of Mill’s ratio* 1.971*** (0.084) Constant − 2.729*** (0.084) N 4,135 0.3531 Adj. R 0.3526 The table reports the regression of the adjustment of the individually selected commuting time after the second move on the difference between the average commutes in the new and the old region. Stand- ard errors clustered by NUTS-3 regions, below parameter estimates. Levels of significance: *1%, **5%, ***10% *Inverse of Mill’s ratio is obtained from the first stage probit estima- tion of moving again within the new region of omitted variable bias and sorting (see Sect.  5.3). Hence, mobile workers com- ing from NUTS-3 regions with longer observed commutes have a greater tolerance for commuting and choose longer individual commutes in the target region. This indicates the presence of a context effect and is therefore consistent with the result obtained by Simonsohn (2006). However, a comparison of the effects with those found by Simonsohn (2006) shows that he overestimates the effect of the context (see Sect.  5.3 Table  8). This is because he does not include individual unobserved fixed effects. In addition, comparing R reveals that the model I consider performs much better than that of Simonsohn (2006) (0.36 vs. 0.15). Since commuting may be endogenous with respect to wages, model 4 excludes daily wages, which has little impact on the size of the coefficient of C . In addi- i,t−1 tion, in model 5 I include time-lagged wages t − 1. In this estimation, too, the result shows no change for the variable of interest C . i,t−1 Thus, the results indicate that workers’ current commuting behavior is affected not only by their own previous commuting time but also by the average commuting time in the region they moved from. Prediction 2: Individuals readjust their commuting times and move again when remaining in the new region If workers relocate from regions with longer commutes to regions with shorter average commuting times ( C > C ), they initially commute longer than the i,t−1 i,t=0 average in the target region. The reason for this is that they have a greater toler- ance for commuting as they come from regions where long commutes are common. Nevertheless, if they remain in the new region and observe fewer commutes, they 1 3 R. Jost Table 8 Robustness check: individually selected commuting time after the move NUTS-3 region Dependent variable: ln(C ) i,t=0 Model 1 Model 2 Model 3 Model 4 Model 5 Ln(C ) 0.531*** 0.226*** 0.226*** 0.227*** 0.225*** i,t−1 (0.004) (0.006) (0.006) (0.006) (0.006) 0.154*** 0.224*** 0.220*** 0.223*** 0.224*** Ln(C ) i,t=−1 (0.023) (0.029) (0.029) (0.029) (0.029) Inverse of Mill’s ratio* 0.092 − 0.009 0.292** − 0.021 (0.078) (0.032) (0.132) (0.032) Ln(wage) 0.061*** 0.088*** 0.034 (0.013) (0.027) (0.024) Medium-skilled 0.039** 0.119*** 0.078** (0.018) (0.040) (0.035) High-skilled 0.035 0.124** 0.040 (0.028) (0.062) (0.048) Migrant − 0.037* − 0.063 − 0.027 (0.019) (0.063) (0.060) Specialist/expert 0.026 0.016 0.027 (0.024) (0.044) (0.044) Trained/professional assistant 0.001 − 0.012 0.001 (0.021) (0.038) (0.038) Age groups Yes Yes Yes Occupation dummies Yes Yes Yes Industry dummies Yes Yes Yes Occupational status Yes Yes Yes Firm size (Number of workers) Yes Yes Yes Year dummies Yes Yes Yes Yes Yes Place of residence type Yes Yes Yes Yes Yes Place of work type Yes Yes Yes Yes Yes Constant 0.190 1.438*** − 0.042 1.476*** 1.194*** (0.310) (0.134) (0.533) (0.132) (0.139) N 45,232 45,232 45,232 45,232 45,232 N (cluster) 15,262 15,262 15,262 15,262 15,262 R 0.3415 0.5768 0.5763 0.5753 0.5776 Adj. R 0.3407 0.3606 0.3595 0.3586 0.3612 The table reports regressions of the individually selected log commuting times after the first relocation on the average log commuting time in the region before the move and control variables. Standard errors clustered by individuals, below parameter estimates. Levels of significance: *1%, **5%, ***10% *Inverse of Mill’s ratio is obtained from the first stage probit estimation of the move become dissatisfied with their initially chosen commutes and their desired commut- ing time changes. Therefore, I expect them to reduce their commutes by relocating again within the new region. To analyze the adjustment of the commuting time after a second move, I consider only individuals who move again within one year after 1 3 Persistence of commuting habits: context effects in Germany relocating to the new region. A total of 4,135 individuals move again within the new NUTS-3 region in t = 1. The regression estimates of Eq. 6 are presented in Table 7, where (C − C ) , i,t+1 i,t=0 the dependent variable, measures the change in the individual commuting time after the second and the first relocation. Therefore, it represents the adjustment of the individual commuting time between t = 0 and t = + 1. The key predictor is the differ - ence between the average commuting time in the new region and that in the previ- ous region (C − C ) . Moreover, as workers may endogenously choose whether i,t=0 i,t−1 to move a second time, I use a two-step regression (Heckman 1979): in the first step, I estimate a probit regression for the decision to relocate a second time in the new region (Eq.  3). The results of this probit regression can be found in the Web Appendix C. They show, for example, that the greater the difference between the average commuting time and the individual’s own selected commuting time in the target region, the more likely a second move is. In the second step, I use the inverse Mill’s ratio from the first step as an additional control variable. The results are pre- sented in Table 7 and are seen to be in line with prediction 2, the greater the differ - ence between the new and the old region (C − C ) the stronger the adjustment i,t=0 i,t−1 of the individually chosen commuting time after the second move is. Comparing the estimated effect of  (Table 7) with the estimation of  in prediction 1 (Table 6 2 2 model 2) it can be seen that the coefficient  of the first prediction is twice as large as  in the second prediction. Thus, second-time movers do not fully reverse the original impact of C , but it is moving in that direction. i,t−1 With this result, I can therefore rule out an explanation for the commuting behav- ior that is based on stable unobserved differences across movers from different regions, as individuals readjust their commuting time by moving again within the new region—they adopt the commuting behavior of the new region. 5.3 Robustness checks Although the presence of stable unobserved differences can be ruled out by con- firming prediction 2, there could be other explanations for the presented results and several issues that might influence the outcome, such as unobserved heterogeneity, omitted variable bias, selectivity, and sorting. However, in the following, I am able not only to reject other explanations, but also to confirm my results by means of several robustness checks. Therefore, the effect of C on C can be interpreted i,t−1 i,t=0 as causal. 5.3.1 Unobserved heterogeneity In fact, unobserved heterogeneity can have an influence on the estimates of C , i,t−1 thereby driving the effect of the context (see Sect.  4). To deal with this issue, I include individual fixed effects in my analysis (see Eq.  2). This is especially impor- tant, and failure to do so generates a bias. This can be observed in Table 8 (model 1). Excluding individual fixed effects overestimates the effect of the individual previous commuting time C , and underestimates the influence of the context of previously i,t−1 1 3 R. Jost observed commutes C . It is therefore important to include individual fixed i,t−1 effects. Failure to do so leads to a bias, as in the study by Simonsohn (2006) which does not include individual fixed effects in the analysis and therefore underestimates the effect of the context. 5.3.2 Omitted variable bias In addition, I conduct several robustness checks excluding individual and firm char - acteristics. In model 2 (Table 8) I exclude firm characteristics, which yields similar results for the context of previously observed commutes to those in Table 6 (model 2), which included all control variables. Also, almost the same results are obtained when firm characteristics are excluded and when both individual and firm character - istics are excluded (models 3 and 4). Thus, the results on the previous average com- muting time are very robust and do not seem to be influenced by observed individual or firm characteristics. This leads me to conclude that there is no evidence of omit- ted variable bias. 5.3.3 Selectivity To control for the selectivity of a relocation—as workers may endogenously choose to relocate—I use a two-step Heckman selection model (Heckman 1979), in which I control for the selectivity of a relocation (Eq. 3). To gain an impression of whether selectivity is important I estimate the model without controlling for selectivity. The results are provided in Table 8 (model 5) and show almost the same effects for pre- viously observed commutes as those in Table 6 (model 2). Only the coefficients for wages and the skill-level variables change. Thus, controlling for the selectivity of the relocation is not important for interpreting the variable of interest but influences other control variables. 5.3.4 Sorting Another issue might be sorting, as workers select themselves into certain regions because of their taste for commuting. To address this issue, I run a reversed regres- sion of Eq. 5. In line with the definition of sorting, I should find a positive correla- tion between the average commuting time in the destination region and the indi- vidual commuting time in the region before the move. However, my results show no significant effect of the average commuting time in the destination regions (Table  9). Thus, there is no sign of a sorting process—individuals do not select themselves into regions because of their taste for commuting—but this once again shows the presence of the context effect. Moreover, workers might also move for job-related reasons, such as higher wages. As wages are highly correlated with commuting in theory, I consider only work- ers who earn almost the same wage before and after the first relocation. Table  10 Wages are rounded to the nearest ten. 1 3 Persistence of commuting habits: context effects in Germany Table 9 Robustness check: NUTS-3 region Depend- individuals select themselves ent variable into regions because of their ln(C ) i,t−1 taste for commuting Ln(C ) 0.082*** i,t=0 (0.008) − 0.109 Ln(C ) i,t=0 (0.077) 0.951*** Ln(C ) i,t−1 (0.059) Ln(wage) 0.086*** (0.021) Medium-skilled 0.049* (0.029) High-skilled 0.078** (0.034) Migrant − 0.109** (0.043) Specialist/expert 0.053 (0.047) Trained/professional assistant − 0.005 (0.042) Woman − 0.070*** (0.019) Age groups Yes Occupation dummies Yes Industry dummies Yes Occupational status Yes Firm size (Number of workers) Yes Year dummies Yes Place of residence type Yes Place of work type Yes Constant − 0.658** (0.297) N 15,262 R 0.056 0.0520 Adj.R The table reports the regression of the individual commuting time in the previous region on the average commuting time in the target region (after the relocation). Standard errors clustered by NUTS-3 regions, below parameter estimates. Levels of significance: *1%, **5%, ***10% shows that the average commuting time in the region before the move has a positive and significant influence on the commuting time of workers who do not achieve an increase in wages after the relocation. This indicates that endogeneity issues with respect to wages do not drive the results. In addition to restricting the sample to per- sons earning the same wage before and after relocating, I also restrict it to workers 1 3 R. Jost Table 10 Robustness check: NUTS-3 region (1) Depend- (2) Depend- movers, who earn almost the ent variable ent variable same wage before and after ln(C ) ln(C ) i,t=0 i,t=0 relocating (1) and who have the same wage as well as the same Ln(C ) 0.202*** 0.199*** i,t−1 task level (2) before and after (0.013) (0.015) relocating 0.381*** 0.361*** Ln(C ) i,t=−1 (0.071) (0.081) Ln(wage) − 0.164* − 0.125 (0.088) (0.109) Medium-skilled − 0.089 − 0.073 (0.102) (0.117) High-skilled − 0.033 − 0.042 (0.151) (0.196) Migrant 0.161 0.060 (0.108) (0.138) Specialist/expert 0.066 0.014 (0.103) (0.197) Trained/professional assistant 0.063 0.018 (0.081) (0.155) Age groups Yes Yes Occupation dummies Yes Yes Industry dummies Yes Yes Occupational status Yes Yes Firm size (Number of workers) Yes Yes Year dummies Yes Yes Place of residence type Yes Yes Place of work type Yes Yes Constant 1.745*** 1.687*** (0.449) (0.559) N 9,193 7,473 N (cluster) 3,094 2,514 R 0.5797 0.5833 Adj. R 0.3603 0.3645 The table reports regressions of the individually selected log com- muting time after the first relocation on the average log commuting time in the region before the move and control variables. Standard errors clustered by individuals, below parameter estimates. Levels of significance: *1%, **5%, ***10% who do not change their task level. Once again, the coefficient of the average com- muting time in the previous region does not change. To sum up, the robustness checks show that it is crucial to include the individual fixed effects when investigating the individual commuting behaviors. In addition, the robustness checks indicate that my results on the average commuting time are not driven by omitted variable bias—as the coefficient is very robust when indi- vidual and firm-specific characteristics are excluded. Furthermore, sorting does not 1 3 Persistence of commuting habits: context effects in Germany seem to influence my results, either. Therefore, the investigated influence of the pre- viously observed commutes on the individually chosen commuting time (Table  6) can be interpreted as causal. 5.4 Eecfft heterogeneity In the following, I investigate the heterogeneous effects of the context on the indi- vidual selected commuting time. I differentiate movers by different age groups, skill levels, and gender. In addition, I consider movers between different types of regions—urban and rural areas—as well as movers between labor market regions. 5.4.1 Age groups, gender, and skill level Since it is possible that individuals differ in their behavior due to their age, gender, or skill level, I take up this point by performing the estimation for different interac- tions (Web Appendix D). In particular, I interact the average commuting time in the region before the relocation C with age, gender, and skill level. The results i,t=−1 show no significant group differences in terms of age and skill level. Nor can any significant differences be observed between women and men. Thus, there is no effect heterogeneity for different groups regarding age categories, skill level, or gender. 5.4.2 Movers between different types of rural and urban regions Considering movers between different types of place of residence, I interact the average commuting time in the previous location C with the different types of i,t=−1 rural and urban regions before and after the move. The results are shown in the Web Appendix E and indicate that the effect of the context of previously observed commutes is strongest for those moving to urban areas, especially for the group moving from a rural to an urban area. This is related to the fact that workers who previously lived in a rural area with long average commutes are used to commuting long distances. Therefore, when moving to urban regions such workers have a higher tolerance for commuting and choose longer than average commutes in the urban region. However, for movers to rural areas the results indicate a smaller or insignifi- cant effect of the context. The reason could be that the majority of workers moving from urban to rural areas do not only relocate their place of residence but also take up a new job in the rural area. Thus, other conditions, such as job availability, are more important than commuting preferences for this group of movers. Hence, the results indicate that the size of the effect of the context depends par - ticularly on the region type of the place of residence before and after the reloca- tion. Considering only movers between metropolitan areas (Simonsohn 2006) might therefore lead to a bias in the estimated effect. I consider only the location of the place of residence. However, as I cannot take into account dense traffic or congestion when calculating the commuting time, the results might underestimate the commuting costs especially in dense urban areas. 1 3 R. Jost 5.4.3 Labor market regions Next, I show the results for individuals moving between German labor market regions (Kosfeld and Werner 2012). The restrictions are the same as for movers between NUTS-3 regions, i.e., workers have to relocate both their place of work and their place of residence to a different German labor market region. Moreover, the labor market region of the place of work and the place of residence must be constant for two years before and after the move. In contrast to the consideration of indi- viduals moving between NUTS-3 regions, I calculate the average commuting time at the level of labor market regions (as a proxy for previously observed commuting options). The results are shown in Web Appendix F and are comparable with the effect of the context for persons moving between NUTS-3 regions (Table 6). 6 Conclusion This study investigates for the first time commuting behavior in terms of a behav - ioral economic concept based on geo-referenced data for Germany. The basis of this investigation is the approach developed by Simonsohn (2006), who examines commuting behavior for the US. However, I can show that his estimated effects are biased due to the absence of individual fixed effects and the consideration only of individuals moving between metropolitan areas. The presented results show that workers’ commuting decisions are influenced by commuting options observed in the past. This explains why individuals who move from different regions to one and the same region initially commute differently: indi- viduals moving from areas with long average commutes have a greater tolerance for commuting and therefore commute more than individuals coming from regions with shorter commutes. However, if they remain in the new region, they adjust their initially chosen commuting times to the average commutes in the new region. This refutes the assumption of stable unobserved differences across individuals. Instead, individuals change their marginal utility of commuting when moving to a new region, as they adjust their commuting time by means of a second relocation within the new region. The reason for this behavior is the change in the context: the origi- nal context was seen as the average commuting time in the previous region, but the context changes with the relocation to a new region. Thus, commuting preferences change. In addition, the results indicate that selectivity and sorting do not influence the effect of the context, but it is crucial to include individual fixed effects. Moreo- ver, the context has different effects depending on the region type of the place of residence: the context effect is greatest for those moving from rural to urban areas. However, the travel time budget can be influenced by remote work that increased during the Covid-19-shock and might increase the expected and acceptable commut- ing distance. Future research could examine whether such increase in remote work influences the effect of the context. Additionally, for future investigation that exam- ine consumer preferences and other labor market decisions, the study highlights the importance of identifying the context of previously observed options and including 1 3 Persistence of commuting habits: context effects in Germany them in the analysis. Finally, the results indicate the essentiality of including indi- vidual fixed effects, as they influence the outcome of commuting decisions. Supplementary Information The online version contains supplementary material available at https:// doi. org/ 10. 1007/ s00168- 023- 01223-4. Acknowledgements I thank Peter Haller, Wolfgang Dauth, Joachim Möller, the ERSA Congress 2019 and the BGPE Workshop in Regensburg (2019) for many helpful comments on this project. Further I appreciate very detailed and helpful comments by Janet Kohlhase and two anonymous reviewers. Funding Open Access funding enabled and organized by Projekt DEAL. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Com- mons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. References Alonso W (1964) Location and land use. Toward a general theory of land rent. Harvard University Press, Cambridge Bettman J, Luce FL, Payne J (1998) Constructive consumer choice processes. J Consumer Res 25:187–217 Bhargava S, Fisman R (2014) Contrast effects in sequential decisions: evidence from speed dating. Rev Econ Stat 96:444–457 Brueckner JK (2000) Urban sprawl: diagnosis and remedies. Int Reg Sci Rev 23:160–171. https:// doi. org/ 10. 1177/ 01600 17007 61012 710 Brueckner JK, Stastna L (2020) Commuting and migration. J Reg Sci 60:853–877. https:// doi. org/ 10. 1111/ jors. 12498 Brunow S, Gründer M (2013) The impact of activity chaining on the duration of daily activities. Trans- portation 40:981–1001 Card D, Heining J, Kline P (2013) Workplace heterogeneity and the rise of West German wage inequality. Q J Econ 128:967–1015 Dargay JM, Clark S (2012) The determinants of long distance travel in Great Britain. Transp Res Part A Policy Pract 46:576–587. https:// doi. org/ 10. 1016/j. tra. 2011. 11. 016 Dauth W, Haller P (2020) Is there loss aversion in the trade-off between wages and commuting distances? Reg Sci Urban Econ. https:// doi. org/ 10. 1016/j. regsc iurbe co. 2020. 103527 Destatis (2017) Commuting in germany: 68% use a car to commute to work. https:// www. desta tis. de/ DE/ Zahle nFakt en/ ImFok us/ Arbei tsmar kt/ Pendl erArb eitsw eg. html Duan Y, Jost O, Jost R (2022) Beyond lost earnings: the long-term impact of job displacement on work- ers’ commuting behavior. IAB-Discussion Paper, 15/2022 Frey B, Stutzer A (2007) Commuting and life satisfaction in Germany. Informationen zur Raumentwick- lung. Heft 2/3 Gimenez-Nadal JI, Molina JA, Velilla J (2018) Spatial distribution of US employment in an urban effi- ciency wage setting. J Reg Sci 58:141–158. https:// doi. org/ 10. 1111/ jors. 12351 Gimenez-Nadal JI, Molina JA, Velilla J (2020) Trends in commuting time of European workers: a cross- country analysis. IZA Discussion Paper No 12916 Hanson S, Johnston I (1985) Gender differences in work trip lengths: Implications and explanations. Urban Geogr 6:193–219. https:// doi. org/ 10. 2747/ 0272- 3638.6. 3. 193 1 3 R. Jost Hartzmark S, Shue K (2017) A tough act to follow: contrast effects in financial markets. NBER Working Paper No. 23883 Heckman J (1979) Sample selection bias as a specification error. Econometrica 47:153–161. https:// doi. org/ 10. 2307/ 19123 52 Heuermann D, Assmann F, Freund F, vom Berge P (2016) The distributional effect of commuting subsi- dies-evidence from geo-referenced data and large-scale policy reform. Reg Sci Urban Econ 67:11– 24. https:// doi. org/ 10. 1016/j. regsc iurbe co. 2017. 08. 001 Huber S, Rust C (2016) osrmtime: calculate travel time and distance with OpenStreetMap data using the open source routing machine (OSRM). Stand Genomic Sci 16:416–423. https:// doi. org/ 10. 1177/ 15368 67X16 01600 209 Huber J, John P, Puto C (1982) Adding asymmetrically dominated alternatives: violations of regularity and the similarity hypothesis. J Consumer Res 9:90–98. https:// doi. org/ 10. 1086/ 208899 Kahneman D, Tveresky A (1979) Prospect theory: an analysis of decision under risk. Econometrica 4:263–287 Kosfeld R, Werner A (2012) German labour markets—new delineation after the reforms of Ger- man district boundaries 2007–2011. Raumforsch Raumordn 70:49–64. https:// doi. org/ 10. 1007/ s13147- 011- 0137-8 Lux M, Sunega O (2012) Labor mobility and housing: the impact of housing tenure and housing afford- ability on labor migration in the Czech Republic. Urban Stud 4:489–504. https:// doi. org/ 10. 1177/ 00420 98011 405693 McQuaid R, Chen T (2012) Commuting times—the role of gender, children and part-time work. Transp Econ 34:66–73. https:// doi. org/ 10. 1016/j. retrec. 2011. 12. 001 Mulalic I, van Ommeren J, Pilegaard N (2014) Wages and commuting: quasi-natural experiments´ evi- dence from firms that relocate. Econ J 124:1086–1105. https:// doi. org/ 10. 1111/ ecoj. 12074 Ostermann K, Eppelsheimer J, Gläser N, Haller P, Oertel M (2022) Geodata in labor market research: trends, potentials and perspective. J Labor Mark Res 56:1–17 Ross SL, Zenou Y (2008) Are shirking and leisure substitutable? An empirical test of efficiency wages based on urban economic theory. Reg Sci Urban Econ 38:498–517 Rouwendal J, Rietveld P (1994) Changes in commuting distances of Dutch households. Urban Studies 31:1545–1557 Ryder H, Heal G (1973) Optimal growth with intertemporally dependent preference. Rev Econ Stud 40:1–33 Shuai X (2012) Does commuting lead to migration? J Reg Anal Policy 42:237–250 Simonsohn U (2006) New Yorkers commute more everywhere: contrast effects in the field. Rev Econ Stat 88:1–9. https:// doi. org/ 10. 1162/ rest. 2006. 88.1.1 Simonsohn U, Loewenstein G (2006) Mistake #37: the effect of previously encountered prices on current housing demand*. Econ J 116:175–199. https:// doi. org/ 10. 1111/j. 1468- 0297. 2006. 01052.x Simonson I, Tveresky A (1992) Choice in context: tradeoff contrast and extremeness aversion. J Mark Res 29:281–295 Tveresky A, Kahneman D (1991) Loss aversion in riskless choice: a reference-dependent model. Q J Econ 106:1039–1061 Van Ham M, Hooimeijer P (2009) Regional differences in spatial flexibility: long commutes and job related migration intentions in the Netherlands. Appl Spat Anal Policy 2:129–146. https:// doi. org/ 10. 1007/ s12061- 008- 9016-2 Van Ommeren J (2005) Commuting: the contribution of search theory. Emerald Group Publishing Lim- ited, Bingley, pp 347–380. https:// doi. org/ 10. 1108/ S0573- 8555(2005) 00002 66013 Van Ommeren J, Fosgerau M (2009) Workers’ marginal costs of commuting. J Urban Econ 65:38–47. https:// doi. org/ 10. 1016/j. jue. 2008. 08. 001 Van Ommeren J, van den Berg G, Gorter C (2000) Estimating the marginal willingness to pay for com- muting. J Reg Sci 40:541–563. https:// doi. org/ 10. 1111/ 0022- 4146. 00187 Zabel J (2012) Migration, housing market, and labor market responses to employment shocks. J Urban Econ 72:267–284. https:// doi. org/ 10. 1016/j. jue. 2012. 05. 006 Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. 1 3

Journal

The Annals of Regional Science – Springer Journals

Published: Mar 1, 2024

Keywords: J60; R10; R19; R23

There are no references for this article.