Women’s Issues in Transportation: Summary of the 4th International Conference, Volume 2: Technical Papers (2011)

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

191 Travel Time and Distance Regarding Gender Patterns in the Paris Region Past Trends and Forecasts to 2030 Ariane Dupont and Zoran Krakutovski, National Research Institute on Transportation and Its Security, Department of Economics and Sociology of Transportation (INRETS-DEST) This paper investigates the specificities and the evolu- tion of mobility patterns of women and aging people in the most densely populated region including Paris and the departments, which has the highest gross domestic product per capita in France. Four house- hold travel surveys available from 1976 to 2001 allow a comparison to be made of travel time during those years and also a forecast of mobility up to 2030. The chosen focus is on the distinction between constrained trips and nonconstrained trips, constrained trips being defined as trips whose destination is work, university, or school. It is observed that the constrained trips have not increased recently in regard to time and distance, after a surge in the 1980s, except for women, whose travel time and distance budgets strikingly increased. In the same period, overall mobility increased dramati- cally, especially as a result of nonconstrained trips for both men and women. It appears, however, that non- constrained and constrained trips for women increased more significantly. This paper investigates the travel behavior of women and aging people in the most densely populated and wealthiest region of France [gross domestic product (gDP) per capita], which includes Paris and the departments. This empirical case study aims to identify the specificities of mobility patterns and their evolution over time (measured by the average number of trips per individual per working day and distance and time budgets1) related to gender and age. Four household travel surveys available for 1976, 1983, 1991, and 2001 allow one to compare travel time for these years and also to forecast mobility up to 2030. This analysis is driven by the investigation of the assumption of recent converging patterns in a highly dense urban area. The focus here is on the distinction between con- strained trips and nonconstrained trips, constrained trips being defined as trips whose destination is work, university, or school. A study has been done of the mobility patterns related to demographic patterns but also to socioeconomic and geographic patterns to identify some key elements determining travel char- acteristics and choice differences between men and women and the evolution of these patterns during a long period. In a standard approach, time budgets and distance budgets were then computed according to the purpose of the trip. Paris results confirmed that the Zahavi law (Zahavi and Ryan 1980) on the constancy of transport time among time and place has to be contested, par- ticularly when density and urban patterns are taken into account, as shown especially for France by Wiel 1 Travel time budget computes a daily time dedicated to transport on a weekday per individual (in minutes, min/individual/day), and travel distance budget computes a daily distance (as the crow flies) dedi- cated to transportation between one location of activity and another (km/individual/day).

192 WOMEN’S ISSuES IN TRANSPORTATION, VOLuME 2 (2002). These budgets must be put into perspective with transport mode and housing location. As a result, it is observed that the constrained trips for Franciliens2 have not increased recently in regard to average time and distance, after a surge in the 1980s, except for women, whose distance budget increased. In the same period, the overall mobility, measured by travel dis- tance budget, is increasing. It appears that this increase is the result of the development of nonconstrained trips and of constrained trips for women. Then, in regard to demographic patterns such as age and gender, the evolution is mixed when introducing gender and age patterns. Indeed, from 1976 in regard to distance, as constrained trips declined for men they strikingly increased for women, and in regard to time, the constrained trips are almost constant. The time dedicated for commuting is also changing according to age class, and huge differences appear between the 1970s and the 2000s. Aging people being far more mobile than their parents at the same age, and the most interesting point shown by census is that aging people, since the end of the 1990s and the beginning of the 2000s, are staying in the Paris region for their retirement period, which was not the case 30 years ago. Also being investigated are the discrepancies in trips and mode choice according to gender for aging people. It will then be seen that location matters. Because of the relocation of economic activities out- side Paris and because of poor accessibility to transit, women’s travel time and distance budgets have signifi- cantly increased in the past years and are forecast to continue to surge. That is why the analysis has to be disaggregated according to the seven subareas of the Paris region and also according to socioeconomic fac- tors. Are travel patterns the same for all socioeconomic categories of women, for all women over the area, for all age categories? The first part of the paper is dedicated to the pre- sentation of the survey that was used and the overall patterns of traveling for the Paris region between 1976 and 2001. In the second part, an analysis is carried out of women’s and aging people’s mobility patterns, with a special focus on constrained and nonconstrained trips. A disaggregated analysis in regard to location and socioeconomic factors helps to identify the deter- minants of these gender differences in travel time and distances but also in modal choices. In a third part, a forecast is developed for travel time, up to 2030, mainly on the basis of demographic patterns in the age-cohort modeling approach (Bush 2003, Berri 2005, gallez 1994). As a result, this forecast supports the investigation of the recent convergence of mobility patterns between men and women. 2 Name of the inhabitants of Ile-de-France. trAvel pAtternS of the pAriS region (ile-de-frAnce) Paris Region The Paris region contains the major labor pool of France. This region is set in a cluster of eight departments: Paris and the inner belt including three departments (Val-de- Marne, Hauts-de-Seine, and Seine-Saint-Denis) close to Paris and directly surrounding it and the outer belt including four departments (Seine-et-Marne, Essonne, Yvelines, Val-d’Oise) up to 100 km from Paris. The subordination of the outer and inner suburbs to Paris is still outstanding in regard to administrative and business activities (see Figure 1).3 In 2008, according to the Organisation for Eco- nomic Co-operation and Development Regional Statis- tical Database, the Paris region, with a population of 11,490,969 inhabitants, accounts for one-fifth of the French population. With a gDP of 552,664 million euros, it is the wealthiest part of France (gDP per capita: 47,155 euros). In 1996, the gross disposable income per inhabitant in Ile-de-France (the IDF) was 110,800 francs (16,891 euros) against 90,100 francs (13,735.65 euros) in the provinces. given in an index relative to all of France, the gDP in the IDF was 152.9 in 1996, and ranked first in the list of the 22 French regions—the rank being con- firmed in 2008. The Paris region’s gDP rose during the period of interest from 983 billion francs (150 billion euros) in 1982 up to 1,750 billion francs (267 billion euros) in 1989, reaching 2,289 billion francs (349 billion euros) in 1996. Between 1995 and 2005, even if other parts of France, especially in the south of the Loire Valley, were economi- cally dynamic and attractive, the Paris region remained a dynamic pole with a gDP yearly growth of 2.5% and the major labor pool, with an employment rate of 59.5% and an unemployment rate of 9.5%. Between 1995 and 2005, the annual growth rate was 2.5%, the activity rate was 59.5%, and the unemploy- ment rate was of 9.5%, below the national average. The job market has been oriented mainly toward ter- tiary activities since the 1990s but still remains the main pole of industrial production. Brière and guilbert (2001) showed that between 1976 and 2001 the working pop- ulation of the IDF was deeply restructured: a decrease in the activity rate of the population, a rise in women’s activity rate, an increasing share of white-collar workers and highly qualified workers in a context of the develop- 3 All results presented in the tables and figures of this paper are derived from the authors’ own computations (French National Institute for Transportation and Safety/Department of Economics and Sociology of Transportation).

193gENDER PATTERNS IN TRAVEL TIME AND DISTANCE IN THE PARIS REgION ment of the production of high value-added goods and services. Blanchard and Ferrer (2007) identified a Paris paradox: the working population of the IDF gathers very highly qualified and skilled individuals (5 to 6 years after graduation) as well as workers with no qualifications at all (who have not graduated). According to the French National Institute of Statis- tics and Economic Studies (INSEE) the share of working inhabitants (International Labor Organization defini- tion) decreased from 53% in 1976 to 49% in 2001 (with the longer time spent at university and a lower legal retirement age since the 1981 law). But for the period, the evolution for men and women was different, with a decrease in the share of working males (from 63% to 54%) while the share of the female working population remained almost constant (44% to 45%). The inhabitants of the Paris region are people mainly between the ages of 20 and 24 years and those up to the age of 55 to 59, because Paris is still viewed as a place dedicated to education and work: the young people move to be educated in Paris (most of the famous gradu- ate schools and universities are still located in the Paris region) and then move back to their native region (or the southern part of France) for retirement. During the period, equality between men and women in regard to access to the job market improved in the Paris region. The share of the nonworking population increased as a result of new regulations concerning the legal retire- ment age and special incentive frameworks for promoting retirement, and also a longer education period. The analysis focused on the residential location shows that the number of inhabitants in all departments of the Paris region increased from 1976 to 2001, except in Paris where the population decreased by 5%. The most impor- tant increase occurred in the Seine-et-Marne Depart- ment (+58% between 1976 and 2001) in the east of the Paris region and in the Val-d’Oise Department (+41% between 1976 and 2001) in the west/northwest of the Paris region. Even so, Paris remains the most populated place, with 1.97 million inhabitants, followed by the Hauts-de-Seine Department, close to the business area La Défense, with 1.31 million inhabitants. These latter departments are those with the less impor- tant nonworking population. The share of nonworking individuals has significantly and continuously increased in the departments of Seine-Saint-Denis and Val-d’Oise, to the north of Paris. The jobs are located mainly in the center and the west of the region, that is, in Paris and Hauts-de-Seine, and housing stock in the east. In recent years, firms have relo- cated their headquarters from inside Paris to the suburbs because of the high commercial rent levels in Paris. Because of the high population density of the Paris region and the cluster location of business and admin- istrative activities and housing, mobility is a key issue in this, the most important economic and labor area of France. Paris Mobility Survey A large survey is regularly conducted on the transportation behavior of households in the Paris region. It contains vari- ous survey instruments to describe the means of transport owned by the household, commuting habits, daily mobil- ity, and so on. Data collected in each of these instruments can be analyzed separately. The purpose of this survey is to describe the trips made by households and individuals who are living in Paris and the nearest departments, as well as their use of public and private transport means. The sur- vey is conducted for a working weekday. The focus is on internal trips, that is, trips whose origin and destination are within the Paris region (the IDF). Analysis of the trips and the mobility of inhabit- ants through the time and distance budgets of the most wealthy and densely populated region of France is based on the data collected by four household mobility surveys carried out in the Paris region in 1976–1977, 1983–1984, 1991–1992, and 2001–2002. For each of the four surveys, the sample size is about 10,000 responding households. For example, from 18,182 households representative of the IDF population drawn in 2001–2002, there are 10,478 responses (i.e., a response rate of 57%). The questionnaire contains 1. Household data (address, class of income, motor- ization, number of persons, and so forth), 2. Surveyed individuals in the household older than 5 years (work, age, place of work, and so forth), 3. Characteristics of trips (origin, destination, mode, time, purpose, and so forth). FIGURE 1 Map of Paris and its inner and outer suburbs.

194 WOMEN’S ISSuES IN TRANSPORTATION, VOLuME 2 The origin and destination of trips are geocoded, which allows the analysis of Parisian mobility in differ- ent zones. Travel Patterns over the Years The analysis favors the distinction between “constrained” and “nonconstrained” trips to identify travel behavior. The label is not coined in relation to the semantic mean- ing of “constrained” or “constraint” as relates to the feelings that an individual has in defining a trip as com- pulsory or not, but is based on the work of Jean-Pierre Orfeuil (2000). A constrained trip is designated as any trip whose destination is beforehand defined and chosen in relation to a daily activity involving a legal constraint or involvement (as a job contract or an activity under contract), the latter being mainly school or work.4 The home destination trips whose share remained constant during the period of the four surveys (about 40%) are considered to be nonconstrained trips (Table 1). If the focus is on the trips market at the global level, note the importance of nonconstrained trips over con- strained trips, which increased during the period and was even more striking when focusing on the average time and distance traveled per trip in the Paris region. During the period the average distance and duration of internal trips increased, but the rise was mainly the result of the growing importance of nonconstrained trips. The year 1991 should be identified as a turning point in travel patterns: the average trip duration went up by 6% in the late 1970s and 1980s and remained constant thereafter, and the average distance went up by 32% between the first survey and the last survey. The evolution of con- strained and nonconstrained trips follows respectively contracting evolutions, with the increasing weight of the share of nonconstrained trips in overall mobility. Although a surge by 20% of the traveled distance for constrained trips occurred between 1976 and 2001, the traveled distance for nonconstrained trips continuously 4 The following trips have been defined as constrained trips: trips whose purpose at destination is “usual and fixed working place,” “business,” “elementary school,” “secondary and technical school,” or “higher/university education.” increased by 39%. The time dedicated to work destination trips diminished considerably (–26%) during the period as the length of nonconstrained trips increased by 5.6%. Travel time in the Paris region remained higher than in the other parts of France. The last National Household Travel Survey, conducted in 2008, showed that although travel times in the Paris region are still the highest in France, they diminished between 1993 and 2008 as they increased dramatically in the provinces (Hubert 2009), showing a convergence of travel time. This brief overview of travel patterns in the overall Paris region shows that the increased number of trips is the result of the rise of nonconstrained trips, and this trend gains strength in regard to the duration and dis- tance of trips. This analysis of past trends will now be investigated by decomposing them according to gender and age at the individual level. evolution of moBility pAtternS of Women And Aging people in the pAriS region: A diSAggregAted point of vieW The total travel time budget5 for men stayed almost con- stant during the period between the surveys of 1976 and 2001; for women it increased by 15%. The breakdown is quite interesting because it appears that if the travel time budgets for constrained trips for men (35 min) are still higher than for women (20 min), they tend to converge with an increase for women by 14%. The increase in travel time budget is due to the rise in time dedicated to nonconstrained trips for women (Table 2). This result has to be put into perspective with the evo- lution of travel distance budgets,6 which had increased during the period at the global level. The growth by 26% of the travel distance budget for the internal trips dur- ing the period is the result mainly of increasing traveled distances for nonconstrained trips. The men’s travel dis- tance budget (19.5 km in 2001) is higher than the wom- en’s (13.5 km in 2001). The interesting point is that the 5 The travel time budget represents the total daily time devoted to traveling, measured as average time budget per individual. 6 Travel distance budget represents the total daily distances devoted to traveling, measured as average distance budget per individual. TABLE 1 Share of Trips, by Destination Purpose Trip Destination (%) Year Home Work Education Other Constrained Nonconstrained Total (1) (2) (3) (4) (5) (6 [3 + 4]) (7 [2 + 5]) (8 [6 + 7]) 1976 40 22 7 32 29 71 100 1983 39 21 8 32 28 72 100 1991 39 19 8 34 27 73 100 2001 40 18 8 34 26 75 100 Source: Authors’ calculations based on the four Paris Mobility Surveys (1976–1977, 1983–1984, 1991–1992, and 2001–2002).

195gENDER PATTERNS IN TRAVEL TIME AND DISTANCE IN THE PARIS REgION travel distance budget for constrained trips decreased for men during the period between the last two surveys. The women’s travel distance budget has increased by 44% at the total level, with a slight increase in the constrained trips distance and an important increase in the noncon- strained trips. A brief comparison with the results for the Lille area7 (surveys in 1976, 1987, 1998, 2006) (Krakutovski and Armoogum 2007; Merle et al. 2009), which presents a more clustered demographic and geographic structure, shows a decrease in the number of trips in 2006 com- pared with 1998 after a continuous increase. It seems to confirm, as in Paris, an increase during the period of travel time budgets and distance budgets, especially for women, and the development of car use for women even if they use mainly public transportation. The results of the Lille region for 2006 are quite close in regard to number of trips, with a daily average per individual of 3.76. The average traveled distance in this very clustered urban and industrial area is 15.8 km per individual per day. It can be observed that the average traveled distances in Lille are shorter than in Paris. This difference is the result of the segmented organization of the activities in the Paris region covering the sectors for housing in the east and the sectors for business in the center and the west in a territory of 12,000 km2, com- pared with the cluster organization of the activities in a territory of 612 km2 in the Lille region. The gender issue is also at stake in that region, with 17.5 km per individual per day for men and 14.4 km per individual per day for women. The difference in trav- eled distance is also explained by the shorter distances for commuting for women. Indeed, occupation matters a great deal in explaining the traveled distance. 7 An area of about 1 million inhabitants in the north of France at the border with Belgium, Lille has been historically industrialized by heavy industry (coal mining) and textiles. When the Paris region results are compared with with national results,8 the gender issue is confirmed for France in 1994, with a larger number of trips, shorter traveled distance, and larger time budgets for women. In 2008, the first results indicate that the average number of trips is decreasing in the urban area, with 3.6 trips per individual per day in 2008 compared with 3.8 in 1994. The average traveled distance per day per individ- ual increased by 6 km/individual/day between 1994 and 2008 in urban areas. The average total traveled time is quite stable with 56 min in 2006 (55 min per individual per day in 1994). Nobis and Lenz showed that in germany in 2002, gen- der issues in transport behavior are still at stake (2006). The number of trips for men is 3.26/individual/day against 3.52 for women (for full-time jobs), with respec- tive average traveled distance of 14.5 km/individual/day for men and 13.8 km/individual/day for women. During the period, male travel distance budgets remained higher than those of women, but they converge with a striking rise in the weight of nonconstrained trips. The female travel distance budget is 55% of the male travel distance budget in 1976 and 69% in 2001. Nobis and Lenz noticed an increasing resemblance in behavioral patterns, however, especially when the house- hold status is single (2006). They also showed that fam- ily life—the presence of children in the household—is an important determinant of travel patterns. gordon et al. (1989) showed that for a large u.S. sam- ple, once having standardized the different categories of 8 The results of the French National Household Travel Survey of 2009 are not yet published, but the first results are presented in Hubert 2009, “Dans les grandes agglomerations, la mobilité quotidienne des habitants diminue, et elle augmente ailleurs,” (INSEE Premiere, No. 1252, July). INSEE-INRETS is still working on the questionnaires, and average distance and average time in rural areas versus urban areas have been obtained. More detailed results, especially on gender issues, will be published in 2010. TABLE 2 Number of Trips, Travel Distance Budget, and Travel Time Budget, by Gender: 1976–2001 1976–1977 Survey 1983–1984 Survey 1991–1992 Survey 2001–2002 Survey Male Female Male Female Male Female Male Female Number of trips (ind./day) Total trips 3.6 3.3 3.6 3.3 3.5 3.4 3.5 3.5 Constrained trips 1.3 0.8 1.2 0.8 1.1 0.8 1.0 0.8 Nonconstrained trips 2.3 2.6 2.4 2.6 2.4 2.6 2.4 2.7 Distance budget (km/ind./day) Total trips 16.9 9.4 17.7 10.8 19.7 12.7 19.5 13.5 Constrained trips 7.4 3.0 7.6 3.4 8.3 4.0 7.7 4.2 Nonconstrained trips 9.5 6.4 10.1 7.4 11.4 8.8 11.8 9.3 Time budget (min/ind./day) Total trips 87.1 64.8 85.7 65.9 89.2 73.5 86.9 74.4 Constrained trips 35.4 17.7 34.0 18.2 34.3 20.4 35.3 20.2 Nonconstrained trips 51.7 47.1 51.7 47.7 54.9 53.1 51.6 54.1 Note: ind. = individual. Source: Authors’ calculations based on the four Paris Mobility Surveys Surveys (1976–1977, 1983–1984, 1991–1992, and 2001–2002).

196 WOMEN’S ISSuES IN TRANSPORTATION, VOLuME 2 population by level of income, occupation, structure of household, location, and traveled modes, women still always show shorter traveled distances for work trips than men and undertake a greater number of trips for nonworking purposes. As shown by Prédali (2001), although the travel pat- terns of men and women seem to be similar again in the Paris region, especially at the turning point of the 2000s, they are not overlapping and are far from being identical.9 This divergence can be explained by sociologi- cal habits and representations. This research shows that women have different mobility practices as a result of the structure of the family core, a variable that has a weak influence on men’s behavior. For equivalent pro- fessional and family patterns, men and women show different mobility behaviors measured through distance budget. The explanation has to be sought in regard to social identity and representations. Prédali shows that (a) even if the number of trips made by women is higher than in the case of men, the distance budgets are always much smaller, and (b) even when a woman is single with no children in her charge, she always chooses her home location to be close to her working location.10 9 The research is supported by face-to-face interviews conducted in the Paris region. 10 The authors would like to investigate whether or not the following sentence quoted in the full text just before the callout for this footnote (“even when a woman is single with no children in her charge, she always chooses her home location to be close to her working loca- tion”) is still valid according to the results of the updated survey. Results show, however, that women are ready to travel longer distances than in the past to secure an occupa- tion and to travel to their occupation place. The gender issue is more striking when investigating the difference between the working and the nonworking population. Time and Distance Budgets Related to Labor Force and Types of Jobs The gender tendencies and differences observed at the general level are confirmed when focusing on the labor force.11 The travel time budget of working men is higher than that of working women, but during the period they converge [103 min/individual/day (with a peak point at 108 min/individual/day in 1991) versus 81 min/indi- vidual/day in 1976, but 103 min/individual/day versus 89 min per individual per day in 2001]. The travel time budget for constrained trips declined for men from 51 min/individual/day in 1991 (peak point) to 45 min/indi- vidual/day in 2001 and increased for women from 31 min/individual/day in 1976 up to 33 min/individual/day in 2001. For both categories, the daily time dedicated to nonconstrained trips went up (Figure 2). The number of trips of working women is now con- sidered by types of jobs to identify the category facing this increase in mobility. For all categories of jobs, the number of trips increased during the period. During the 11 In France, the labor force or working population is composed of (a) people with jobs and (b) unemployed people looking for a job and ready to work within a month. 120 Tr av e l t im e bu dg et (m in. /in div idu al/ da y) 100 80 60 40 20 0 1976 30.9 Total trips, men Constrained trips, men Constrained trips, women Nonconstrained trips, men Nonconstrained trips, womenTotal trips, women 48.5 48.6 51.0 45.3 50.3 51.4 102.7 108.1 102.8102.9 89.589.6 82.381.2 57.457.1 54.254.5 56.756.2 30.9 33.4 32.7 1983 1991 2001 FIGURE 2 Evolution of time budget according to gender patterns in the working population: 1976–2001.

197gENDER PATTERNS IN TRAVEL TIME AND DISTANCE IN THE PARIS REgION period, the number of constrained trips was almost sta- ble, but for the number of nonconstrained trips, the rise concerns craftsmen and merchants and workers (labor) and pensioners. In 1976 the number of trips for women belonging to the category “staff and upper middle class” was at its highest peak and declined thereafter. The peak occurred at an important moment in the emancipation of women (Figure 3). In comparing these results with working men’s cat- egories, the same trends are observed for men except for the decrease in constrained trips. The curves are flat- tened and the staff, upper middle class category did not face such an increase in the case of men compared with women. Also observed is that the mobility of pension- ers and “other inactives” was higher for men than for women. When crossing age, gender, and occupation, it is interesting to draw a comparison with the evolution in similar urban community, for example, the San Fran- cisco Bay Area. gossen and Purvis (2004) obtained interesting results for the 1990–2000 decade. They confirmed the fact that occupation matters in explain- ing longer traveled distance and time per day per indi- vidual. That is, in 2000, working women younger than 60 years of age spent more time on travel than unem- ployed men did. But they found that “differences in travel time expen- ditures between working women and men in 2000 equal- ized from 1900 to 2000” (gossen and Purvis 2004, p. 4). Although in 1990 women between 18 and 22 years of age and between 40 and 64 spent less time in travel than working men, the difference disappeared in 2000 except for women in their 50s. During the decade, as in the Paris region, the nonworking population increased its travel time expenditures in 2000 compared with 1990. Travel Time and Distance Budgets for Women Related to Housing Location Measuring mobility in regard to number of trips with a Paris region destination at the total level yields mixed results: although the frequency of trips whose desti- nation is Paris decreases in some of the nearest areas (Val-de-Marne, Hauts-de-Seine), the fringes of the Paris region faced a huge increase in the number of trips. This increase is confirmed when looking at both travel dis- tance and time budgets (see Figure 4). The travel time budgets of the different parts of the region converged to reach an average of 80 min/individ- ual/day between 1976 and 2001. The most important increase (14%) in travel time budget and travel distance budget (50%) occurred for residents living in the Seine- et-Marne Department, where the housing density has risen as a result of this department’s low real estate prices (purchasing and renting). The latter is a semirural area N um be r o f t rip s pe r w om an p er d ay 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0 Fa rm er Cra ftsm an . m er ch an t Sta ff, up pe r m idd le cla ss Pro fes sio na l–in ter m ed iat e Em plo yee La bo r Stu de nt Pe ns ion er Oth er ina ctiv e Nonconstrained trips 2001 Nonconstrained trips 1983 Nonconstrained trips 1991 Nonconstrained trips 1976Total trips 1976 Total trips 1983 Total trips 1991 Total trips 2001 Constrained trips 1976 Constrained trips 1983 Constrained trips 1991 Constrained trips 2001 FIGURE 3 Evolution of number of trips for women by social categories: 1976–2001.

198 WOMEN’S ISSuES IN TRANSPORTATION, VOLuME 2 with little access to public transportation except in the so-called “new cities” of Melun and Marne la Vallée. The travel time budget of inhabitants in Paris increased by 3% and the travel distance budget by 18% during the overall period. In regard to distance, it is easy to identify three zones: (a) inner Paris with an average travel distance budget of 11.0 km/individual/day; (b) the inner belt with an average of 13.5 km/individual/day, and (c) the outer belt (semiru- ral areas) with an average of 21.7 km/individual/ day. The gender differences are still at stake in each area with lower travel time and distance budgets for women and converging trends. But both men’s and women’s travel behavior seem to be ruled by the following “law”: the farther you live from Paris, the greater is your daily traveled distance and time budget for transportation. The converging trends are more striking in the fringes of the Paris region than in the center and areas close to Paris. Schwanen et al. (2001) find similar results for the Netherlands, showing that the type of urbanization (monocentric versus polycentric) has a clear impact on travel patterns: there is more use of public transport in the core of the cities and less in the suburbs, and shop- ping distances and exchange commuting distances are longer in the suburbs than in the centers. They show that the urban structure affects the traveled distances only for commuting trips. Concerning the issue of the type and level of urbanization, personal attributes such as gender and the structure of the household matter a great deal: in any case, women travel shorter distances, confirming the results of Johnston-Anumonwo (1992) and Turner and Niemeier (1997). Modal Choice Housing location explains not only increasing time and distance budgets but also modal choice, at the global level. The weight of the modal share of cars is over- whelming, from 31.2% of trips in 1976 up to 43.6 % in 2001. The share of public transportation increased slightly from 18.6% to 19.5 %. The share of soft modes (walking and cycling) decreased from 41.9% to 34.5 % in 2001 as a result of the new urban framework and the room dedicated to cars, especially in the suburbs. Results are mixed between the different areas: indi- viduals walk more in the center of Paris than in the other areas and, not surprisingly, the car is the main mode of transportation in the outer belt areas. Looking closer at the gender issue in modal choice, it can be seen that women are using public transport and walking more than men as modes of transportation. This result is not surprising, even common. Dupont-Kerlan and Fontaine (2002) show that women have a higher propensity and willingness to use public transport, espe- cially in the Paris region (see Figure 5). In Lille (Merle et al. 2009), 61% of the total daily distance traveled by men is by car, whereas for women it is only 51%. In the last survey of 2006, however, men traveled by public transportation more than women compared with 1998. Of the average traveled distance of 17.5 km/individual/day for men, 12.7 km are by car— 10.7 km as driver and 2.0 km as passenger—and 3 km are by public transportation. In the case of women, of the average traveled distance of 14.4 km/individual/day, 10.4 km are by car—7.4 km as driver and 3.0 km as passenger—but only 2.7 km are by public transporta- 35 30 25 20 15 10 5 0 75 77 78 1976 1983 1991 2001 91 Departments—IDF 92 93 94 95 Pa rt (% ) FIGURE 4 Share of number of trips in the IDF with Paris as the destination, by department: 1976–2001. See Figure 1 for departments corresponding to numbers given, e.g., Paris = 75.

199gENDER PATTERNS IN TRAVEL TIME AND DISTANCE IN THE PARIS REgION tion. Men and women walk and bike almost equally, with 1.2 km for men and 1.3 km for women. The interesting point for the Paris region is the increase in car use among women during the period. This evolu- tion of modal choice and travel patterns toward more car dependency may be related to the improvement in women’s access to driver’s licences in the past 20 years (see Figure 6). Nobis and Lenz also showed an increasing share of women with a driver’s licence in germany in 2002 compared with the past decades and also increasing car availability, especially for women having part-time jobs (2006).12 12 Most employed women in germany are turning to part-time jobs because school ends at 1:00 p.m. and they have to pick up their chil- dren at school, most of the time by car. Travel Time and Distance Budgets Related to Age Class The analysis of travel time budget by age class shows sta- bility in the mobility of the 5- to 9-year-old age class and the 10- to 14-year-old and 15- to 19-year-old classes. In the past two surveys, the travel time budget had increased mainly for the age classes between 20 and 24 and 40 and 44 years. During the period the increase in the mobility of seniors is noticeable. For all classes beyond the 20- to 24-year-olds, the time dedicated to nonconstrained trips rose during the period, especially in the last two surveys. The curve of travel time budget for nonconstrained trips has two maximums: the 25- to 29-year-olds and the 65- to 69-year-olds, which can be explained by life cycle and job activities. The most important travel time budgets for constrained trips 25 20 15 10 5 0 Nonmobile Public transport Car Two wheels Walk Other Men 1983 Women 1983 Men 2001 Women 2001 Pe rc e n t FIGURE 5 Modal shares for men and women in 1983 and 2001. 40 35 30 25 20 15 10 5 0 Without license With license Without response 1983 1991 2001 Pe rc e n t FIGURE 6 Share of women having a driver’s license in 1983, 1991, and 2001.

200 WOMEN’S ISSUES IN TRANSPORTATION, VOLUME 2 are identified for the age classes of the middle of the life cycle until the 55- to 59-year-old class. In 1976, the 60- to 64-year-old and 65- to 69-year-old age classes had a more important travel time budget for constrained trips than later because of the higher legal age of retirement at that time (the legal age was then 65 years but was changed to 60 in 1981; the law applied from 1982 to 2008). FORECASTS SUPPORTING CONVERGING MOBILITY PATTERNS In this concluding section, results are presented briefly on the basis of past trends and demographic patterns and also on future mobility trends in the Paris region. The travel time budget and the travel distance budget are identified and forecast with the help of an age cohort model.13 13 These are the results obtained by the age cohort model Mi,p = aik0 · Ai + gk · Ck + ei,k,p where Mi,p = mobility expressed either as a travel time budget (min/indi- vidual/day), or travel distance budget (km/individual/day). The index i designates the age class of individuals during period p. aik0 = parameters measuring the influence of the age of the individual on mobility. The fundamental assumption is that the behavior generated by the age of the individual is stable over time. These parameters define a curve of profile-type for the life cycle for the generation of reference k0. Ai = age indicators of individuals (age classes are in 5-year incre- ments from 5 to 9 years old up to 80 years and over). gk = parameters measuring the lags between each cohort and the profile type of the reference generation gk0 = 0. Ck = indicators of 5-year cohorts (before 1911, from 1911 to 1915, …, 1991 to 1995). bp = parameters measuring the distortion of profile type along the life cycle due to period effects. ei,k,p = error term. The modeling is supported by demographic forecasts provided by INSEE until 2030. The latter underlines an aging population even in the Paris region, especially in the Seine-et-Marne Department. The number of people over 80 years of age will double; however, the share of people under 20 years will be more important in the Paris region than in the other parts of France. The labor force between 20 and 59 years old will be stable. The forecasts determined show a slight decrease in travel time budget from 79.7 min per individual per day down to 78.9 min per individual per day, this trend being a consequence of a decreasing travel time budget for men and for motorized households. Forecasts show a decrease in total travel time budgets for the age classes of 25 to 29, 30 to 34, and 35 to 39 and an increase in travel time budgets for the age classes beyond 55 to 59. The increased trend of travel time bud- gets for nonconstrained trips is confirmed in the fore- casts; especially for the age classes over 40 to 44. The travel time budget for constrained trips should decrease for males between 2010 and 2030, and such trips will be stable for women (even when broken down into constrained and nonconstrained trips). The travel time budget for nonconstrained trips should increase for all parts of the population according to either gender or age patterns, even though gender differences are still at stake: the rise will be by 7% for men and only by 1% for women for the period between 2010 and 2030. The important change in travel patterns is forecast to occur in the travel distance budget for women. Indeed, at the total level, the travel distance budget is stable, but it Forecasts are done from 1975 to 2030 in 5-year steps. The generation of reference is that of 1966–1970. The database is a pseudo-panel made up of the four Parisian mobility surveys and the demographic forecasts made by INSEE. 80 70 60 50 40 30 20 10 0 Total trips Constrained trips Nonconstrained trips 1975 1980 1985 1990 1995 2000 2005 2010 2015 2020 2025 2030 Tr av e l t im e bu dg et (m in. /in div idu al/ da y) FIGURE 7 Forecasts of travel time budgets for women in the Paris region until 2030.

201gENDER PATTERNS IN TRAVEL TIME AND DISTANCE IN THE PARIS REgION increases continuously during the period for women (see Figures 7 and 8). The travel distance budget of the population should stabilize from 17.5 km/individual/day in 2010 to 18.5 km/individual/day in 2030, the distance of constrained trips being stable at about 5.0 km/individual/day and the distance for nonconstrained trips increasing from 12 km/individual/day up to 13.3 km/individual/day. The travel distance budget for total trips for women is forecast to increase by 2030 by 9%, the distance traveled for constrained trips by 6.8%, and the dis- tance traveled for nonconstrained trips by 14.8%. For the same period, the distance budget for men should rise by 4% for total trips, with a decrease of 12% for constrained trips and an increase of 11.4% for non- constrained trips. concluSion The forecasts based on a demographic model confirm past trends with an increase in travel distance budget for women, for both purposes. Although men’s noncon- strained trips are increasing, it seems that they are not a substitute for those usually made by women (driving children, shopping for groceries). These additional trips made by men are not partly replacing those traditionally done by women. They are done for their own purposes and activities (sport, shopping). This analysis of Parisian individuals’ time used for transport activities should be completed by deeper socio- logical investigation, the analysis of the impact of the evo- lution of working hours (from 40 h to 35 h per week), and the impact of the geomorphologic evolution of the area, with a new cluster of housing and working activities. AcknoWledgmentS The authors thank the anonymous referee of the 4th Transportation Research Board Conference on Wom- en’s Issues for his or her comments and, for their care- ful proofreading, Kay Axhausen, Yoram Shiftan, Alan Pisarski, Jean-Loup Madre, Wendell Cox, Jean-Paul Hubert, Patrick Bonnel, Françoise Potier, and Bernard Quételard. referenceS Berri, A. 2005. Dynamiques de la motorisation et des dépenses de transport des ménages. Analyses sur données indivi- duelles et semi-agrégés. PhD dissertation. French National Research Institutie for Transportation and Its Security. Blanchard, N., and A. Ferrer. 2007. L’Ile-de-France parmi les régions européennes-une population jeune mais présente sur le marché du travail. INSEE, Collection Ile-de-France à la page, n° 282, juin. Brière, L., and Y. guilbert. 2001. Population active, emploi et chômage. INSEE Ile-de-France, Etudes, n° 40, juin. Bush, S. 2003. Forecasting 65+ Travel: An Integration of Cohort Analysis and Travel Demand Modeling. Depart- ment of Civil and Environmental Engineering, Massachu- setts Institute of Technology, Cambridge, uSA. Dupont-Kerlan, E., and H. Fontaine. 2002. Femmes, villes et transports. Communication Colloque femmes et villes. Conference on women and urban areas, 8 et 9 march 2002, Tours, France. gallez,. C. 1994. Identifying the Long Term Dynamics of Car Ownership: A Demographic Approach. Transport Reviews, Vol. 14, No. 1, pp. 83–102. gordon P., A. Kumar, and H. W. Richardson. 1989. gender 18 16 14 12 10 8 6 4 2 0 Total trips Constrained trips Nonconstrained trips 1975 1980 1985 1990 1995 2000 2005 2010 2015 2020 2025 2030 Tr av e l d ist an ce b u dg et (m in. /in div idu al/ da y) FIGURE 8 Forecasts of travel distance budgets for women in the Paris region until 2030.

202 WOMEN’S ISSuES IN TRANSPORTATION, VOLuME 2 Differences in Metropolitan Travel Behaviour. Regional Studies, Vol. 23, pp. 499–510. gossen R., and C. L. Purvis. 2004. Activities, Time and Travel: Changes in Women’s Travel Time Expenditures, 1990– 2000. Metropolitan Transport Commission, Dec. Hubert, J-P. 2009. Dans les grandes agglomérations, la mobi- lité quotidienne des habitants diminue, et elle augmente ailleurs. INSEE PREMIERE, n°1252, July. Johnston-Anumonwo, I. 1992. The Influence of Household Type on gender Differences in Work Trip Distance. Pro- fessional Geographer, Vol. 44, No. 2, pp. 161–169. Krakutovski, Z., and J. Armoogum. 2007. La mobilité quo- tidienne des Lillois à l’horizon 2030. Population, INED, Vol. 62, No. 4, pp. 759–787. Merle, N., A. Dupont, B. Quételard, and L. Hivert. 2009. Dia- gnostic Énergie Environnement des Déplacements (DEED) 1987/1998/2006 pour la Communauté urbaine de Lille. INRETS et CETE NP pour LMCu, Région Nord et ADEME Nord, published as Inrets/Ademe report in Oct. 2009. Nobis, C., and B. Lenz. 2006. gender Differences in Travel Pat- terns: Role of Employment Status and Household Struc- ture. In Conference Proceedings 35: Research on Women’s Issues in Transportation: Report of a Conference; Volume 2: Technical Papers, Transportation Research Board of the National Academies, Washington, D.C., pp. 114–123. Orfeuil, J.-P. 2000. L’évolution de la mobilité quotidienne. Les collections de I’NRETS (Synthèse n° 37), Paris, 146 p. Prédali F. 2001. La mobilité comme révélateur de l’évolution des modes de vie des femmes. PhD dissertation. Document available at the library of INRETS (French National Insti- tute for Transportation and Safety). Schwanen T., F. M. Dieleman, and M. Dijst. 2001. Travel Behav- ior in Dutch Monocentric and Policentric urban Systems. Journal of Transport Geography, Vol. 9, pp. 173–186. Turner T., and D. Niemeier. 1997. Travel to Work and House- hold Responsibility: New Evidence. Transportation, Vol. 24, pp. 397–419. Wiel, M. 2002. Ville et automobile. collection ‘les urbanités.’ Ed. Descartes et Cie. Zahavi Y., and J. M. Ryan. 1980. Stability of Travel Compo- nents over Time. In Transportation Research Record 750, TRB, National Research Council, Washington, D.C., pp. 19–26. reSourceS Armoogum, J., and J.-L. Madre. 1997. Accuracy of Data and Memory Effects in Home Based Surveys on Travel Behav- iour. Presented at 76th Annual Meeting of the Transport Research Board, Washington, D.C. Berri, A., J.-L. Madre, and Y. Bussière. 2005: Etalement urbain, contexte économique et croissance de la circulation auto- mobile des ménages en France: Projections à l’horizon 2020. Recherche Transports Sécurité, Vol. 89 (Oct.–Dec.), pp. 305–326. Bresson g., J. Dargay, J.-L. Madre, and A. Pirotte. 2004. Economic and Structural Determinants of the Demand for Public Transport: An Analysis on a Panel of French urban Areas using Shrinkage Estimators. Transportation Research Part A, Vol. 38, pp. 269–285. Dargay J., J.-L. Madre, and A. Berri A. 2000. Car Owner- ship Dynamics Seen Through the Follow-up of Cohorts: A Comparison of France and the uK. In Transportation Research Record: Journal of the Transportation Research Board, No. 1733, TRB, National Research Council, Washington, D.C., pp. 31–38. Krakutovski, Z. 2004. Amélioration de l’approche démographi- que pour la prévision à long terme de la mobilité urbaine. Thèse de doctorat. Direction de thèse: Jean-Pierre Orfeuil et encadrée par: Jean-Loup Madre et Jimmy Armoogum. Institut d’urbanisme de Paris de l’université Paris XII. Madre J.-L., and J. Maffre. 1997. La mobilité régulière et la mobilité locale en 1982 et 1994. Insee Résultats—Con- sommation Modes de vie, n° 88–89. Madre, J.-L., A. Pirotte, and g. Bresson. 2000. Comment faire converger les approches économique et démographique de la demande de transport? Actes des Journées de Méthodo- logie Statistique (INSEE), Paris. All errors remain the authors’.