Navigation – Plan du site

AccueilnumerosHors-sérieFor a temporal ecology of rush-ho...

For a temporal ecology of rush-hours: a survey of choices of work hour in the Paris region

Pour une écologie temporelle de l’heure de pointe : enquête sur les choix d’horaires de travail en Île-de-France
Emmanuel Munch
Traduction de Richard Hillman
Cet article est une traduction de :
Pour une écologie temporelle de l’heure de pointe : enquête sur les choix d’horaires de travail en Île-de-France [fr]


Une part de plus en plus importante de la population active française accède à des horaires de travail qui ne sont plus soumis au contrôle explicite de l’employeur. Pourtant, en Île-de-France, la concentration des déplacements à l’heure de pointe du matin ne cessent de s’intensifier. Ainsi, avant de chercher à diluer la congestion des transports à l’heure de pointe, il devient nécessaire de comprendre les logiques d’action individuelles qui fondent un choix d’horaire de travail. Pour quelles raisons un salarié aux horaires flexibles arrive-t-il au travail en heure de pointe ? Guidées par une approche compréhensive et centrées sur les temporalités du quotidien, nos investigations s’appuient sur les résultats d’un questionnaire (3202 répondants) et d’entretiens (29) réalisés auprès de cadres du pôle d’activité de la Plaine Saint-Denis. Afin de décrire les stratégies temporelles qui expliquent une arrivée au bureau en heure de pointe, nous formulons trois hypothèses de recherche liées à trois dimensions : (I) Il subsiste toujours des contraintes de couplage (horaires d’école, de réunion) qui poussent les salariés flexibles à se rendre au travail durant la période de pointe. (II) Les salariés flexibles qui n’ont pas de contraintes de couplage préfèrent (loisir en fin d’après-midi, synchronisation avec les proches) arriver avant ou pendant l'heure de pointe. (III) Il existe des normes sociales d’horaires de travail (normes des salariés disciplinés et du cadre dévoué) qui canalisent les pratiques horaires en sanctionnant les arrivées trop tardives au bureau. Et finalement, c’est en dévoilant les entrelacs organiques de la synchronisation au travail que l’article ouvre des pistes opérationnelles pour désaturer les transports à l’heure de pointe du matin.

Haut de page

Texte intégral

The results presented in this article are the fruit of doctoral research funded by SNCF Recherche & Innovation and SNCF Transilien. My thanks to them for their confidence. I am also grateful to Anne Aguiléra and Marianne Thébert for re-reading the article. Finally, my thanks also to the two anonymous readers for the journal EPS.


1While the proportion of employees in the Paris region (Île de France) able to exercise some choice in their working hours is on the increase (Bouleau & Leroi, 2016), this article seeks to explain the persistence there of the morning rush-hour.

2This objective is motivated by the renewal of a paradoxical observation: in the Paris region, flexibility in working hours, far from leading to a spreading out of the times of arrival at the place of work, has been accompanied by an increased temporal concentration of arrivals in the office [Munch & Proulhac, 2019]. Indeed, in a region particularly subject to problems of transport congestion [Koning, 2012], from the beginning of the 1990s, employees have benefitted from more and more autonomy in deciding when they will come to work, yet arrival times at the workplace have not become staggered; on the contrary, with respect to public transport, they are increasingly concentrated between 9:00 and 9:30 a.m., the most critical point. The objective of this article is to set out the reasons for this paradox, adopting a comprehensive procedure: how to explain that the shift to flexible working hours coincides with an increased concentration in the times of arrival at work?

3This observation requires explanation, for it goes contrary to most of the common representations of the causation of rush-hour phenomena. One usually supposes that it is the factory sirens, the time-clocks at the entrance of the building, the supervision of the bosses and, more generally, the standardisation of work which are responsible for the convergence of working hours and the rush-hour phenomena [Godard, 2007; Thompson, 1969].

4As long as working hours were prescribed by in regulated and uniform fashion by employers, there was little mystery about the rush-hour phenomenon: it was accounted for by the imposed synchronisation of working hours. From the point of view of Économie des Transports [Small, 1982; Vickrey, 1969] and, at the level of the urban transportation networks, the freeing up of collective and regulated scheduling was perceived as a necessary and sufficient condition for resolution of the problem [Thoemmes, 2013; Hines, 1984; Baudraz, 1971]. In an age of the expansion of services and greater flexibility of working hours, the persistence of rush-hour phenomena calls for re-examination of these postulates and the theoretical models which underlie them.

5To examine the problem otherwise, this article positions itself in the disciplinary perspectives of Sociology of Time [Zerubavel, 1985] and Time Geography [Lenntorp, 1977], and adopts the scale of individual logics of action. The essential scientific issue resides in the association between a time of arrival at work and the reasons which justify this choice from the employee’s point of view.

6To achieve this objective, we will enquire into the motivation of salaried executives and users of public transport. All of them work in a centre of tertiary activity in the first Parisian ‘ring’: La Plaine Saint-Denis.

7In addition to addressing a theoretical objective, our subject of research also serves a practical purpose. We consider that by analysing the individual motivations which determine a choice of work scheduling, we will be able to identify new means of diminishing rush-hour flows. Indeed, in a context where more and more employees enjoy the freedom to choose their work hours, such means go beyond the staggering of schedules and a simple intervention in the form and mechanics of daily routines. Today, to produce a staggering of work hours, it has become important to understand what, in the substance and organic dimension of the daily lives of employees, explains their choice of schedule.

8In order to identify these explanatory factors, we will base ourselves first on a review of the literature dealing with programmes of daily activities and the causes governing the choice of work hours. In the second part, using quantitative and qualitative methods, we will present the outline of our investigative methodology and specify the particularities of our field of enquiry. Then we will successively test three different research hypotheses. In the third part, we will seek to verify that couple-related constraints (schedules involving school, meetings, a spouse’s work) remain, which impel employees on flexible hours to arrive at the office during the rush-hour. In the fourth part, we will test the idea that employees who are not compelled to conform to a couple’s needs generally prefer to arrive at work before or during the rush-hour. Finally, beyond the question of preferences and constraints, we will see whether social norms for working hours exist which impel employees to arrive at the office before 9:30.

1. Review of the literature

9The individual motivations justifying a choice of work schedule are still very often studied in a compartmentalised manner within different disciplines of the social sciences [Thorauge & al., 2016; Schwanen, 2008]. Consequently, most studies furnish only partial responses to the question of the reasons justifying these work schedule practices. Instead of being mutually exclusive, these models should complement each other in order to improve knowledge of scheduling practices and their implications for transport policies [Munch, 2017a].

1.1 The economy and Scheduling Preferences models

10Most research focusing on the question of journeys during rush-hours and avoidance of periods of congestion is based on econometric methods which seek to model the choice of a schedule for travelling to work. They generally refer to models of ‘Scheduling Preferences’ [Small, 1982], which themselves rely on the ‘Bottleneck Model’ [Vickrey, 1969] as a means of describing road congestion.

11According to these models, individuals choose their time for arriving at work, and therefore for the journey to get there, by performing a cost/advantage calculation [Börjesson, 2008] between a time during and outside rush-hours. They calculate the ‘cost’ to them of a rush-hour journey, in terms of time spent on the road or the discomfort experienced in public transport, compared with what an earlier or later arrival at work ‘would cost’ them, mainly in terms of sanctions or professional penalties. If these models accurately describe a part of the motivations for choosing a work schedule, they have the drawback of considering these motivations only superficially. In particular, one does not know what individuals are taking into account when they evaluate the sanctions and professional penalties to which they open themselves in deviating from standard work hours (9 a.m.-6 p.m.). This is part of what we seek to shed light on in this article.

12These models subsequently focus on the socio-demographic characteristics of the subjects of the enquiries to explain the choices made [Yang & al., 2016; Saleh & Farrell, 2005; Yoshimura & Okumura, 2001; Small, 1982]. The socio-demographic variables considered generally have reference to daily life experiences in society, in the family, at work and in public transport [He, 2013; Chin, 1990; McCafferthy & Hall, 1982; Ott & al. 1980]. Here, too, the motivations that explain the work schedule followed by an individual are deduced by supposition. For example, if it is observed that women are more likely to travel during rush-hour, the conclusion is quite frequently drawn that this is indirectly due to the fact that they, more often than men, drop off the children at school before arriving at work [He, 2013; Lavielle, 2008]. Yet empirically, nothing informs us that these women actually drop off the children at school before arriving at work. Therefore, even if analysis of behaviour according to socio-demographic variables remains interesting, it cannot suffice to understand with precision the raisons for one scheduling choice or another. Studying individuals’ daily ‘programmes of activities’ is indispensable for explaining their work schedule choices.

1.2 The Time-Geographic models

13In geography, on the basis of the precursor work of the Scandinavian school of the Geography of Time [Hagerstrand, 1970; Lenntorp, 1977], we are witnessing substantial development of new representational models involving a so-called ‘activity-based approach’ to travelling [McNally, 2000]. These models draw more on individuals’ programmes of daily activities than on their socio-demographic situation to model the temporal aspect of their journeys. For example, for parents of young children, the choice of the time to travel to work will be the result of coordination between the children’s school schedule, the time of a meeting on arrival at work, and the evening closing time of the supermarket [Drevon, 2016].

14Here individuals optimise management of their daily agendas by taking into account the ‘temporal constraints’ imposed on them [Lenntorp, 1977] – foremost among which one finds the ‘coupling constraints’ [Adoue, 2016]. These constraints relate to the absolute necessity of being on time for an ‘appointment’ in the broad sense, that is, of being in a place at a precise moment for a generally agreed-on duration (school hours, meeting hours…). Parallel to this, individuals also take account of their preferences when the activities in question offer the possibility of ‘temporal accessibility’ – that is, an extended time period during which the activity can be pursued (as is the case, for example, with the opening hours of shops). Consequently, with reference to these motivations oriented toward goals in an instrumental manner [Weber, 1921], we will seek to identify the couple constraints and the individual preferences which explain the decision to arrive at work during the rush-hour.

1.3 Sociology of time and work

15Certain sociological studies, however, have shown, contrary to the models previously described, that our management of time rather often depends on actions not oriented towards goals, or not in an optimal manner [de Coninck & Guillot, 2007] – notably because an ethical quality exists in individuals’ relation to time. When they decide on the amount of time an activity should take, or when it should take place [Zerubavel, 1985], individuals more or less consciously refer to ‘normes temporelles’ (temporal norms) [Rosa, 2013], to sets of values, to what may be perceived as ‘good’ or ‘bad’ timing.

16As Alvin Toffler [1971, p. 43] says, in the behaviour of adults, all the activities which we practice are based on explicit or implicit assumptions about their duration and their chronology. The standardisation of the durations of activities is closely tied to standardisation of the time at which the activity is supposed to occur. For example, the fact that a university course lasts an hour is often linked to the fact that it takes place between 2 p.m. and 3 p.m. Specifying the duration of an activity is necessarily associated with specifying the time-slot during which the activity will take place. In Western countries, this coordination between chronology and chronometry is based on the process of ‘scheduling’. In other civilisations, the temporality of activities is generally less pre-arranged and more spontaneous [Pritchard, 1940].

17In the same way, in industrialised countries, one does not necessarily eat when one is hungry, but when it is the proper time for dining. One does not necessarily go to bed one one is tired, but when the time is appropriate. The time for going to bed or for other social activities is the result of a normative prescription. This is a function of a reinforced conception of the notion of ‘punctuality’. Being late also amounts to deviance from the prescribed norm, with the potential consequence of a loss of social reputation [Weber, 1904].

18On the precise question of work, the sociological notion of a ‘norme sociale de temps de travail’ (social norm of working time) – 35 heures, 7 heures par jour… – exists and is often analyzed [Thoemmes, 2012; Bouffartigue, 2012]. Even if, at present, there is no concept defining the ‘social norms of work hours’, for our part we consider that analogous motivations should be detectible from the point of view of work scheduling. We assume that, for employees on flexible hours, early morning arrival – before the rush-hour – is associated with positive values, whereas, on the other hand, late arrival – after the rush-hour – is associated with negative symbols and values. This, moreover, is where the third objective of our research lies. In fact, this will partly consist in demonstrating the existence of ‘social norms of work hours’, which, by way of the social control of colleagues and the hierarchy, render complicated arrivals after 9:30 a.m.

19Certainly, there is a theoretical gap between the sociology of work, the sociology of time and the objects of research mid-way between working hours and the phenomenon of the rush-hour. Nevertheless, a review of the literature on the temporal aspects of journeys from home to work reveals especially strong conceptual barriers between economy, geography and sociology. This lack of connection between the different disciplines devoted to studying social phenomena in their temporal dimension had already been observed by William Grossin in his Introduction à l’écologie temporelle [1996]. Because this observation applies to our object of research, this article aims at furnishing a unified and coherent representation of the motivations of individual behaviour which account for a decision to arrive at work during the rush-hour.

2. Field and methodology

20To verify our research hypotheses, we chose to survey a population of employees working in the headquarters of large service-sector companies. We also opted for a method combining quantitative investigation by questionnaire and qualitative investigation by semi-structured interviews. The qualitative investigation permits deeper analysis of the constraints and preferences accounting for a choice of work schedule, but also a response to our third research hypothesis concerning the influence of social norms on work schedule practices.

2.1 The commercial centre of La Plaine Saint-Denis

  • 1 The regional transportation authority which manages the RER (Réseau Express Régional) is also the o (...)
  • 2 Number of usable responses after verification of the database. The questionnaire, to be completed a (...)

21The location of our study has the advantage of accentuating and catalysing the chief characteristics bearing on our subject of research. It is situated in the commercial zone of La Plaine Saint-Denis, a centre of service-sector activities to the north of the city of Paris, which assembles some 40,000 workers and is in close proximity to two railway stations (SNCF), La Plaine-Stade de France (RER B), to the east, and Stade de France-Saint-Denis (RER D) to the west. Within this area of our study (Map 1), we benefitted from partnerships established with SNCF Transilien1 and eight tertiary enterprises. These eight enterprises concentrate somewhat more that 12,000 employees, most of whom are executives benefitting from flexible work-hours and mainly using public transport to get to work. On the basis of this research field, we rely on the combined application of a quantitative survey (3,202 questionnaire responses2) and a qualitative investigation (29 semi-structured interviews) involving employees of these eight enterprises.

Map 1 – Localisation of the enterprises surveyed

Map 1 – Localisation of the enterprises surveyed

Source: BDD GEO SIRENE. List of enterprises updated as of 1 January 2017. BD ORTHO de l’IGN. Photo 2011. Created by E. Munch and C. Pivano. Note : Filtered by establishments with a minimum of 50 employees.

2.2 The investigation by interview

  • 3 5 persons presented a profile too indistinct to permit them to be identified in the questionnaire d (...)

22The process of recruiting employees surveyed by semi-structured interview (29 interviews) followed the investigation by questionnaire, with volunteers able to give their contact information at the end of the questionnaire. The average profile of those interviewed deliberately varies from that of those surveyed by questionnaire, since interviews included only people who felt more particularly concerned by the issue of the rush-hour and the organisation of their daily agenda. We note that, with the agreement of these interviewees, we were able to cross the comments (in the interview) and the responses (to the questionnaire) provided by the same person surveyed.3 But the principal methodological difficulty presented itself with regard to the construction of the questionnaire, that is, the formalisation of the factor to explain and the explanatory factors.

2.3 The survey by questionnaire

23First, a determination of what we consider the rush-hour of arrivals – that is, the period of one hour including the greatest number of arrivals in the area studies in the course of a single day – was made, based on exit validation data from the stations of La Plaine Saint-Denis. It occurs between 8:30 a.m. and 9:29 a.m. This time-slot includes practically half the arrivals recorded in the zone for a Tuesday business-day morning (Figure 2). Moreover, this period also corresponds to the rush-hour arrivals at work by public transport as compiled on the level of the Paris region (Île-de-France) [Munch, 2017b].

Figure 2 – Distribution of exit times at the Stade-de-France Saint-Denis station, by 30-minute periods

Figure 2 – Distribution of exit times at the Stade-de-France Saint-Denis station, by 30-minute periods

Reading: At the Stade-de-France Saint-Denis station, 22.6 % of the exits effected during the morning (6 a.m.-11 a.m.) on business-day Tuesdays in 2015 took place between 9:00 a.m. and 9:29 a.m.

Source: Station exit data for all the business-day Tuesdays in 2015. Centre TVV–SNCF Transilien.

  • 4 Twelve reasons identified as well as one response ‘Others’; The respondents selected on average som (...)

24Secondly, we decided to ask the employees directly to classify, in order of importance, the factors determining their time of arrival at work. The respondents could choose to classify between one and thirteen responses4 (Table 3). In order to facilitate interpretation of the reasons for an arrival between 8:30 a.m. and 9:29 a.m., for each of the surveys, we selected the reason classified in first position.

Table 3 – Distribution of reasons accounting for working hours, depending on whether they related to couple-related constraints or individual preferences

Areas of daily life

Decision-makers directly involved

Reasons for the schedule

Couple-related constraints


The enterprise and its economic environment

Production norms

Schedules of economic partners


The individual and his enterprise

Hours fixed by the employer

Meeting hours

Hours of colleagues


The individual and his environment outside work

Day-care hours

School hours

Transport hours

Partner’s schedule

Extra-professional activity



The individual

Avoidance of rush-hour

Personal preferences

Source: The author

25The other explanatory variables present in the analysis are all control variables, which make it possible to associate the hours of arrival at work with certain categories of population. These population categories are envisaged from three perspectives defining the identity of the respondent with regard to his or her work schedule.

26The first perspective permits definition of individuals in the sphere of social and family life through the following variables:

  • zone of residence, sex, age, marital status, number of children, educational level.

27The second describes their identity at work through questions on:

  • the socio-professional category, the enterprise they belong to, the frequency of distance-working, their theoretical working time, their actual working time.

28The third draws their profile as commuter and their journey to get to work:

  • modes of travel used, combination of modes, long journey on the RER or not, total duration of journey.

29In order to be exhaustive, we could have selected other control variables based on exploitation of our questionnaire, but we chose to limit the number of explanatory variables to those that proved especially useful. The variables not selected are those which, based on correlation analyses and preliminary modelling, do not improve – or do so too little – our ‘explanatory model’ of the hour of arrival at work.

30Multinomial logistic regression of the period of arrival at work [McCafferthy & Hall, 1982] makes it possible to effect multivariate analyses of the determinants of a time-slot for arrival at work. This type of modelling permits us to integrate the characteristics of individuals with couple-related constraints or individual preferences in order to explain the choice of a schedule. The principal interest of logistic regression here is precisely to isolate – all other things being equal – the influence of a variable on the schedule followed.

31The time-slot of reference chosen is the period from 9:00 a.m. to 9:29 a.m., since that is the period which assembles the largest number of arrivals.

3. Arriving in the rush-hour as a response to couple-related constraints

32Basing ourselves on the date of our questionnaire and the interviews undertaken with employees, we sought to verify the hypothesis that the employees benefit only in appearance from greater scheduling liberties and that constraints of coordination remain in daily life that impel them to travel during the rush-hour. We will show that these couple-related constraints are mainly of three kinds, and that they especially concern certain parts of the population.

33The table presented below summarises the effect of each of the modes chosen on the probability of travelling to work in each of five time-slots of thirty minutes. The time-slots on a red background represent the morning rush-hour (8:30 a.m.-9:29 a.m.).

Table 4 – Marginal effect of explanatory variables on the probability of arriving at work during the time-slot indicated

Probability for the individual of reference of arriving during the time-slot

Before 8:00 a.m.

8:00 a.m.-8:29 a.m.

8:30 a.m.-8:59 a.m.

9:00 a.m.-9:29 a.m.

9:30 a.m. and after

(in %)








Reason stated for the time (Fixed schedules)

Production norms






Hours of economic partners






Meeting hours






Hours of colleagues






Day-care hours






School hours






Transport hours






Spouse’s hours






Extra-professional activité






Avoidance of rush-hour






Personal preferences












Zone of residence (Paris)

Inner ring






Outer ring






Outside Paris region






Sex (Man)







Age (35-44 year)

Less than 25 years






25-34 years






45-54 years






More than 55 years






Marital status (Maried/legal couple)







Living as a couple


















Number of children supported (None)

1 child






2 children






3 or more children






Diploma (secondary)

Baccalaureate or under






First degree






  • 5 Public transport
  • 6 Personal vehicle

Probability of arriving during the time-slot

Before 8:00 a.m.

8:00 a.m.-8:29 a.m.

8:30 a.m.-8:59 a.m.

9:00 a.m.-9:29 a.m.

9:30 a.m. and after

(in %)








Socio-professional category (executive)













Team Leader






Upper-level executive






Enterprise (Insurance)











































Frequency of distance-working (Never)

Certain weeks






Every week






Theoretical work hours (Fixed schedule)

Arrival time-slots






Free hours






Work hours effected (Fixed schedule)

Arrival time-slot






Random hours






Duration of journey on RER (0 or <20 min.)

Superior or equal to 20 min.






Mode of transport (Only PT5)

Only PV6






Only others






PV and PT






Journey time (average = 58 min.)






Effect of an additional minute






Significativity of coefficients: ***to 0.001; ** to 0.01; to à 0.1; otherwise absence of significativity.
R² of McFadden =0.188 Note: The situation of reference corresponds to the modes in parentheses. The modes of reference are the most frequent characteristics for each of the variables.

34Reading: the individual possessing all the reference characteristics serves as a point of comparison. He or she has all the characteristics indicated in parentheses. For the individual possessing all the reference characteristics, the probability of arriving at work before 8:00 a.m. is 3.7 %. The values indicated in the columns ‘Deviation from the Reference’ describe, for each of the variables, the distance from the reference mode. Consequently, all other things being equal, persons residing outside the Paris region have, relative to Parisians, a 11.4 % greater chance of arriving at work before 8:00 a.m. In other words, a non-Parisian otherwise having all the reference characteristics has a 15.1 % (3.7+11.4) chance of arriving at work before 8:00 a.m. Source: Rush-hour survey - Saint-Denis - May 2015 (3,202 respondents).

3.1 Rush-hour appointments: school hours, meeting hours, the partner’s schedule

35The rush-hour, because it assembles the greatest number of employees travelling, is the period symbolising the ‘common denominator’ of social life and daily interactions. It sums up the coordination of couple-related constraints for the largest number. It permits the entire population to come together at work while retaining the possibility of making arrangements with other individuals in one’s social circle (family, friends, shopkeepers…) before or after work.

36Thus, for the parents of school-age children, school hours fundamentally structure the beginning of the working day. In Paris, the hours are all effectively the same: start of classes at 8:30 a.m. and opening of the gates at 8:20.

37The times of arrival at school are stringent. At most, one may drop off one’s child ten minutes before the buzzer. The working hours of parents who drop off their children at school consequently conform to the same stringency, to which must be added the duration of the journey from home to school to work.

38Based on this reading, the arrival at work of parents dropping off their children at kindergarten or primary school around 9:00 a.m. is easier to understand. By comparison with a reference individual whose hours of work are fixed and determined by the employer, salaried parents who benefit from flexible working hours and whose schedule is determined by the school-hours of their children actually have an 8.1 % greater chance of arriving between 9:00 a.m. and 9:29 a.m. at the office (Table 4).

39The time of the first meeting of the day also constitutes a real determinant of arrival during the rush-hour: those surveyed whose time of arrival is chiefly determined by the times of meeting have an 8.5 % greater chance of arriving at work between 9:00 a.m. and 9:29 a.m. than persons whose working hours are fixed. As Jean explained, by convention the first meeting almost always takes place between 9:00 a.m. and 9:30 a.m.:

The first meetings are always at 9:00 a.m., 9:15 or 9:30. During my whole career, I have had only one meeting before 9 o’clock, and that was because we were in teleconference with people in another time-zone. (Jean, about thirty years-old, unmarried without children, resident in the 94th Département, arrival around 9:00 a.m.)

40The pressure of meetings as a factor determining the time of arrival in rush-hour is dominant in our population, constituted mainly of executives. As a result, this pressure often comes in conflict with the children’s school-hours. Indeed, for working parents, these two factors can coincide and enter into opposition. How to manage so as to drop off one’s child at school on time and not be late for the meeting?

Dropping off the children is a real problem. With my wife, we have arrived at the following arrangement: every other week I drop them off at school and she picks them up, and the following week we do the opposite. From time to time this may be adjusted. When I have meetings before 9:00 a.m. and I cannot change them, we reverse the order normally in place for dropping off and picking up the children. (Florent, about thirty years-old, married with children, executive, living outside Paris region, arrival around 8:30 a.m.)

41Finally, again mid-way between couple-related constraints and personal acquaintances, we find the pressure of the partner’s work schedule. With reference to an individual possessing all the reference characteristics, those surveyed who stated that their time for starting the day was mainly determined by the working hours of their partner have a 13.4 % greater chance of arriving at work between 8:30 a.m. and 8:59 a.m. The (work) schedule of the partner becomes determinant, if he or she is subject to stronger constraints than the work schedule of the other member of the couple. By a ricochet effect, one of the members then adapts to the strict constraints imposed on the other. These constraints most often apply with regard to arrivals just before 9 o’clock. The schedule of the partner does not exercise its ‘constraining’ role in the couple, however, except when other activities or persons ‘more constraining’ do not carry more weight in the schedule:

Since we no longer have a child of school age, it is really the partner’s schedule that has become the determining factor. (Alexis, about fifty years-old, maried with one child, executive, living in the 91st Département, arrival around 8:00 a.m.)

3.2 Sociology of the rush-hour: women at the centre of the debates

42Along with the couple-related constraints previously studied (school hours, meeting hours, the partner’s schedule), individual characteristics also have an impact on the choice of time for arriving at work, even if these seem less clearly to lead to arrivals during the rush-hour. According to Table 4, the majority of individual characteristics (age, marital status, socio-professional category, educational level, zone of residence) seems in general to separate the population between those coming before 9:00 a.m. (‘morning persons’) and those coming after 9:00 a.m. (‘late-comers’). While couple-related constraints (school, meeting, partner) separate individuals more between the rush-hour and otherwise, the characteristics particular to the individuals seem to separate them around a symmetrical axis situated at the high point of the rush-hour – at 9:00 a.m.

43There is only one socio-demographic variable which, according to the response considered, conduces either to more arrivals during the rush-hour or to more arrivals outside the rush-hour – namely, gender.

44Independently of all the other characteristics and constraints which might characterise them on the questionnaire, women have a greater chance than men of arriving during the rush-hour. The time-slots between 8:00 a.m. and 8:59 a.m. do not yield any statistically significant probabilities. By contrast, for the earliest time-slot, prior to 8:00 a.m., women have less chance (–1.5 %) than men of arriving during this period. Similarly, and again in relation to men, women also have less chance (-3.6 %) of arriving after the rush-hour. On the other hand, women have a 3.0 % greater chance of arriving between 9:00 a.m. and 9:29 a.m., that is, during the rush-hour.

45The main interest of the observations derived from our logistic regression is that they make it possible to isolate the effect of each variable. Thus, in our model, women have a greater tendency to arrive during the rush-hour, all other things being equal. This is why one could not affirm that women arrive more often during the rush-hour because they more often drop off the children at school. Given that the effect of the variable ‘school hours’ is fixed when the scheduling practices of women and men are compared, it is therefore necessary to look for other possible explanations.

46Even if the gaps are being reduced, women in France, on the average, still spend 57 % more time than men in domestic activities [Brousse, 2015]. Moreover, working women are engaged in a range of spheres of daily life which require a great many interactions [Adam, 2004; Davies, 2001] but also coordination among them. Even if men seem almost equally responsible for dropping off the children in our enquiry (56 % of women and 44 % of men), women, on the whole, have more interactions with the children. They are also more involved in taking care of the elderly in their entourage and run more errands [Schintler, 2001]. They therefore have more daily constraints to orchestrate:

Reconciling private and professional life is, at any rate, tremendously important. We are always trying to do that. For women who didn’t work in the past, now we do, but we also have children, so we also need to take care of them. (Marianne, forty years old, divorced with children, executive, living in the 92nd Département, arrival around 8:50 a.m.)

47At an individual level, an arrival during the rush-hour very often reveals strategies of coordination and reconciliation of numerous couple-related constraints. It follows that, if women are faced with a greater number of couple-related constraints than men [de Palma & al. 2009; Schwanen, 2007], it is highly probable that, to manage all of them in the course of the day, they would be induced to arrive at work during the rush-hour. Given that a certain number of these constraints were not listed on the questionnaire (caring for children, for elderly persons, house-keeping, shopping…), it is possible that the pressure of these imperatives on arrival at work during the rush-hour is reflected in the sex of the person surveyed.

4. Arriving before or during the rush-hour as a response to preferences?

48At present, we will focus only on individuals who explained their time of arrival at work by their preferences. This sub-population includes 962 individuals, or a little more than 30 % of the population surveyed by questionnaire. In parallel, we will rely largely on the words of the interviewees who ticked one of the boxes on the questionnaire indicating that they choose their schedule mainly according to their inclinations. The employees studied are divided into two groups depending on the type of inclination selected. The group studied in the first section includes 412 who stated that they choose their schedule according to a specifically identified preference: they seek mainly to ‘avoid the periods when the transport network is too congested’ when they choose their working hours. The second group contains 550 individuals who stated that their schedule is chiefly determined by preferences much more general: ‘personal preferences’. Study of these two types of preferences will validate the hypothesis that, when they are not explicitly required to connect with others during their work, individuals prefer to arrive before or during the rush-hour.

4.1 Avoidance of congestion: a way of killing two birds with one stone

49First, the individuals who state that they seek to avoid the rush-hour are highly likely to arrive before the rush-hour and hardly ever after it. Compared with an individual whose schedule is determined by his or her employer, individuals preferring to avoid rush-hour congestion have a 31.3 % greater probability of arriving at work before 8:00 a.m. This strategy of advancing one’s schedule proves more effective than that of delaying it when faced with the string of inconveniences induced by congestion: discomfort, slow-downs, unpredictability of the length of the journey, difficulty of making connections, and even the risk of terrorist attack.

50Indeed, the time spent passing through ‘bottlenecks’ causes disruptions to be protracted and to have a greater impact on travellers who left after the rush-hour [Vickrey, 1969]. To benefit from the thinning out of the flood after the rush-hour, one must allow the bulk of the traffic to pass and arrive, not at 9:30 a.m., but rather at 10:00 a.m. or even 10:30 [Tribout, 2014]. Inversely, in leaving before the bulk of the traffic, it is sufficient to start out just before the rush in order to avoid its effects.

51Nevertheless, crossing the results of the questionnaire with the statements of the interviewees revealed that the choice of a work schedule is in reality determined only very rarely by a strict search for improved travelling conditions – even for the populations most sensitive to travel conditions (long journeys, older persons). In the light of the explanations of the interviewees, it turns out that it is instead preferences regarding the timing of periods of activities and rest which motivate – somewhat by chance – the avoidance of periods of heavy transport usage. In fact, the objective of avoiding the rush-hour does not seem possible to achieve except when it is ‘doubled’ by an interest and preference which go beyond the simple question of travel:

What do you consider the reasons determining your time of arrival at work?
- ‘I thought about it on the way to the interview, because I never asked myself the question.
There’s no one in the RER with me, whereas my husband, he likes to sleep, so he leaves a little later, but he has much more of a crown in the RER.’
Then it would be mainly to benefit from greater comfort in the RER that you arrive at 8:40 ?
- ‘It’s especially that I’m rather a morning person, or let’s say I’m very fond of long evenings, so of getting home early from work, and so of getting there early.’ (Léna, about fifty years-old, married with no children at home, living in the 78th
Département, arrival around 8:40 a.m.)

4.2 Scheduling preferences or respect for the social norms of working hours? The example of the morning coffee

52The observation leads us to the study of our second group of employees who freely chose their work schedule according to their preference. They did not state on the questionnaire that they sought to avoid the congestion, but simply that they decided according to their personal preferences.

53On the one hand, these employees, who have the profile of independent workers, are more likely (+12.4 %) to arrive early at work – before 8:00 a.m. The interviewees who ticked the box ‘personal preferences’ along with an early morning arrival again explain their schedule preference by the benefits gained from leaving work sooner. In addition, some explain the interest of an early morning arrival at work in terms of profiting from the calm at the office, especially in an ‘open space’ arrangement.

54On the other hand, and again by contrast with employees whose schedule is fixed by their employer, the employees choosing their schedule as a function of their personal preferences are also more likely (+4.9 %) to arrive during the rush-hour (between 9:00 a.m. and 9:29). Here again it is the interviews with our survey sample that permit us to decipher this apparently paradoxical preference for the rush-hour, generally so deplored and negatively connotated [Orfeuil, 2000; Ricroch 2013].

55During our interviews, certain individuals indicated that they preferred to arrive during the rush-hour to profit from moments of sociability at that time in their work-place. And it would seem that, in almost all the establishments investigated, these moments of sociability take place around the coffee machine:

Between 8:30 and 9 o’clock, everyone goes down to have a coffee, so there is no one left in the offices. (Nadia, fifty years-old, unmarried without children, executive, living in the 78th Département, arrival around 8:00 a.m.)

56Sophie talks to us about this preference; she says:

 I don’t like to take my coffee all alone. I remember, too, that when my best friend was on holiday, I delayed my time of arrival at work to arrive at the same time as my other friend. (Sophie, about thirty years-old, married with children, executive, living in the 91st Département, arrival around 8:00 a.m.)

57The preference for the morning coffee with colleagues would thus be one of the determinants of an time of arrival at work during the rush-hour. However, do these moments of synchronisation around the coffee machine close to 9:00 a.m. always reflect the individual preferences and the practices really desired by the employees? In certain situations, should they not also be read as a form of submission to what we term the ‘social norms of work hours’?

5. Everyone at work before 9:30 a.m.: the social norms of work hours

58In this final section, we propose to investigate the interest of the idea of ‘social norms of work hours’ as an explanatory factor in the choice of work schedule. The social norms of work hours are usually expressed silently, in the mirror of the glances of others [Mead, 1934] and the labels which everyone imagines that everyone else is imposing: what will my colleagues think, and my superiors, if I decide to arrive at the office at 10 o’clock? Or, better still: that one there gets in early; he seems professional and engaged. By considering mainly our reports of interviews, we attempt here to verify the existence of social norms of work hours as a factor limiting the possibilities of travelling to work after 9:30 a.m.

5.1 Two social norms of work hours

59Behind the social norms of work schedules lies a group of practical rules of behaviour which translates collective values into modes of conduct in the work-place. One thinks mainly of the fact that it might look bad – or that we imagine it looks bad – to arrive too late at the office, and that therefore a practical unwritten rule exists in business life whereby one should be in place before or at the same time as everyone else. Moreover, this rule may also take concrete form in certain cases as a requirement to be present to have coffee with the work-team.

60This is what Julie tells us. She lives at some distance and drops off her children, and she cannot be there until 9:45 a.m. She considers that her absence at the coffee hour is held against her:

For example, the fact of not being there for coffee at the start of the day makes me invisible, and that may limit advancement in a career. One may be evaluated as someone who works, but if you are never seen… (Julie, about thirty years-old, living with a partnership, with children, upper-level executive, living in Paris, arrival around 9:40 a.m.)

61Beyond the tangible sign of the importance of synchronisation at the moment of the morning coffee, we show here that symbolic representations influence the conduct of employees by inducing them to arrive before or during the rush-hour, while discouraging arrival after 9:30 a.m. However, these symbols are very often tacit. Whether they are conveyed by hierarchical superiors or the employees themselves, validated by social surveillance or a form of internalised ethic, they are never discussed within the work force.

62On the basis of our interviews, it is also notable that the time of arrival at work is a signal indicating the quality of the employee, that is, the quantity of work that he or she accomplishes: someone who arrives earlier is supposed to work a great deal; someone who arrives at the same time as everyone else is supposed to work normally; somewhat who arrives late is supposed to accomplish little. To put it otherwise, and with regard to representations of the work ethic [Thrift, 1990; Weber, 1904], an individual who arrives at work before the others is a devoted employee; one who arrives at the same time as all the others is disciplined; and one who arrives after them is a shirker. With reference to these temporal symbols, our interviews reveal the presence of two social norms of work hours (Table 5) which channel the hours of arrival before 9:30 a.m.

Table 5 – The two social norms of work hours

Period of arrival

Distinguishing mechanism

Effect on the group

Schedule norm for disciplined employees


Sanction for those individuals arriving after 9:30 a.m.

Employee cohesion

Schedule norm for devoted executive

Before rush-hour

Promotion for those individuals arriving before 8:30 a.m.

Competition between executives

Source : Auteur

5.2 Schedule norm of disciplined employees

63The first is the central and historic norm of employees: the ‘schedule norm of disciplined employees’, which enjoins individuals to arrive at work at the same time as everyone else, and especially not after 9:30 a.m. in our case. According to the double aspect of social norms, the standard schedule is, on the one hand, justified in a positive sense by the values pertaining to salaried employment, such as effort, discipline, and, more generally, the scientific organisation of work time. Internalisation of the norm of disciplined employees, if it is diffused among all the workers, has the effect of assuring the social coherence of the work force.

64On the other hand, the obverse of this norm is that the employees who do not respect the values of work-hour discipline and arrive after the 9:30 a.m. limit see themselves labelled as idlers. ‘Did you take your morning off?’ an office-mate will ask, on seeing his colleague arrive at 10 o’clock. In this way, late arrivals are potentially open to sanctions from the hierarchy or their colleagues, sanctions that may range from simple teasing to exclusion from the work unit:

I’ve always had comments made about me about my time of arrival but never about my time of leaving! No one is there to see when I leave in the evening. The people who’ve said that are not my superiors. The way colleagues look at me… I think that sometimes the looks of colleagues can have an influence, since certain colleagues might pass their complaints about my time of arrival up to the hierarchy. (Aude, about 50 years-old, divorced with children, mid-rank employee, living in Paris, arrival around 9:00 a.m.)

5.3 The work-hour norm of the devoted executive

65The second work-schedule norm, while less common in practice, is nevertheless probably already dominant in the more current representations of the values underpinning managerial status in France. In a transitional phase, this peripheral norm is clearly in competition with the ‘norm of disciplined employees’. This is the ‘work-hour norm of the devoted executive’. Contrary to the majority norm applied to the mass of employees, this norm does not have the consequence of sanctioning those who do not respect it, but rather of distinguishing and rewarding executives who show proof of dedication to their work:

In France we prefer to be more or less within the norm, to follow the crowd and not stand out too much. But this is true only in a sense. If you arrive at 8:00 a.m. and leave at 9:00 p.m., that’s not necessarily something that will be held against you – on the contrary. (Anne, about forty years-old, married with children, executive, living in the 91st Département, arrival around 9:00 a.m.)

66The rules of the game concerning this norm put pressure on executives, then, to engage in a sort of contest over the number of hours worked. The object is to put in the most work hours possible and potentially to arrive before everyone (prior to 8:30 a.m.) in order to make a good impression:

- Before working at […] I worked in a consulting and auditing firm, and there certainly was that sort of unhealthy competition: who is already here when the director arrives? And who stays longest? A pressure linked to the question of being efficient, productive, or not, and which was tied directly to the number of hours spent at work. You felt obliged, then, to come in early and leave late.
- What would have been the origin of that norm?
- For me, it’s quite clear. In the consulting firm where I was, there was a turnover of 25 % per year, and basically it was never stated, never put in words, but it was the spirit of competition. Because every year you were supposed to advance or leave! So necessarily everyone hung on to be one of those who would be continued or have his salary raised. (Jean, about thirty years-old, unmarried, executive, living in the 94th
Département, arrival around 9:30 a.m.)

6. Conclusion and discussion

67This article furnishes information on the reasons impelling employees on flexible hours to travel during the rush-hour.

68On the theoretical level, it furnishes information favouring an interdisciplinary approach and a grouping of explanatory models of work-schedule choices.

69First, with reference to the Geography of Time models, our results show that the majority of workers (70 % of those we surveyed), while in appearance free to chose their work hours, remain in fact limited in selecting their schedule by the need for coordination with their social environment at certain times in the day. Independently of the rule mandating flexible working hours, the ‘couple-related constraints of daily life continue to induce arrivals at work during the morning rush-hour (8:30 a.m.– 9:29 a.m.). This is why employees ‘coupling’ with their social environment around their children’s school hours, the first meeting of the working day and/or the beginning of the work-day of their partner have a much greater probability of arriving during the rush-hour. For this reason, and because in the course of a single day they seem to have to reconcile more couple-related constraints than men do, the working women in our survey have a greater probability than the men of travelling during the morning rush-hour.

70Secondly, and on the contrary, 30 % of the employees seem, for their part, actually to have real manoeuvring room in deciding on their hour for arriving at the office. Among these employees less constrained by interactions at work or within their families, 42.8 % (that is, 12.8 % of all those surveyed) stated that they sought primarily to avoid the period of congestion when deciding on their work schedule. Here, the trade-offs made and the preferences expressed by individuals lead us to question the hypotheses on which the econometric models of ‘Scheduling Preferences’ are founded.

71In staggering their working hours and avoiding the rush-hour, individuals do not perform a cost/benefit calculation of such a binary kind as is implied by those models. The issue is not a mere simple weighing of the benefits expected from the point of view of travelling (gaining time and comfort) against the disadvantages potentially expected from the professional point of view (sanction by the employer because of lateness). In fact, in the face of potential professional disadvantages, for the scale to tip on the side of avoidance of the rush-hour, the improvement in travel conditions expected must be ‘doubled’ by an expected improvement in the daily programme of activities. Our interviews even demonstrate that avoiding periods of congestion is primarily motivated by a search for improvement in the daily programme of activities, rather than by a search for comfort or a time-saving in transportation. For example, the persons who stated that they avoided the morning rush by advancing their schedules explained to us that they sought primarily to benefit from ‘long evenings’ back at home more than from better travel conditions on the way to work.

72Mid-way between avoiding congestion and improving the daily programme of activities, these motivations revealed for a choice of work schedule favour a combination of the models of the Geography of Time and of ‘Scheduling Preferences’. Yet to understand completely the motivations of a choice of work schedule, as well as the reasons for the persistence of the rush-hour phenomenon, this double theoretical combination cannot suffice.

73Thirdly, we reveal in this article that on the other side of the scale – this time behind the sanctions and professional penalties to which individual expose themselves when they deviate from standard schedules – lie hidden ‘social norms of work hours’. With reference to a theoretical universe closer to the sociology of time [Rosa, 2013] and of work [Bouffartigue, 2012], this concept makes it possible to explain the diversity of reasoning and logics that individuals appeal to when they decide on their work schedule. These motivations are revealed by the spoken responses of those surveyed who stated that they chose their schedule, not with a view to avoiding periods of congestion, but simply in response to their personal preferences (17.1 % of the sample). If a large majority of them arrive before or during the rush-hour, it is also because behind their personal preferences is concealed a submission to the social norms of work hours, themselves underpinned by adhesion to certain collective values. First, one finds the ‘social norm of disciplined employees’, whose main effect is to produce a form of social sanction with regard to those who would arrive at the office after the rush-hour. Secondly, and more as a peripheral norm, one finds the ‘social norm of the devoted executive’, which entails a form of reward for employees arriving at work early. One can surmise, then, all the interest accruing to an arrival because of the social norms and their associated values as a means of explaining the persistence of rush-hours and the choice of work schedules.

7. Perspectives

74Transport specialists or geographers of time who are interested in problems of transport saturation have not until now integrated the question of social norms in their understanding of the rush-hour phenomenon. This opens large perspectives for a research action to re-examine the weight of social norms regarding work schedules in the persistence of the rush-hour phenomenon.

75In Japan, moreover, a country in which public transit is very crowed during rush-hours, the governor of Tokyo, Uriko Koike, inaugurated, in the summer of 2017, a campaign aimed at employers and employees labelled ‘Jisa Biz’ (‘jisa’ means ‘staggered schedule’ and ‘biz’ is the abbreviation for English ‘business’). It advocates a ‘réforme des modes de travail’ (reform of work modes) providing for staggered office hours and distance-working to avoid jammed trains during the rush-hour. But, above all, it proliferates declarations in the media to explain that ‘le fait d’arriver au travail à l’heure ne signifie pas forcément qu’on travaille’ (arriving at work on time does not necessarily mean that you are working) [Kubota, 2018, p. 75] – just as arriving after the rush-hour does not mean that one is working less, or less well, than the others.

76It becomes more and more clear that, as long as the question of the social norms of work hours is not debated, this idea will not make headway, and the overloading of transport systems will continue to proliferate. These unspoken factors carry fundamental weight, for the norms often work against purely rational motivation. They can subsist only if their arbitrary aspect is not too strenuously called to attention: is it really necessary for all the inhabitants of the Paris region to be at their place of work before 9:30 a.m.?

Haut de page


ADAM Barbara, 2004, Time, Hoboken, Ed.Wiley, 192 p.

ADOUE François, 2016, La mobilité connectée au quotidien – Les usages du smartphone dans les transports en commun franciliens, Thèse de doctorat : Aménagement et Urbanisme, Université Paris-Est, 322 p.

BAUDRAZ Jean-François, 1971, L'horaire variable de travail, Vaud, Impr. Vaudoise, 135 p.

BOUFFARTIGUE Paul, 2012, Temps de travail et temps de vie. Les nouveaux visages de la disponibilité temporelle, Paris, PUF, 240 p.

BOULEAU Mireille, LEROI Pascale, 2016, « Nouveaux modes de travail et mobilité », Document de l’IAU Île-de-France, avril, 41 p.
[En ligne] URL :

BÖRJESSON Maria, 2008. “Joint RP/SP data in a mixed logit analysis of trip timing decisions”. Transport. Res. Part E : Logist. Transport. Rev. vol. 44, n° 6, pp. 1025–1038.
[En ligne] DOI :10.1016/j.tre.2007.11.001

BROUSSE Cécile, 2015, « La vie quotidienne en France depuis 1974. Les enseignements de l’enquête Emploi du temps », Économie et statistique, n° 478, pp. 79-117.
[En ligne] URL :

de CONINCK Frédéric, GUILLOT Caroline., 2007, « L’individualisation du rapport au temps, Marqueur d’une évolution sociale »,¿ Interrogations ?, n° 5.
[En ligne] URL :

DAVIES Kevin, 2001, “Responsibility and daily life: reflections over timespace”, MAY Jon. & THRIFT Nigel, Timespace: Geographies of Temporality, Abingdon-on-Thames, Routledge, pp. 133–148.

DREVON Guillaume, 2016, Mobilité quotidienne et stratégies d’adaptation spatio-temporelles des ménages : une approche comparée des espaces transfrontaliers et non frontaliers des métropoles luxembourgeoises et grenobloises. Thèse de doctorat : Géographie. Université Grenoble Alpes ; CEPS/INSTEAD (Luxembourg), 331p.
[En ligne] URL :

GODARD Francis, 2007, « Vie publique et vie privée : de nouveaux régimes temporels », Réseaux, n° 140, pp. 29-65.
[En ligne] DOI : 10.3917/res.140.0029

GROSSIN William, 1996, Pour une science des temps – Introduction à l’écologie temporelle, Toulouse, Octares, 168 p.

HÄGERSTRAND Torsten, 1970, » How about people in regional science?” Papers of the Regional Science Association, n° 24, pp. 7-21.

HE Sylvia, 2013, « Does flexitime affect choice of departure time for morning home-based commuting trips? Evidence from two regions in California », Transport Policy, n° 25, pp. 210–221.
[En ligne] DOI :

HINES Emery. J, 1984, « Estimating Users and Impacts of a Regional Alternative Work Schedule Program », Transportation Research Record, n° 845.
[En ligne] URL :

KONING Martin, 2012, Essais sur la congestion dans les transports à Paris, Thèse de doctorat en Economie, Université Paris 1, 250 p.
[En ligne] URL :

KUBOTA Masaki, 2018, “Le train-train d’enfer des Japonais”, Courrier International, Hors-Serie avril-mai-juin, pp. 75-79.

LAVIELLE Jean-Pascal, 2008, Évolutions des rythmes sociaux et étalement de l’heure de pointe, Travail de fin d’études, ENTPE.
[En ligne] URL :

LENNTORP Bo, 1977, « Paths in space-time environments: A time-geographic study of movement possibilities of individuals », Environment and Planning A, vol. 9, n° 8, pp. 961-972.

McCAFFERTY Desmond, HALL Fred L., 1982, « The use of multinomial logit analysis to model the choice of time to travel », Economic Geography, n° 58, pp. 236–246.
[En ligne] URL :;jsessionid=0E6791B21E2DB4D760A4B0CA7F?sequence=2

McNALLY Michael G., 2000, “The activity-based approach”, HENSHER David A. & BUTTON Kenneth J., Handbook of transport modelling, Oxford, Pergamon, pp. 53-70.

MEAD George Herbert, 2015, [1934], Mind, Self and Society, Chicago, University of Chicago Press, 536 p.

MUNCH Emmanuel, 2017a, Mais pourquoi arrivent-ils tous à la même heure ? Le paradoxe de l’heure de pointe et des horaires de travail flexibles : enquête sociologique auprès de cadres franciliens. Thèse de doctorat : Aménagement de l'espace, Urbanisme, Université Paris Est, 455 p.
[En ligne] URL :

MUNCH Emmanuel, 2017b, « Les périodes de fréquentation des transports collectifs urbains. Quels enjeux pour la planification ? », ZEMBRI-MARY Geneviève, Concevoir une offre de transport public, Territorial Editions, Dossiers d’experts, n° 813.

MUNCH Emmanuel, PROULHAC Laurent, 2019, « Le paradoxe de l’heure de pointe et des horaires de travail flexibles », Territoire en Mouvement, n° 2019-2.

ORFEUIL Jean-Pierre, 2000, L’évolution de la mobilité quotidienne. Comprendre les dynamiques, éclairer les controverses (Synthèse Inrets, n° 37), Cachan, Ed. Lavoisier, 146 p.

OTT Michael, SLAVIN Harvey, WARD Daniel, 1980, « Behavioral impacts of flexible working hours », Transportation Research Record, Journal of the Transportation Research Board, n° 767, p. 1–6.
[En ligne]URL :

de PALMA André, DUNKERLEY Fay, PROOST Stef, 2009, “Trip chaining, who win, who loose?”, Journal of Economics and Management Science, vol. 19, n° 1, pp. 223-258.
[En ligne] DOI:

PRITCHARD Evans, 1980, [1940], The Nuer: A Description of the Modes of Livelihood and Political Institutions of a Nilotic People, Oxford, Nabu Press, 341 p.

RICROCH Layla, 2011, « Les moments agréables de la vie quotidienne », INSEE première, vol. 1378, n° 4.

ROSA Hartmut, 2013, Social Acceleration: A New Theory of Modernity, Columbia University Press, 512 p.

SALEH Wafaa, FARRELL Séona, 2005, « Implications of congestion charging for departure time choice: Work and non-work schedule flexibility », Transportation Research Part A, vol. 39, n° 7-9, p. 773–791.
[En ligne] URL :

SCHINTLER Laurie A., 2001, « Women and Travel », HENSHER David A., Handbook of Transport Systems and Traffic Control (Handbooks in Tranport, Volume 3), Oxford, Pergamon, p. 351-58.

SCHWANEN Tim, 2007, “Gender Differences in Chauffeuring Children among Dual-Earner Families”, The professional Geographer, vol. 59, p. 447–462.
[En ligne] DOI:

SCHWANEN Tim, 2008, “Struggling with Time: Investigating Coupling Constraints”, Transport Reviews, vol. 28, n° 3, p. 337-356,
[En ligne] DOI:

SMALL Kenneth A., 1982, « The scheduling of consumer activities: Work trips ». American Economic Review, vol. 72, n° 3, p. 467–479.
[En ligne]URL:

THOEMMES Jens, 2012, « La fabrique des normes temporelles au travail », La nouvelle revue du travail, n° 1.
[En ligne] URL :]

THOEMMES Jens, 2013, « L'histoire oubliée des horaires individualisés : de la désynchronisation choisie à la flexibilité pour l'entreprise », Revue Française de Socio-Économie, n° 11, p. 35-53.
[En ligne] DOI: 10.3917/rfse.011.0035

THOMPSON Edward P., 1967, “Time, Work-Discipline and Industrial Capitalism”, Past & Present, n° 38, p. 56-97. [En ligne] URL:

THORHAUGE Mikkel, CHERCHI Elisabetta, RICH Jeppe, 2016, « How flexible is flexible? Accounting for the effect of rescheduling possibilities in choice of departure time for work trips », Transportation Research Part A, vol. 86, p. 177–193.
[En ligne] DOI:

THRIFT Nigel, 1990, “The making of a capitalist time consciousness”, HASSARD John, The Sociology of Time, New York, St. Martin’s Press, p. 105-129.
[En ligne] URL:

TOFFLER Alvin, 1971, Future Shock, New York, Bantam Books, 540 p.

TRIBOUT Jacques, 2014, « Les déplacements ferrés lourds en Île-de-France – Une analyse de l’EGT 2010 », Document interne SNCF Transilien, 97 p.

VICKREY William S., 1969, “Congestion theory and transport investment”, American Economic Review, Papers and Proceedings n °59, p. 251-260.
[En ligne] URL :

WEBER Max, 1994, [1904], L’éthique protestante et l’esprit du Capitalisme, Paris, Pocket, 285 p.

WEBER Max, 1971, [1922], Économie et Société Tome 2 - L’organisation et les puissances de la société dans leur rapport avec l’économie, Paris, Pocket, 410 p.

YANG, Ying, STEINER, Ruth L., et SRINIVASAN, Sivaramakrishnan, 2016, “The Impact of Flexible Work Hours on Trip Departure Time Choices in Metropolitan Miami”., In: 16th COTA International Conference of Transportation ProfessionalsTransportation Research BoardInstitute of Transportation Engineers (ITE) American Society of Civil Engineers.

YOSHIMURA Mitsunori & OKUMURA Makoto, 2001, « Optimal commuting and work start time distribution under flexible work hours system on motor commuting », Eastern Asia Society for Transportation Studies (EASTS), n° 10.
[En ligne] URL :

ZERUBAVEL Eviatar, 1981, Hidden Rythms, University of California Press, 224 p.

Haut de page


1 The regional transportation authority which manages the RER (Réseau Express Régional) is also the organism financing this research, since the research was part of a doctoral project involving a Convention Industrielle de Formation par la Recherche (CIFRE) established with SNCF Transilien.

2 Number of usable responses after verification of the database. The questionnaire, to be completed anonymously online, was transmitted to the employees by each enterprise. The enterprises were able to choose a distribution period of two weeks between 7 April 2015 and 22 May 2015. At the beginning of the period chosen, the senior management of the participating organisations sent an invitation, in the name of the enterprise, to respond to the questionnaire to all of the workers (both permanent and temporary) possessing an e-mail address. Of the total of 12,038 employees contacted, 3,358 responded to the whole of the questionnaire.

3 5 persons presented a profile too indistinct to permit them to be identified in the questionnaire database.

4 Twelve reasons identified as well as one response ‘Others’; The respondents selected on average somewhat less than three reasons.

5 Public transport

6 Personal vehicle

Haut de page

Table des illustrations

Titre Map 1 – Localisation of the enterprises surveyed
Crédits Source: BDD GEO SIRENE. List of enterprises updated as of 1 January 2017. BD ORTHO de l’IGN. Photo 2011. Created by E. Munch and C. Pivano. Note : Filtered by establishments with a minimum of 50 employees.
Fichier image/jpeg, 992k
Titre Figure 2 – Distribution of exit times at the Stade-de-France Saint-Denis station, by 30-minute periods
Légende Reading: At the Stade-de-France Saint-Denis station, 22.6 % of the exits effected during the morning (6 a.m.-11 a.m.) on business-day Tuesdays in 2015 took place between 9:00 a.m. and 9:29 a.m.
Crédits Source: Station exit data for all the business-day Tuesdays in 2015. Centre TVV–SNCF Transilien.
Fichier image/png, 18k
Haut de page

Pour citer cet article

Référence électronique

Emmanuel Munch, « For a temporal ecology of rush-hours: a survey of choices of work hour in the Paris region »Espace populations sociétés [En ligne], Hors-série | 2021, mis en ligne le 06 novembre 2021, consulté le 30 novembre 2021. URL : ; DOI :

Haut de page


Emmanuel Munch

Laboratoire Ville Mobilité Transport UMR T9403 ENPC-IFSTTAR UPEM
6-8 Avenue Blaise Pascal, 77455 Marne-la-Vallée cedex 2.

Articles du même auteur

Haut de page

Droits d’auteur

Licence Creative Commons
Espace Populations Sociétés est mis à disposition selon les termes de la licence Creative Commons Attribution - Pas d'Utilisation Commerciale - Pas de Modification 4.0 International.

Haut de page
  • Logo Université de Lille
  • Logo Laboratoire TVES
  • Logo Laboratoire Clersé
  • DOAJ - Directory of Open Access Journals
  • Revue soutenue par l’Institut des sciences humaines et sociales du CNRS
    CNRS - Institut national des sciences humaines et sociales
  • OpenEdition Journals
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search