the journey to work: a century of change

The Journey to Work: A Century of ChangeAuthor(s): Colin G. Pooley and Jean TurnbullSource: Area, Vol. 31, No. 3 (Sep., 1999), pp. 281-292Published by: The Royal Geographical Society (with the Institute of British Geographers)Stable URL: http://www.jstor.org/stable/20003993 .

Accessed: 16/06/2014 21:35

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

.

The Royal Geographical Society (with the Institute of British Geographers) is collaborating with JSTOR todigitize, preserve and extend access to Area.

http://www.jstor.org

This content downloaded from 185.2.32.49 on Mon, 16 Jun 2014 21:35:34 PMAll use subject to JSTOR Terms and Conditions

http://www.jstor.org/action/showPublisher?publisherCode=rgs

http://www.jstor.org/stable/20003993?origin=JSTOR-pdf

http://www.jstor.org/page/info/about/policies/terms.jsp


Area (1 999) 31.3, 281-292

The journey to work: a century of change

Colin G Pooley and Jean Turnbull Department of Geography, Lancaster University, Lancaster LA1 4YB. Email: [email protected]

Revised manuscript received 1 1 December 1998.

Summary The paper uses original quantitative and qualitative evidence to examine

changes in the distance travelled and time spent on the journey to work in twentieth

century Britain. Emphasis is placed on variations by gender and location, and on the

implications of modal change. We suggest that analysis of past commuting trends may have

implications for contemporary transport policy.

The significance of the journey to work

The journey to work is a process that has economic,

social, cultural and environmental significance, both for individuals and for society in general (Pickup 1983; European Foundation 1987; OECD 1995;

Gibbs 1997). For instance, the need to travel to work imposes costs in terms of time and money, which limit household budgets and restrict employ

ment opportunities. Long hours spent commuting

affect the time available for household and family responsibilities, and have implications for personal

relationships and childcare. For some, the journey

itself can take on cultural significance, forming an important part of an individual's life and generating a community based around fellow travellers; meanwhile, the environmental costs of commuting by private car are evidenced in high levels of congestion, pollution and damage to human health

from vehicle exhaust emissions. These issues have received much attention from contemporary academics, planners and policy-makers alike (Daniels 1980; Daniels and Warnes 1983; Breheny and

Rockwood 1993; DoT 1993; Banister 1994; Whitelegg 1997; DETR 1998a), but in the late

twentieth century, a long journey to work, with all its

associated problems, has become an inevitable con

sequence of modern lifestyle decisions, particularly the process of counterurbanization, which is seen

most clearly in a British context in south-east England (Champion 1989; McGhie 1997).

Although these general trends, and their associ ated significance, can be demonstrated for the 1990s, we know little about how lifestyle decisions

affecting commuting changed over time during the twentieth century. For instance, we know that in

the mid-1990s the average commuting distance in

Britain is approximately 145 kilometres, and that those commuting into Central London travel further and for longer than people who live elsewhere. We also know that men travel further to and from work than women, and that men are more likely to use

motor vehicles for commuting (1991 Census; see

also DoT 1994-96; Office of National Statistics 1998; DETR 1998b). However, we do not know

when such trends developed, or their stability over time. Although we also know that, in general, commuting distances have increased during the

twentieth century, there is no directly comparable quantitative evidence for the century since 1890 (Lawton 1963; 1968; Warnes 1972), and there are scant data on how other aspects of commuting have changed (Liepmann 1944; Westergaard 1957; Daniels 1970; Warnes 1975; Davies and Musson 1978; National Travel Surveys). Some further evi dence comes from mid-twentieth century planning reports and related studies, which focus on the growing problems of traffic in towns (Barlow Report 1940; Sneed 1961; Ministry of Transport 1963), but there has been no systematic attempt to chart

changes in the journey to work over the twentieth century.

This paper uses original quantitative and quali tative data from the twentieth century to examine

changes in the distance and travel time of commut ing journeys in Britain, major changes in the main

mode of transport used, and variations in commuting

patterns by gender and geographical location. We

ISSN 0004-0894 ? Royal Geographical Society (with The Institute of British Geographers) 1999



282 Pooley and Turnbull

suggest that a better understanding of how behav iour has changed in the past may aid the solution of contemporary problems associated with traffic in towns.

Methods The main challenges facing this project were to collect a large enough sample of data relating to people who entered the labour force at different times during the twentieth century to make some quantitative statements about changing commuting patterns, and also to record sufficient detail to reveal qualitative aspects of the decision-making process and its relation to individual circumstances. These twin aims have been met firstly by collecting infor

mation on journey-to-work life histories for 1834 individuals who began work after 1890, recording details of a total of 12 439 separate journeys on standard (but detailed) recording schedules, and secondly by conducting in-depth semi-structured interviews with 90 respondents to explore the structure and context of journey decisions. There is no simple and unbiased way to identify a sample of respondents for a study such as this; our intention

was to collect data on approximately equal numbers of people who began work in each decade from the 1890s to the 1980s, and to construct as represen tative a sample as possible. As few people who entered the labour market before about 1920 are still alive, two separate data collection strategies were adopted, as follows.

Data on those who began work in the late nineteenth and early-twentieth centuries were col lected through a network of contacts with family historians, which had been established in a previous research project on residential migration (Pooley and Turnbull 1998). Requests for information were placed in the magazines of family history and genealogy societies in all parts of the country, and respondents from the previous project were con tacted directly. Family historians were asked if they were able to provide detailed data on their ancestors' (usually their parents') commute. As we were asking for material from the recent past, it was assumed that there would be a heavy reliance on personal memory and family papers rather than archival research. Respondents who began work

more recently (from the 1920s onwards) were con tacted in a variety of ways. Family historians were also asked to provide data on themselves and living relatives; a number of large employers were con

tacted; advertisements or letters were placed in a number of local newspapers and other publications; and some pensioners' clubs were approached. Companies were selected on the basis that they employed a wide range of workers (from manual staff to professionals), and included local authorities, institutes of higher education and large private com panies. Our request for respondents was publicized in a variety of ways: some firms attached a letter from us to every payslip, others published a request for help in an internal newsletter, whilst some placed posters on staff noticeboards. In each case, people who were prepared to participate in the project were asked to contact us directly. We collected informa tion on people who worked in any location in Britain, but targeted three large labour markets in particular (London, Manchester and Glasgow), which were also to form the focus of the qualitative study.

Family historians provided by far the best response to our call for data, contributing 77-4 per cent of all returned forms. Public-sector employees and con tacts made through public libraries and old peoples' homes provided the majority of the remaining data, but the response from newspaper advertisements was very poor. Everyone who expressed an interest in the project was sent a detailed data entry form, divided into four sections. These forms requested information on a person's employment history (including periods out of the workforce) from the time they began work to retirement (where relevant); residential history; details of every new commute, including mode of transport, time taken, cost and reasons why a particular form of transport was chosen; and personal characteristics. Respondents completed a form either for themselves or (in the case of some family historians) for a close relative. Those completing forms for themselves were asked if they were prepared to participate in an in-depth interview at a later date. Almost half of all people

who were sent a form completed and returned usable information.

A data-collection procedure such as this, which relies on a self-selected sample responding to an open request for information, creates many problems and potential biases. It is not possible to use such data to relate journey-to-work character istics to specific labour markets as, in most areas, the number of responses was too small. For analysis,

we have divided respondents into those working in London, other towns with a population of over 100 000 (in 1951) and other places; there is a fairly even split between these three categories. We also



The journey to work: a century of change 283

Table 1 Characteristics of the dataset

Decade Sample size*

Male Female Total

No No No %t

1890-99 77 819 17 18-1 94 5-1 1900-09 101 73 7 36 26-3 137 75 1910-19 108 65-9 56 34-1 164 8-9 1920-29 94 60-6 61 39-4 1 55 8-5 1930-39 166 58-9 116 41-1 282 15-4 1940-49 193 52-2 1 77 47-8 370 20-2 1950-59 100 41-3 142 58-7 242 13-2 1960-69 57 39-0 89 61-0 146 8-0 1970-79 63 46-0 74 54 0 137 7-4 1980-98 51 47-7 56 52 3 107 5 8

Total 1010 55-1 824 44-9 1834 100-0

Notes: *Number of males and females in the sample who started work in each decade

**Percentage of total starting work in that decade

tPercentage of total sample

failed to get an even response between decades, with least information for the 1 890s and most for the 1 930s and 1 940s. The relative lack of data on people

who are no longer alive is understandable; slightly more surprising is the low response from people who began work in the 1970s and 1980s. Older people were much more likely to respond than those

who began work more recently (Table 1). The char acteristics of all those in our sample who were in the

workforce in 1931, 1951 and 1971 were compared with the relevant census workforce characteristics (Table 2). The sample is quite representative of the total population with respect to gender, has some under-representation of the young and single, and contains a substantial bias towards those in higher socio-economic groups. It is almost inevitable that a self-selected sample will be biased towards those

with more education and in higher-status occupa tions. It can be suggested that family historians (our

main providers of information) are themselves most likely to come from middle-class and professional backgrounds, and that white-collar employees are most likely to receive and respond to calls for information disseminated within a company. It can also be suggested that there will be other demo graphic biases in data provided about the ancestors of family historians-in particular, there will be a bias towards those who married and had children. In

addition, all research that collects life-history data from individuals will be affected by errors of recall and possible misrepresentation.

All the biases and omissions noted above are inevitable in a project such as this, and must be taken into account in interpretation. We do not claim that the dataset is totally representative of the population as a whole, and the data are not being used for inferential purposes. However, we argue that there are sufficiently large numbers of responses for most decades to enable secure conclusions to be drawn. The fact that journey distances and transport modes recorded in this project for the 1 990s are very similar to those found in recent surveys (Office for National Statistics 1998) suggests that the biases have not unduly distorted the picture of commuting. In addition, the second stage of the research (in-depth interviews) was used to try to correct some of the biases. Those interviewed all began work in London,

Manchester or Glasgow, and in this way we could relate the detailed accounts to specific labour

market characteristics. In selecting those to be interviewed, we also over-sampled respondents in lower-status occupations to ensure that the second stage sample was more broadly representative of the population as a whole.

Moreover, there were some inevitable defi nitional and analytical decisions that affected the




Table 2 Comparison of sample and census population characteristics* (%)

Characteristic 1931 1951 1971

Census Sample Census Sample Census Sample

Gender Male 703 81 4 69-2 72 8 63-3 65 6 Female 297 186 30-8 272 367 344

Age 14-24 years 330 240 232 191 210 46 25-34 years 230 27-3 218 28-9 194 114 35-54 years 32-8 45-7 42-7 37 6 41 8 68 8 55-64 years 108 3 0 12 3 144 1 78 15 2

Marital status Single 48 7 360 34-3 34-1 23 3 15 6 Married 46 8 61 6 61 2 63-9 66 5 80-3 Widowed 4 5 2-4 4 5 2-0 10-2 41

Socio-economic status Professional 2-2 174 3-1 21 2 3-6 41 0 Intermediate 12-9 130 13 7 156 178 170 Skilled 48-9 47 5 52-6 46-2 49 5 34-7 Semi-skilled 18-2 11 6 160 9-1 209 3-5 Unskilled 1 7-8 10 5 14 6 7 9 8 2 3-8

Sample size 339 607 665

Note: *Comparison of the employed sample population alive in 1931, 1951 and 1971 with the employed population in the relevant census for England, Wales and Scotland

interpretation of results. In this paper, analysis is based on all those instances where a specific journey could be identified and measured. Thus, people who worked from home are excluded from analysis, as are those whose place of work varied from day to day (though information exists on both categories of workers). Analysis of transport mode is based on the main mode of transport used, although again, we have data on multi-mode journeys. Due to the characteristics of the dataset, there will be additional biases when the data are divided into time periods. Thus, in the 1890s, we only have data on people who entered the labour market in that decade, whereas in later decades we have data on people in different stages of their life course. This is potentially important, as journey characteristics vary over the life course. To take this discrepancy into account, data

were analysed both by decade and by cohort (com paring groups that entered the labour market at the same time). In this paper, results are presented by

time periods, as analysis by cohort does not make a significant difference to results. In this paper, too,

most emphasis is placed on quantitative data, with only limited use of qualitative information.

Distance and time

It is usually assumed that during the twentieth century the journey to work has become longer both in time and distance. Our data suggest that this is

only partly true. Over the century since 1890, there has been a four-fold increase in the mean one-way journey, from 3-6 kilometres to 14 6 kilometres, with distances increasing steadily over the decades. How ever, by contrast, the length of time spent travelling to work has barely doubled over the past hundred years, from 1 77 minutes to 34-5 minutes, and most of this increase occurred between 1890 and the 1920s; the average length of time spent travelling has increased little since the 1 930s (Table 3). These




Table 3 Average distance (kms), time (mins) and speed (km/h) of journeys since 1890, by gender

Decade Male Female All

Distance Time Speed Distance Time Speed Distance Time Speed

1890-99 4-0 17 0 14-1 1-8 21-3 5-1 3-6 17-7 12-2

1900-09 3-9 21-5 10-9 3-2 25-4 7-6 3-8 22-4 10-2

1910-19 6-2 27-0 13-8 51 26-8 11-4 5 9 27-0 13-1 1920-29 6-8 28-2 14-5 6-1 31 3 11-7 6-7 29-0 13-9

1930-39 7-0 305 13-8 6-8 31 9 12-8 70 309 13-6 1940-49 8-2 33-8 14-6 7-3 33 1 13-2 7-8 33-5 14-0

1950-59 10 1 33-6 18-0 7-4 34-4 12-9 9-0 33 9 15-9 1960-69 12-1 34-6 21-0 7-5 32-1 140 10-2 33-5 18-3 1970-79 13-1 34-5 22-8 7-6 28-5 16-0 10-3 31-5 19-6

1980-89 15-5 37-3 24-9 8-8 29-4 18-0 12-0 33-1 21-8

1990-98 19-4 39-1 29-8 105 30 7 20-5 14-6 34-5 25-4

Source: Details of 12 439 journeys, taken from 1834 individual life histories. Statistics relate to all modes of transport and

are calculated for the decade in which a particular journey started

trends are further emphasized if frequency distri butions are examined. Whereas the modal distance travelled one way to work was less than 2 kilometres in 1890-1929, rising to 2-4-9 kilometres in 1930 59, increasing again to 5-9 9 kilometres in 1960-89, and reaching 10-19-9 kilometres in 1990-98, the modal time taken to travel has remained at 10-29 minutes, from the 1 890s to the 1 990s. Between 1890 and the 1920s, there was a decline in the proportion of commuters travelling for under 10 minutes, and an increase for over one hour, but from the 1 930s onwards these proportions also remained relatively unchanged. We suggest that most people appear to have a threshold of time up until which they are prepared to travel, that this threshold has remained relatively constant over a long period of time, but that the distance travelled in this time has increased as transport modes have changed and

more people have gained access to faster forms of transport. These conclusions are consistent with research elsewhere (Pucher and Lefevre 1996; Schafer and Victor 1997).

Such attitudes to the time spent commuting were evident from some of the respondents interviewed for the research. A married man travelling by car in

Manchester in the 1 980s commented on the variability of the length of his journey:

On a good day, I should think about 25 minutes to half

an hour. On a bad day, an hour and a half. And I didn't

like that. A waste of time. I was too busy, too many things to do. (R24)

Forty years earlier, in the 1 940s, a single man working in London was quite happy to travel by bus or bicycle for over half an hour:

Oh, it was as quick as the bus really. There wasn't much traffic on the road ... it used to take about three quarters of an hour, I suppose, on a bicycle. (R26)

There are well-known gender differentials in contem porary travel patterns, with women having, on aver age, shorter journeys to work than men. Our data suggest that these differentials are of long standing, but that the implications are complex. Both travel times and distances for men and women have followed the same general trend-with distances increasing and times remaining relatively static-but

whereas women have consistently travelled shorter distances than men, their time spent travelling was at least as long as that for men, and in some decades longer (Table 3). Distance differentials were particu larly large in the 1890s, then narrowed until the 1930s, but then widened substantially so that from the 1 970s men were travelling almost twice as far as

women. With respect to time travelled, women had longer mean journeys in most decades before the 1960s, with men travelling for substantially longer times only from the 1980s. These differences are




clearly seen when the average journey speeds of men and women are compared (Table 3). Men had a faster mean journey speed in all decades, with the greatest differentials before the 1920s and after the 1 950s. Women's travel speeds only approached those of men in the 1 920s, 1 930s and 1 940s, a period when women were entering a wider range of jobs, but before the widespread use of the motor car for commuting. The same trends are apparent in all labour markets, suggesting that such gender vari ations are a strongly embedded aspect of journey patterns. Given that women were most likely to be employed part-time (85 7 per cent of all journeys by part-time workers were undertaken by women) and also to have additional childcare and household demands on their time, the fact that they consistently travelled by slower forms of transport than men must have had a significant impact upon their lifestyles and employment opportunities (Hanson and Pratt 1991; Masser etal 1992; Orfeuil and Solomon 1993;

McDowell 1997). Women's dependence on relatively slower

forms of transport was also commented on by a number of respondents. A married woman working in Harrow on the outskirts of London in the 1950s commented:

If you've got your own transport, which we hadn't in those days, it was okay, but you know it was very difficult to get there. You're sort of waiting for buses and of course, because it was sort of out of London, they didn't come very frequently. It was a nightmarish time but when you're younger, you know, you just do the best you can. (R29)

Another single woman working in Enfield (also on the outskirts of London) in the 1960s also com

mented on the slowness and inconvenience of the buses:

Well, I found that travelling into Enfield was pretty well hopeless on public transport. There's a lot of schools en route, the children were always on the bus or at my stop, and I got very frustrated that I couldn't get on a

bus ... leaving home earlier and earlier. (R30)

Contemporary statistics suggest that journey patterns into and out of London are quite different from those of other places. Our data confirm that this is a well-established trend, with those working in London travelling both further and for longer in all decades. In each of the five time periods used in this analysis (to avoid small numbers in some decades), London

commuters travelled almost double the distance of those employed in towns of under 100 000 people,

with the greatest differential occurring in the 1920s and 1930s. In terms of time travelled, London com

muters travelled for approximately half as long again as workers in towns of under 100 000 in the period 1890-1919, and for approximately twice as long in all other time periods (Table 4). It is also instructive to examine the ratio of travel distance to time in settlements of different size. The average speed of the journey (by all modes) has increased steadily in each time period and in all locations (Table 4).

Moreover, from the 1 890s to the 1 970s, mean travel speeds were greater in London than in other places; yet this differential has gradually been closing over time, and in the period after 1980, settlements of under 100 000 recorded faster travel speeds than London (which, in turn, had faster commuting flows than other big cities). This suggests that, although London's transport system has become slightly less effective relative to other towns over the century, improvements in transport technology and traffic

management, combined with wider access to faster forms of transport, have maintained and increased the average speed at which a commuter has travelled (by all modes of transport).

Respondents who worked in London emphasized both the longer commute, and the way in which individuals adjusted their threshold of travel tolerance to cope with an average journey of almost an hour, rather than 30 minutes or less in other places. Thus, one married man who worked in London in the 1 960s said of his journey by train and underground:

I found out that wherever you go in London it nearly always seems to take an hour. It seems to be a standard

travelling time, so an hour didn't worry me, no. If it had been an hour and a half or something I might have

thought more about it. (R28)

Another man working in London in the 1940s also adjusted to his extended journey by bus and train:

I must say, I just regarded it as a routine that didn't

worry me very much. If there'd been an easier way I'm

sure I would have found it, but I didn't, like, as I say,

really mind it very much. In fact it didn't at that time

occur to me that an hour's wasted time was, perhaps

'cause I didn't waste all the time . .. 'cause I was doing

part-time study, I used to do a lot of reading on the

trains and buses. (R26)




Table 4 Average distance (kms), time (mins) and speed (km/h) of journeys since 1890, by location of workplace

Workplace Time period

1890-1919 1920-39 1940-59 1960-79 1980-98

London Distance 68 111 143 180 205 Time 290 43 3 50 7 522 51 5 Speed 141 154 169 207 239

Other cities > 100 000 population

Distance 4 3 5 6 6-5 8-3 10-2 Time 25-3 27-4 28-8 29-3 30 3 Speed 102 123 135 170 202

Towns < 100 000 population

Distance 3 7 44 64 79 109 Time 202 21 7 264 251 261 Speed 110 122 146 189 251

All locations Distance 50 68 8 5 102 12 8 Time 247 303 33 7 326 336 Speed 122 135 15-1 188 229

Source: Details of 12 439 journeys, taken from 1834 individual life histories. Statistics relate to all modes of transport and are calculated for the decade in which a particular journey started

Changing transport modes

Many of the trends outlined above can be explained by changing transport technology and differential access to particular forms of transport. It is well known that the twentieth century has seen a major shift away from public transport and towards the private motor car, and that there are contemporary differences in levels of car use amongst men and

women (DoT 1981; Dasgupta et al 1990; Jansen 1993; White 1995; Beatty and Haywood 1997).

However, we have few data on precisely when these changes occurred or on the stability of such trends over time. There appear to have been two major shifts, occurring in the 1 930s and the 1 960s, in the

main modes of transport used for commuting during the twentieth century (Table 5). From 1890 to the 1930s, walking was the most common experience,

with more than 40 per cent of those in employment walking before 1920. Of those using transport, jour neys were quite evenly divided between trams/ trolley buses, trains and buses, with bicycle use increasing rapidly from about 1910. From the 1 930s to the 1950s, the incidence of walking declined

rapidly, whilst the use of buses and bicycles increased substantially. Train use remained quite stable, and trams and trolley buses gradually disap peared. From the 1960s, the use of the motor car became dominant, with more than 40 per cent of those in employment commuting by car from the 1970s. Walking continued to decline in importance and travelling by bicycle and bus also declined sharply, though train and underground use remained almost unchanged, accounting for about one-fifth of

journeys. There were marked and consistent differences in

the main modes of transport used in different locations (Table 6). Walking, cycling and car use have always been at a very low level in London (though cycling has become relatively more import ant since the 1980s), with most people commuting in and out of London by underground and mainline trains and, especially before the 1 940s, by bus.

Trams and trolley buses were available, and used most extensively, in other cities of over 100000 population, with walking, cycling and (after 1960) car use most common in towns of under 100 000. Use of the bicycle between about 1920 and 1950 was




Table 5 Main mode of transport for journeys since 1890 (%)

Decade Transport mode

Walking Bicycle Tram/trolley bus Bus Train (overground) Underground Motorcycle Car/van

1890-99 594 2-0 16-8 5 0 9-8 5 0 00 00 1900-09 49 4 11-2 11-6 14-6 10-2 0-4 0-0 1-1 1910-19 40-6 13-3 16-0 9 9 15-4 1-9 0-6 1-9 1920-29 28-5 17-5 10-6 15-3 17-8 2-3 3-9 5-2 1930-39 22-5 19-1 9 7 13-8 18-4 4-1 2-3 9-1 1940-49 17-2 19-6 6-7 23-0 18-3 5-4 2-2 6-0

1950-59 13-4 16-0 2-5 23-3 18-9 4-4 3-0 16-3 1960-69 14-0 5-2 0-2 18-8 16-2 5-3 2 6 35-8 1970-79 13-4 4-5 0-1 15-8 13-2 5-5 1-9 44-5 1980-89 10-3 6-1 0.0 11-7 15-4 5-4 1-8 48-5

1990-98 7-9 61 0-2 10-1 1 7-2 4-5 0-6 52-8

Sample size 2083 1379 466 2073 2002 564 264 3108

Source: Details of 12 439 journeys, taken from 1834 individual life histories. Statistics are calculated for the decade in

which a particular journey started

particularly notable in smaller settlements, and was the single most important means of commuting in such towns in the 1 940s. Such trends obviously reflect the availability of particular forms of transport (with trains, trams and trolley buses more commonly available in larger cities), the morphology of particu lar urban areas, and the choices of individual com

muters who, for instance, found walking and cycling more conducive in smaller places. However, the combination of limited transport infrastructure and lack of traffic congestion meant that it was also in smaller towns that the switch from walking and cycling before 1960 to use of the motor car after 1960 was most dramatic. Thus, in the period 1940 59, only 14 7 per cent of commuting journeys were

made by car in settlements of under 100 000 popu lation; yet this figure increased to 50 8 per cent in the following 20 years. By and large, those (mostly larger) places with better-developed public transport systems have had consistently lower rates of car use for commuting journeys.

There were both similarities and differences in the main forms of transport used by men and women to commute, but with a consistent trend for women to be more dependent on walking and slower forms of public transport such as buses, trams and trolley buses. Before the rise of the motor car, men were much more likely to commute by bicycle and train (Table 7). However, it should be stressed that, before

1930, when transport choices were more limited, walking was the single most important means of transport for both men and women. During the 1930s and 1940s, cycling was the most important form of transport for men, with women increasingly travelling by bus and (in London) underground train, especially in the 1 940s, 1950s and 1 960s. Commut ing by car became the most important form of transport for men in the 1 950s, but this was not the case for women until the 1970s-though from the 1980s the proportion of men and women commut ing by car has been broadly similar. Overall, men have tended to use faster, more independent and private forms of transport (bicycles and cars),

whereas women have used slower and more public forms of transport (walking, buses and trams). The greater similarity of travel times and distances for men and women in the 1930s to 1950s (noted above) can partly be explained by the fact that, in this period, many men travelled by bicycle, whilst

women tended to travel by bus. Men thus had more independence and flexibility in their daily transport, but travel speeds were similar. Male attitudes to private transport, particularly the company car, were summed up by one respondent who worked in outer London in the 1980s:

I mean, once you've got a car you've got complete freedom, especially if they're paying for it.... I mean, if




Table 6 Main mode of transport for journeys since 1890, by location of workplace (%)

Mode Time period

1890-1919 1920-39 1940-59 1960-79 1980-98

London Walk 287 115 6-6 47 34 Bicycle 52 40 56 15 58 Tram/trolley bus 14 9 9 5 3-7 0-1 0.0 Bus 20 1 20 6 16 2 10 2 12 5 Train (overground) 242 389 44-7 45-7 41 8 Underground 48 99 178 23-4 21 8 Motorcycle 00 1 2 1-3 1 6 17 Car/van 0 7 42 3-5 12 5 12 7

Sample size 289 506 954 792 464

Other cities > 100 000 population

Walk 466 256 134 123 122 Bicycle 112 196 182 52 53 Tram/trolley bus 26 1 18-4 8 1 02 0.1 Bus 75 116 312 276 154 Train (overground) 7-5 13 1 106 76 12 7 Underground 00 1 1 02 0 7 04 Motorcycle 04 2 9 2 8 2 5 09 Car/van 0 7 62 142 42 9 52 1

Sample size 268 550 1316 1260 755

Towns <100 000 population Walk 619 344 222 191 103 Bicycle 1 75 290 249 62 69 Tram/trolley bus 3-8 3 4 1 6 0.1 0.0 Bus 4-5 102 205 130 6-9 Train (overground) 7 0 6 3 8 9 5-4 5.3 Underground 0 4 0 3 0 7 0-4 0 5 Motorcycle 0 7 4 2 3 5 2-3 1-6 Car/van 2-8 11 6 14 7 50-8 675

Sample size 286 649 1478 1651 889



you have an accident ... all that happens is that they have to find you another vehicle quickly. (Ro2)

Conclusions: implications for contemporary transport policy

Because the journey to work is an everyday event, experienced by most people at some time in their

lives, it is a topic on which many individuals hold strong views and opinions. Perhaps the results pre sented in this paper will come as no great surprise to most readers, in that they reflect common-sense assumptions about how the journey to work has changed throughout the twentieth century, usually constructed from personal experiences and historical anecdote. However, it is important to stress that such changes, especially their precise nature and




Table 7 Main mode of transport for journeys since 1890, by gender (%)

Mode Time period

1890-1919 1920-39 1940-59 1960-79 1980-98

Males Walk 44-3 23 2 12 3 8-0 7 2 Bicycle 13-2 21 6 21 4 5-5 71 Tram/trolley bus 13 8 90 3 3 00 00 Bus 98 108 177 113 58 Train (overground) 144 175 189 17-9 21 9 Underground 1 2 3 2 3 8 3 9 40 Motorcycle 05 3 5 3 9 3-3 1-8 Car/van 1 6 100 16 3 48-1 509

Sample size 646 1272 2291 2026 1000

Females Walk 503 293 19 7 206 116 Bicycle 50 95 11-7 40 5-1 Tram/trolley bus 178 13 2 62 02 01 Bus 142 222 317 24-8 161 Train (overground) 86 203 18-1 11 1 101 Underground 3-6 3 9 6-6 7 2 6 1 Motorcycle 00 09 0-6 1.1 10 Car/van 0-5 0 7 4-4 29-9 49 0

Sample size 197 433 1457 1677 1108



timing, have not previously been demonstrated. The fact that, by and large, they reflect commonly held assumptions makes them no less significant. The data clearly show the extent to which many aspects of the journey to work are of long standing, and the ways in which individual decisions about travel modes relate to the availability of transport technology, urban morphology and individual preferences.

We consider that the historical analysis of such trends also has contemporary relevance for the

management of traffic in cities. The 1998 Govern ment White Paper on transport stated that 'People ... want more choice on whether to use their cars ... they want a better public transport system ... they want better protection for the environment and they want less pollution' (DETR 1 998a, 5). The integrated transport policy described in the White Paper claims to offer 'the framework to produce better public transport and easier access to work

places and other everyday facilities for all' (5). The analysis presented in this paper has demonstrated a number of trends that have implications for the fulfilment of the government's stated objectives on transport.

First, it has been shown that inequalities in access to and use of transport for commuting are deeply embedded and of long standing, with women con sistently utilizing slower forms of transport, despite changes in technology. Thus, any strategy to improve access to fast, cheap and efficient transport for all

must recognize and tackle these deep-seated inequalities. Second, the analysis also shows that, especially amongst men, there appears to be a strong and long-standing preference for the use of independent and private forms of transport rather than more communal and public transport. Men have often chosen to use first bicycles and then cars because of their flexibility and privacy; any integrated public transport system must deal with such attitudes




and preferences. Third, and more positively, the history of commuting in the twentieth century clearly

demonstrates that people are prepared to undertake major shifts in the forms of transport that they predominately use. Given the right transport environment, it should be possible to encourage commuters to make further changes in their patterns of commuting. Fourth, it has also been demonstrated that, throughout the second half of the twentieth century, settlements offering the greatest range of public transport have had consistently lower levels of car usage. This suggests that greater investment in

public transport can persuade a substantial pro portion of people to switch from private to public transport. The history of the journey to work in the twentieth century thus offers both encouragement and caution for attempts to develop an integrated transport policy that provides a more sustainable pattern of commuting into Britain's major cities.

Whilst historical analysis cannot provide all the answers, we argue that lessons from the past should not be ignored in the formulation of contemporary transport policy.

Acknowledgements Thanks to the Leverhulme Trust who provided funding for the project, and to all respondents who supplied data for the research.

References

Banister D (1994) 'Reducing the need to travel through planning' Town Planning Review 65, 349-64

Barlow Report (1940) Report of the Royal Commission on the distribution of the industrial population Command 6153 (HMSO, London)

Beatty C and Haywood R (1997) 'Changes in travel

behaviour in the English Passenger Transport Executives

areas 1981-91' Journal of Transport Geography 5, 61-72 Breheny M and Rockwood R (1993) 'Planning the

sustainable city region' in Blowers A (ed) Planning for a

sustainable environment (Earthscan, London), 150-89 Champion A (1 989) Counterurbanization: the changing

pace and nature of population deconcentration (Arnold, London)

Daniels P (1970) 'Employment decentralization and the journey to work' Area 1, 47-51

(1980) Office location and the joumey to work (Gower, Aldershot)

Daniels P and Warnes A (1983) Movement in cities (Methuen, London)

Dasgupta M, Frost M and Spence N (1990) Mode choice in travel to work in British cities 1971-81 Research Paper,

Department of Geography, London School of Economics

Davies W and Musson T (1978) 'Spatial patterns of

commuting in South Wales, 1951-71' Regional Studies 12, 353-66

DETR (Department of Environment, Transport and the Regions) (1 998a) A new deal for transport: better for everyone. Government White Paper on the future of transport (Stationery Office, London)

-(1998b) Transport trends (Stationery Office, London) DoT (Department of Transport) (1981) Transport statistics

1970-80 (HMSO, London) - (1993) Reducing transport emissions through planning

(HMSO, London) - (1 994-96) National travel survey (HMSO, London) European Foundation (1987) Commuting in the European

community (Office for the Official Publications of the European Communities, Luxembourg)

Gibbs W (1997) 'Transportation's perennial problems' Scientific American October, 32-5

Hanson S and Pratt G (1991) 'Job search and the occu

pational segregation of women' Annals of the Association of American Geographers 81, 229-53

Jansen G (1993) 'Commuting, home sprawl, job sprawl and traffic jams' in Salomon I, Bovy P and Orfeuil P (eds) A

billion trips a day: tradition and transition in European travel patterns (Kluwer Academic, Dordrecht), 101-28

Lawton R (1963) 'The journey to work in England and

Wales: forty years of change' Tijdschrift voor Econo mische en Sociale Geographie 44, 61-9

(1968) 'The journey to work in Britain: some trends and problems' Regional Studies 2, 27-40

Liepmann K (1944) The journey to work: its significance for industrial and community life (Kegan Paul, London)

Masser I, Sviden 0 and Wegener M (1992) The geography of Europe's futures (Belhaven Press, London)

McDowell L (1997) Capital culture: gender at work in the city (Blackwell, Oxford)

McGhie C (1997) The Sunday Telegraph guide to com muterland: country life within reach of London (Arum Press, London)

Ministry of Transport (1963) Traffic in towns: a study of the long term problems of traffic in urban areas (Buchanan report) (HMSO, London)

Office for National Statistics (1998) Social trends 28 (Stationery Office, London)

Orfeuil J and Solomon I (1993) 'Travel patterns of the

Europeans in everyday life' in Salomon I, Bovy P and

Orfeuil J (eds) A billion trips a day: tradition and

transition in European travel patterns (Kluwer Academic, Dordrecht), 33-50

OECD (1995) Urban travel and sustainable development (Organization for Economic Cooperation and Develop

ment, Paris) Pickup L (1983) A European study of commuting and its

consequences (Shankill, Dublin) Pooley C and Turnbull J (1998) Migration and mobility in

Britain since the eighteenth century (UCL Press, London)




Pucher J and Lefevre C (1996) The urban transport crisis in Europe and North America (Macmillan, Basingstoke)

Schafer A and Victor D (1997) 'The past and future of

global mobility' Scientific American October, 36-9 Sneed R (1961) The traffic problem in towns (Manchester

Statistical Society, Manchester) Warnes A (1972) 'Estimates of journey to work distances

from census statistics' Regional Studies 6, 315-26 (1975) 'Commuting towards city centres: a study

of population and employment density gradients in

Liverpool and Manchester' Transactions of the Institute of British Geographers 64, 77-96

Westergaard J (1957) 'Journeys to work in the London region' Town Planning Review 28, 37-62

White P (1995) Public transport: its planning management and operation 3rd edition (UCL Press, London)

Whitelegg J (1997) Critical mass: transport, environment and society in the twenty first century (Pluto, London)



the journey to work: a century of change

Documents