Driving Forces behind the Growth of Per‐capita Car Driving Distance in the UK, 1970–2000

Abstract

Although per‐capita car trip distance (measured in passenger‐km) and car driving distance (measured in vehicle‐km) in the UK have kept increasing, their growth rates slowed considerably in the 1990s when compared with the 1970s and 1980s. The paper investigates the main driving forces behind the changes in car trip and car driving distances, and it examines the determining factors for the slow down of growth in the 1990s on the basis of the analysis of data from the National Travel Survey (1975/76, 1989/91, 1992/94, 1995/97 and 1999/2001). In particular, it emphasizes the significance of changes in car ownership levels as a key driving force and attempts to separate this ‘car ownership effect’ from other effects. The log‐mean Divisia index decomposition method is applied to measure the relative contribution of each effect. Separate analyses are undertaken according to trip purpose. Other underlying causes, such as changes in fuel price and road capacity, are also examined.

Notes

1. Car driving distance (measured in vehicle‐km) is determined by car trip distance (measured in passenger‐km) divided by the average occupancy rate (passengers per vehicle), which will be analysed below.

2. The definitions of ‘car ownership effect’ and ‘car use effect’ are presented below.

3. Data sources for all the figures and analysis are from the NTS data set unless other data sources are specified.

4. This figure, which is based on NTS data, is different from the passenger‐km data from the National Transport Statistics Great Britain (NTSG). The latter relies on the vehicle‐km data and occupancy rate to calculate the passenger‐km. Figure 1 was based on NTSG data.

5. This proportion is based on the NTS. It might be slightly different from Census data.

6. Ang et al. (1998, 2003) present the general form of the decomposition equation using the additive log‐mean Divisia index decomposition method.

7. Trip purposes are classified here into six categories: commuting, business, education, personal business (including shopping), leisure and holiday. This grouping is based on the judgement that trip patterns of sub‐categories would be similar in each group. This classification is also generally compatible with NTS reports by the Department for Transport. Commuting trips are trips to a usual place of work from home, or from work to home (‘Commuting’ in the trip purpose categories (data code J28) of the NTS data). Business trips are trips in the course of work (‘Business’ and ‘Other work’ in data code J28 of the NTS). Education trips are trips to school or college (‘Education’ in data code J28 of the NTS). Personal business trips include visits to services, e.g. hairdressers, launderettes, solicitors, banks, libraries, churches and for medical treatment. They also include trips for eating and drinking, unless the main purpose is entertainment or social (Department of Transport, Local Government and the Regions, 2001a). In this study, shopping trips are also included in this category, which therefore includes ‘Shopping’, ‘Personal business medical’, ‘Personal business eat/drink’ and ‘Personal business other’ in data code J28 of the NTS. Leisure trips in this study include ‘Visit friends at a private home’, ‘Eat/drink with friends’, ‘Other social’, ‘Entertain/public activity’, ‘Sport: participate’, ‘Day trips’, ‘Just walk’ and ‘Other non‐escort’ in data code J28 of the NTS. Holiday trips are ‘Holiday: base’ in data code J28 of the NTS. All escort trips are distributed according to the trip purposes of the accompanied persons. Since the NTS 1975–76 data set does not include the trip purposes of Escort trips, they are distributed into each trip purpose roughly by a similar proportion with other data sets.

8. Although the discussion here is based mostly on the analysis of NTS data 1975/76, 1989/91 and 1999/2001, analysis of other NTS data sets (1992/94, 1995/97) was also used to check the consistency of trends. For example, the data on holiday trips had an irregularity in the trend, which was a key reason for the exclusion of holiday trips from the analysis in the following section. This seems to be due to small samples of holiday trips in the NTS survey. According to Department of Transport, Local Government and the Regions (2001a), for trip estimates using NTS data, sub‐samples under 300 should not be used to generate estimates, whilst sub‐samples of under 1000 should be used cautiously. According to the grouping of this paper, only car trips for holiday purpose in the population group with no household car had less than 1000 observations (more than 300). The sample size of other categories generally far exceeds 1000.

9. Office for National Statistics (various years).

10. This definition of a ‘teleworker’ is quoted from Department of Trade and Industry (2002a).

11. Department of Trade and Industry (2002a).

12. Department of Trade and Industry (2002a).

13. Department of Trade and Industry (2002b).

14. Department for Environment, Food and Rural Affairs (DEFRA) (2003).

15. ‘Other work’ category, which was included from the 1989/91 NTS survey, is classified into the Business trip category in this study.

16. Due to the data problem mentioned above, the period 1975/76–1989/01 was excluded from the discussion.

17. Kwon (2004) provides more detailed figures.

18. Dodgson et al. (1997) estimated that video conferencing could reduce car business travel by between 5 and 20% by 2007, and by between 10 and 40% by 2017. However, Dodgson et al. (2000) revised this figure to 3% by 2005 and by 5% by 2010. The estimation of Dodgson et al. (1997, 2000) is the impact of the increased use of telecommuting technology in reducing traffic growth of different types below what it would otherwise be.

19. This point was also mentioned by Noble and Potter (1998).

20. For detailed figures, see Kwon (2004).

21. Office for National Statistics (2001).

22. Department for Transport (2002b).

23. Department of the Environment, Transport and the Regions (2000b); Department of Transport, Local Government and the Regions (2001b).