See discussions, stats, and author profiles for this publication at: [Link]

net/publication/222276916

Park and ride: Could it lead to an increase in car traffic?

Article in Transport Policy · January 1995

DOI: 10.1016/0967-070X(95)93242-Q

CITATIONS READS

107 3,847

1 author:

Graham Peter Parkhurst

University of the West of England, Bristol
99 PUBLICATIONS 3,176 CITATIONS

SEE PROFILE

All content following this page was uploaded by Graham Peter Parkhurst on 19 April 2016.

The user has requested enhancement of the downloaded file.

 Tmnsport Policy, Vol. 2, No. 1, pp. 15-23, 1995
Copyright 0 1995 Elsevier Science Ltd
Printed in Great Britain. All rights reserved
0967-070X/95 $ IO.00 + 0.00

Park and ride: could it

lead to an increase in car traffic?

Graham Parkhurst
ESRC Transport Studies Unit, University of Oxford, I I Bevington Road, Oxjord OX2 6NB

The detailed and long-term effects of park and ride are more complex than generally acknowl-
edged. In 1994, studies of park and ride were carried out in Oxford and York. The success of
park and ride in attracting users was confirmed, and these users reported high levels of satis-
faction. However, some users had switched from modes other than car and others were found
to be making additional trips which had been attracted by the opportunity to park and ride.
Congestion has remained persistent in the cities studied and there is concern that total travel
may have been increased rather than reduced, which was the intention. It is concluded that, for
park and ride to be successful, ‘package’ policies will need to be strengthened to favour conven-
tional public transport and that particular care will need to be taken by other local authorities
adopting park and ride.

Keywords: park and ride, public transport, integrated transport policy, interchange

This paper considers some effects of bus-based park and ists and is successful in attracting users. In fact more
ride systems on the host city and surrounding area. detailed research has been carried out on this system.
Bus-based park and ride in the UK originated in the In the first years of operation, researchers at the
1960s. Great interest followed in the 1970s and this has Transport Studies Unit of Oxford University (TSU)
re-emerged with considerable influence in recent years. monitored the introduction of park and ride and were
The number of cities considering schemes continues to critical of the low
grow, as does the range of settlement size and type for levels of patronage and the corresponding financial
which park and ride is considered an appropriate strat- liabilities (Papoulias and Heggie, 1975; Heggie and
egy. Nottingham has more than 2000 permanently Papoulias, 1976; Papoulias and Heggie, 1976). They
available spaces. Many places, such as Bath and York, were also the first to show, for Oxford’s western site,
although having less than a thousand dedicated spaces, that although most patronage had been derived from
can offer more on Saturdays by negotiating the use of motorists who would otherwise had driven to the centre
car parks used in the week for other purposes. Bristol, a (57-68%), people who had formerly travelled by bus
city of 400 000 inhabitants, currently has one site with were also using park and ride (58%). They also found
1500 spaces, but plans to provide 12 000 spaces in ten that ‘informal’ park and ride had previously been prac-
sites. Winchester, Guildford and Brighton have all tised by IO-14% of their sample (254 users), and that
initiated schemes recently. 12-14% had switched allegiance from Oxford’s first,
Furthermore, the scale of existing and planned southern, site when the western one opened.
schemes continues to increase. Although not in the same Later, when the future of this western site in parti-
league as Bristol, the cities of York, Chester and Oxford cular was in jeopardy, White (1977) found that
are developing proposals to provide more than 4000 depending on the day of use, 48-5 1% of a sample of
spaces each. At Oxford, 20% of central area parking 415 users would drive to the centre instead, whilst
supply is now provided in park and ride car parks. If 22-30% would travel by bus if the site were to close,
only public parking space is considered, then this figure which was rather more than the number who would use
approaches 50%. another park and ride site (13-16%). This suggested that
Despite this fast and widespread expansion, relatively bus services were seen by many to be a better option
little research exists to demonstrate the wider impact of than travelling (an extra) three miles by car.
park and ride policies. The Oxford system provides a Future surveys would tend towards White’s account
case-study which is 2 1 years old and has often been seen of significantly overlapping demand for the various
as a test-bed, if not quite a blue-print, for many other mode-sectors, rather than the modest levels identified by
schemes. However, observation alone can only provide Heggie and Papoulias. In any case, the strength of politi-
the conclusions that park and ride is popular with motor- cal will in favour of park and ride proved to be the most
Park and ride: could it increase car traffic:‘: G Parkhurst

important factor in deciding the fate of park and ride in to move on to other concerns and the usual machinery of
these early years, and the system survived further trials dissemination has not operated.
to give rise to more favourable patronage statistics. Other contributory reasons may be that the apparent
Since the mid-1970s, research on the system has benefits of park and ride were irresistible, or that those
largely been the preserve of another Oxford-based with an interest in over-viewing the subject lacked the
group, Bixby and his students at the Department of necessary survey-budgets for the provision of empirical
Town Planning, Oxford Polytechnic.’ Examination of evidence. Certainly, local authorities seem to have been
their survey reports shows that, in addition to recording primarily concerned with collecting data to confirm the
the build up of park and ride patronage over the decades, viability of existing park and ride sites or to assist in
they identified persistent behavioural responses that negotiating their implementation.
were not anticipated, or not intended, by the architects For these reasons, it was felt that a large-scale and
of park and ride. objective overview of park and ride was long overdue.
In 1978, four years after the first of the present sites To this end, studies of two park and ride operations were
opened, Devonald et (11. (1978) surveyed 1000 users. carried out as part of a project to consider the current
From their Wednesday survey, around two-thirds of state of transport interchange, funded by the Rees
their sample saw the car as their alternative mode, whilst Jeffreys Road Fund, with additional support from the
a quarter would travel by bus. However, on Saturday, ESRC. The case-studies selected were the schemes at
more would use the car, and bus use would be trivial. Oxford and York. Oxford’s is currently the largest in the
They noted that for the two sites, at least a quarter and a UK, with 3700 spaces, and the oldest, this being the 2 1st
third of patronage respectively seemed to originate from year of continuous operation. York’s, with 563 spaces,
relatively urban areas. Also, there was evidence that was a success soon after launching as a permanent, full-
walk trips were making up 5-10% of the trips to one time facility in 1990.
site. Bixby (1978) was sufficiently confident of these The survey in York contacted 600 users and was con-
results to conclude that about a quarter of users did have ducted by York City Council and analysed by TSU. The
a bus alternative and also suggested that park and ride Oxford survey, involving 1750 users, was a TSU ven-
had the power to generate travel. ture. Both surveys involved a Friday and a Saturday.
Collins et al. (1987) in a survey of 550 users on the Survey forms were handed to users of the bus service
re-launched east-west route in Oxford found that 4 1.5% for self-completion whilst waiting to depart and col-
of users had switched from the car, but 39.3% from con- lected at the alighting stop. Buses were sampled at the
ventional bus routes and 7.3% had already been using same rate from the start of operations until the late after-
park and ride. To some extent this seems to have been noon, with the intention of approaching a representative
due to significant numbers of users walking to the sites sample of users arriving at different times. The data
to take advantage of a cheaper service. were weighted to correct for different overall sampling
Newson (1993) discusses another study carried out by rates between the two survey days, and to provide an
the same institution a year later which contacted 745 estimate of weekly use.
users and found that 58% saw driving to the centre as It should be noted that although the size and historic
their alternative, whilst 29% would use bus or rail ser- quality of the cities are at least comparable, there are
vices and 8% would suppress their journeys. other important regional, socio-economic and contextual
More recently Cooper (1993), working in York, found policy differences. In addition the surveys were similar,
that 15% of trips to the site by her sample of 150 high but not identical; there were a few questions that were
frequency users had been made on foot and that 19% difficult to ask in Oxford, due to the length of time since
had previously travelled by bus. She further notes that opening. Conversely, some questions could not be asked
35% of the sample thought their alternative to park and in York because sufficient experience does not yet exist.
ride was a conventional bus, but the full implications of Other questions had to be asked differently in each city.
the implied mode-shifts are not considered.’ For these reasons, the analysis should not be considered
Although some professionals may have had concerns to be a direct comparison of the two cities, but rather an
that the impact of park and ride was not being fully investigation as to whether similar conclusions can be
monitored, only in a very few cases have these concerns drawn despite these differences.
been publicised. Huntley (1993), for example, notes the
absence of traffic reduction effects.
The effects of park and ride
One reason for this seems to be that the work carried
out on park and ride has often been in the form of stu- Transport policies can be expected to have their
dent projects. Once completed, the authors have tended intended beneficial effects, but also unexpected neutral
effects and perhaps unintended negative effects. The
benefits of park and ride are well established, accepted
’ Now School of Planning, Oxford Brookes University. and publicised. However, the dis-benefits have received
?However, after completing this research, a study carried out in much less concern or discussion. In coming to a bal-
Chester came to light (Jones, 1994) which had considered many simi-
anced view, we need to consider potential disadvantages
lar issues. Due to resource constraints, this research was contined to
shopping trips from one of the three Chester sites. however there were or problems, in order to judge in which direction to
similar findings. some of which are noted later in this report. strengthen park and ride in the future, and whether suc-

16
 Park and ride: could it increase car traflc?: G Parkhurst

cess requires changes in other related aspects of trans- replaced by new car trips and a congested equilib-
port policy. rium has been maintained (Parkhurst, 1993).
(iii) A cost-benefit assessment of overall resources may
Summary of benefits show that a subsidy is justified, and County
Councils are increasingly prepared to promote and
(9 The most important benefit of park and ride that has
fund schemes. However, on occasion in the past,
emerged is the enabling effect on plans for eco-
the costs of park and ride have fallen on the urban
nomic and environmental enhancement. In political
resident. Some schemes have sought to discourage
terms, a successful park and ride scheme can also
people travelling on park and ride buses unless
help to win support for other plans for transforming
arriving at the site by car.s In these cases, there is
town centre environments, such as pedestrianisa-
an equity problem since local people are denied the
tion, which might otherwise be resisted.
opportunity to walk or cycle to the site, to take
(ii) Some vehicle movements can be removed from that
advantage of a favourable bus service, despite con-
part of the urban road network ‘inside’ the park and
tributing financially to it.
ride site. This provides local congestion relief,
reduced energy consumption, and some reduction
(iv) Although the total market for bus travel inside the
site is increased, and the park and ride bus service
in air pollution. can often be run commercially once patronage is
(iii) The parking space provided at the edge of the city established, some form of subsidy is typically
can increase the overall supply, whilst allowing required for the site itself. Bus services outside the
central land to be used for a more economically site may have to compete for some marginal users
beneficial purpose. There is also the potential to for whom park and ride makes it easier to travel the
reduce central parking supply, rather than just long-haul part of the journey by car. This implies
restrain use, although there are no convincing less revenue to the public transport operator,
demonstrations of this having happened in the UK, reduced bus energy efficiency and in some cases
in some cases because of the predominance of pri- either service cuts or a greater need for public sub-
vately controlled parking supply. sidy.
Traffic and parking capacity can be restrained in the (VI Although increased accessibility may favour the
centre of a major urban area without needing to centre of the host urban area, the natural result is
reduce the overall access to the centre of the city also more trips, from greater distances. As park and
from outside. Even if the volume of car use in the ride is often associated with an increase in total
centre is not reduced, it is possible to increase the parking stock, additional trip-end opportunities
total numbers of travellers to the centre without the will exist. Furthermore, the separation of home,
additional congestion and environmental intrusion work and other travel-generators becomes more
that would have happened if those extra people had acceptable as the penalties of access are reduced.
come by car. The cost of tackling transport problems by focusing
Motorists report favourable experiences with park only on the city in isolation may be increased rural
and ride. The opportunity to travel in a stress-free development.
manner, often via bus lanes, is valued. Upon arri-
val, preferential access may be given to the
shopping area and difficulties encountered with
Survey evidence
multi-storey parking are avoided. At the same time, The information from the Friday surveys is marked in
the benefits of car use - flexibility, immediacy of the tables by an ‘F’, Saturday surveys by an ‘S’ and the
access, speed and privacy - are enjoyed for the derived 6-day estimation by a ‘6’.
main part of the journey. It is possible that a
favourable experience with public buses will Current demand for park and ride
encourage further use of public transport. Table 1 shows the journey purpose for visits to the two
By increasing the accessibility of the central core, cities. Commuter journeys are the most important travel
pressures for suburban development and out of purpose, especially for Oxford where study trips were
town relocation may be reduced. separately identified. In both cities, work and shopping
trips account for three-quarters of park and ride demand.
Summary of d&benefits Table 2 identifies how users arrived at the site. Car
(i) The most obvious drawback is the environmental arrivers account for nearly all park and ride bus trips in
impact of constructing large car parks in sensitive Oxford, but in York, nearly one in five walk or cycle to
land, often green belt, on the urban fringe. the site. From the figures for car drivers and passengers,
(ii) However, a more serious shortcoming is probably car occupancy to the sites in both cities is around 1.25
the ineffectiveness of park and ride in reducing traf- per vehicle, although it is lower on weekdays and much
fic downstream of the sites. No long-term higher on Saturdays.
reductions in traffic levels have been attributable to
park and ride schemes. In both Oxford and York, ‘At Canterbury, for example, a parking ticket is required for travel on
the trips removed from the road network have been the bus; tickets are issued only at a vehicle-detecting barrier.

17
Park and ride: could it increase car trafJic,YJ:
G Parkhurst

Table 1. Journey purpose of bus service users

City Oxford (%) York (%)

Day surveyed F s 6 F S 6
To/from place of work 47.9 8.8 39.7 44.4 15.0 38.6
To/from place of study 6.1 2.5 5.3 N/A N/A N/A
Shopping 27.8 71.2 6.9 31.0 58.4 36.4
Personal business 6.9 2.4 6.0 7.3 2.4 6.3
Social/leisure 6.4 I I.1 7.3 11.9 17.1 12.8
Tourism/overnight stay 1.5 2.2 1.7 I.9 6.3 2.9
Other 3.5 2.0 3.1 3.4 0.7 2.9

Table 2. Mode of arrival at car park of bus users

City Oxford (%) York (%)

Day surveyed F S 6 F S 6
Drove self 75.4 62.X 72.9 58.3 70.3 60.7
Car passenger 15.8 4.0 19.7 14.9 18.5 15.6
‘Kiss & Ride’ 2.0 1.3 I .9 4.3 I .3 3.7
Cycled 0.7 0. I 0.4 I .4 I .o I .4
Walked 5.1 I .7 0.6 19.2 7.9 17.0
Other 0.9 0 0.1 1.8 1.o 1.6

The mean distance of trips to the sites in Oxford was (in the early part of that period) bus services became
20.2 km, with the median trip-length being 14.4 km. In less attractive and usage fell. Therefore the current users
York, the crude calculation yielded a lower figure of who said they previously (10 to 20 years ago) used
13.2 km, with a median of 3.2 km, due in part to those buses include an unknown proportion who, without park
walking to the site. However, the median for car driver and ride, might now be using a car anyway.
and passenger trips alone was still lower than Oxford, at Furthermore, the settlement structure in the Oxford
9.4 km. region was, and remains, more favourable to operating
bus services than that around York.
Previous travel hehaviour- of park and ride users Jones (1994) found for 124 Chester park and ride
In ideal circumstances, each car parked at a site repre- users travelling for shopping that the vast majority,
sents a car removed from the inner part of the city road 84.8%, drove all the way but that 14.3% had switched
network. Clearly, the real situation must fall short of this from bus and train. Also, 7.1% had switched from park
ideal to some degree. Assessing how many users fit into and ride at another site, and a proportion of those prob-
this description is vital for a balanced assessment, but ably did not use a car to access Chester prior to park and
fraught with difficulties. Two questions were asked ride commencing. Shoppers would be expected to be
which can give an indication; first, about the mode that one of the most public transport-averse groups due to
respondents reported they used before the park and ride their ability to alter their destination.
site was opened, and secondly, their alternative to park As mentioned in the introduction, Cooper (1993) also
and ride. found a shift from buses in York. It is notable that in both
For Oxford, most current users had no directly com- the example of previous travel (Oxford) and more recent
parable experience of travel to the city centre before travel (York), a significant switch from bus use to park
park and ride. Consequently, for this particular analysis, and ride occurred. Although the national popularity of car
only those respondents were included who claimed to and bus travel have changed between the two opening
have been commuting since before park and ride ser- dates, there are still a significant number of bus passen-
vices began, and from the same settlement (27.9% of gers who have a car available for journeys they make by
respondents - 369 users). For York, respondents were bus. This phenomenon is considered further below.
simply given the option to specify that they didn’t visit
the city prior to park and ride opening. Just 8.3% fell in Main reason identified for- deciding to use par-k and ride
this category. The benefits to motorists of using park and ride were
Table 3 shows that, of those who met the criteria for strongly identified from prompted options and are
the Oxford survey, little more than half (56%) had pre- shown in Table 4.
viously travelled by car, with most others arriving by For the majority, the push factors of perceived high
bus. In York, where the situation is different for a num- cost and low availability of central parking space were
ber of reasons, more had arrived by car (70%). None the the principal motivations. For Oxford, cost is much
less, nearly a third of York respondents had previously more important than availability, but in York, availabil-
travelled by bus, train or cycle, and most by bus. ity is as important, mainly because parking charges are
In interpreting this result, it is necessary to take account typically one-third of those in Oxford. In both cities, a
of the fact that over the whole period of park and ride in sizeable minority valued most the reduced stress on
Oxford, car ownership has increased substantially, and vehicle and self when travelling by park and ride. The
 Park and ride: could it increase cur traffic?: G Parkhurst

Table 3. Mode of travel of park and ride bus users prior to introduction

City Oxford (%) York (%)

Year of Full-Time Operation 1973,1973,1976,1985 1990
Day Surveyed F s 6 F S 6
Drove and parked in centre 5 1.o 44.3 49.8 60.0 71.8 62.6
Car passenger 3.8 13.9 5.1 6.4 12.7 7.5
Conventional bus 35.2 30.4 34. I 23. I 7.3 20. I
Train 0.7 4.2 I.1 3.2 6. I 3.7
Cycle 55 4.2 5.4 3.2 1.2 2.7
Other 3.8 3.0 3.9 4.1 0.8 3.4

Table 4. Park and ride bus users rationale for choosing this mode

City Oxford (% ) York (%)

Day surveyed F S 6 F S 6
Lack of parking 21.9 23.3 22.2 26.9 32.8 27.9
Cheaper than parking in centre 41.7 38.6 41.1 31.2 21.0 29.3
Avoid stresses on self/vehicle 9.6 13.1 10.3 13.4 24.0 15.7
Easier access to destination 13.0 12.0 12.7 19.4 13.7 18.2
Good for the environment 2.8 2.3 2.7 5.9 2.6 5.3
Unfamiliar with city 2.5 5.0 3.0 I .2 3.7 1.7
Bus service more attractive 6.2 3.3 5.5 N/A N/A N/A
Other 2.5 2.6 2.5 2.0 2.2 2.0

Table 5. Alternative travel behaviour of users if park and ride unavailable

City Oxford (o/o) York (%)

Day surveyed F S 6 F S 6
Not travel today/travel elsewhere 7.2 21.0 10.1 10.5 14.9 Il.3
Travel to city by car 57.0 56.6 56.9 55.5 67.9 58.0
Travel to city by bus 26. I 15.8 23.9 24.3 9.3 21.3
Other 9.7 6.5 9.2 9.7 8.0 9.3

relatively attractive service provided by park and ride For Oxford, a third of the 56.9% who claimed they
buses was a factor for Oxford users, presumably reflect- would still travel to the city by car indicated that they
ing comparisons by former bus users. This option was would seek parking space in the suburbs and then walk
not specified in the York survey. or catch a bus to the centre (informal park and ride).
However, whilst this may be their intention as individu-
How dependent are users on park and ride? als, it is not clear how much parking could actually be
Surveys of intended behaviour do not necessarily give accommodated in the suburbs. Many possibilities have
accurate figures of what people would actually do, but been removed by residents’ zones, but some still exist.
with this caveat, questioning users about the alternative This leaves 32% of all Oxford users who report that they
mode of travel they would use in the absence of park would use a car and park in the centre. In York, where
and ride does reflect their current preferences and per- the option to park in the suburbs was not prompted,
ceptions, and draws on any experience they have of somewhat more than half would choose to travel by car
travel by different modes to the city on other occasions. to the centre, whilst nearly a third would use public
The results provide an interesting comparison with the transport or other means.
analysis above of travel prior to park and ride, together From his sample of shoppers only, Jones (1994)
with an impression of how essential the park and ride found that, of those who would continue to visit, 71.8%
service is perceived to be by its users. These results are identified the car as their alternative, and 15.5% public
shown in Table 5. transport (essentially the bus). However, direct compari-
Notably, 10% of users thought they would either sup- son with the York and Oxford case-studies is difficult.
press their journey or re-direct it to another location. First, shopping trips are discretionary, and secondly, the
This suggests their trips to the study cities were either option to transfer to another site was offered, and this
attracted away from another place, or entirely freshly would attract almost a fifth of users.
generated, which is important if one of the objectives of Significant differences in second-choice mode of
local economic policy is to increase the total number of travel were evident between users travelling for different
visitors to the city. However, some respondents may journey purposes when surveyed. Table 6 shows these
have answered tactically to emphasise park and ride’s results for Oxford. Few trips for work or study could
have been avoided or redirected. Conversely, those on
importance, or might intend to make the journey on a
shopping trips accounted for three-quarters of the travel-
different day.

19
Pur-k und ride: could it increase car traffic.:‘: G Parkhur-st

Table 6. Alternative mode of travel of users travelling for different number of users living in rural areas together with a pro-
purposes
portion of those living beyond 4.5 km also have access to
Pu l-pose (%) Work/study Shopping Other All
high frequency services. However, for this analysis, vil-
lages smaller than 1 km’ or outside 45 km were assumed
Not travel/travel elsewhere X.2 76.4 15.4 I 00 to have poor services.
Travel by car 41.6 36.6 21.x I00 Table 8 indicates that public transport is less likely to
Travel by bus/train/cycle 64.3 23.Y I I.8 100
be an acceptable alternative as the area of origin
Other 40.2 43.Y 15.9 100
becomes more rural, and there is a higher chance of trips
not being made to Oxford. However, for areas with
Table 7. Degree of consideration by Oxford users during residential ‘very good’ services, half see their next best alternative
relocation
as being public transport or cycling.
Day Surveyed (%) F s 6
This analysis has two important implications. First, it
is an indication that public transport policies within the
None X3.3 90.X X4.7 Oxford catchment area may be considered to be notably
Shght 5.0 I.9 4.3 successful, at least for some of the settlements served. It
One of a number of factors 3.5 1.9 3.3
is by no means always the case that the hinterland of a
The most important factor 0.9 0.3 0.8
Didn’t exist when last moved 7.3 5.1 6.9 city the size of Oxford offers reasonable standards of
bus service over such an area. Secondly, it does tend to
support the plausibility of the respondents’ replies, since
lers who claimed they would not travel at all. The most of those reporting ‘bus’ as their alternative mode
majority of those who would travel by bus, train or cycle did have reasonable bus services, and very few of those
were making work journeys, but a significant number of in areas served by poor bus services chose that response.
shoppers would also come by bus. Clearly, park and ride users from urban areas with
Table 7 shows the influence park and ride has had on very good or good public transport are still deriving ben-
residential location. Accessibility is only one of a num- efits. These perhaps take the form of greater
ber of considerations when moving home, so it might be convenience, comfort or perceived cost, rather than jour-
expected that one aspect of travel would not receive ney time by park and ride. However, for this particular
extreme prominence during decision making. None the group, the benefits of car use itself are not so great that
less, nearly 10% of those who had relocated thought it in the absence of park and ride they would still travel by
had been a factor. When asked if they would have to car; for some of them it would be the least favoured
consider moving if park and ride became unavailable, alternative option due to high parking charges, time
7% of the sample thought they would have to. spent parking and loss of access to bus priorities.
The result that needs to be scrutinised most carefully The intention that transport policies should be tailored
is the rather high proportion of people who claim that. in to enhance, rather than detract from, the appeal of public
the absence of park and ride, they would be using public transport services is widely, but not universally shared.
transport for the same journeys. They might be making It might be argued that park and ride may represent an
rather unrealistic assessments of their own behaviour. In opportunity largely to abandon public transport outside
order to test this, we investigated how many users really the urban areas, except for social transport fulfilling a
had a reasonable level of public transport service avail- minimum-access function. However, the travel-demand
able to them, measuring the level of service offered in implications of such a strategy are worrying, especially
each of the settlements from which park and ride users if the levels of trip generation suggested by this survey
came, by an analysis of timetables for services to settle- were to be generalised.
ments within a 45 km radius of Oxford, for the weekday Stokes (1994) has carried out a provisional analysis of
morning peak.” the implications of the stated hypothetical alternative
Of the sample, 36.1% were identified as having mode on total distance travelled. Although it is by no
access to either *very good’ public transport (7 areas) or means straightforward to consider changes to park and
‘good’ public transport (9 areas). In addition, a small ride alone, within the total complex transport system, it
_ seems likely that the park and ride provision may imply
’ For details of this procedure, see Parkhurst and Stoku, 1994. increased travel. A significant factor behind this is the

Table 8. Choice of alternative modes by attractiveness of public trampon

Origin Within Urban (%) Rural ~4.5 km

Oxford (% ) V Good Good Poor (%) 1%)
% of Sample 5.9 20.3 9.9 12.2 39.8 11.9

Not visit Oxford 5.6 6.3 11.3 14.x Il.9 6.8

Car-borne 22.5 43.6 53.3 61.1 61.5 79. I
Bus/Train/Cycle 64.3 49.2 33.8 I x.3 24.3 13.8
Other 7.5 0.9 I .6 5.8 2.2 0.2
All I 00 IO0 100 100 100 I00

20
 Park and ride: could it increase car trajjic.~: G Parkhurst

large mean distance travelled using the private mode native modes to the car for travel to the city. Table 9
before switching to public transport. considers, for all users travelling for any purpose when
His estimate for the sample, assuming that people interviewed, the effect of park and ride not being avail-
who suppress their visit to Oxford would travel the same able for shopping trips to Oxford.
distance somewhere else, is that 10% more vehicle-kilo- Hence, the loss of park and ride would have economic
metres would be travelled (around 30 ooo impacts. There would be a reduction in the total number
vehicle-km/week in absolute terms). In addition, more of shopping trips with some being made in favour of
than 20 000 bus-km are operated per week by park and other towns. However, from Table 6, it appeared that a
ride bus services, whilst in the post-regulation operating fifth of shopping trips might be made differently, but
environment there is typically capacity available on con- when all users were asked, more reasonable levels of
ventional bus services, even at peak times. change were implied. This may be partly because some
A fuller analysis would need to consider the impact of users of park and ride choose other modes when they
the extra car trips on the Oxford road network. Some visit Oxford for shopping anyway.
car-km would replace bus-km saved. However, on some Jones (1994), who interviewed only those users actu-
routes, and at some times outside the peaks, capacity is ally travelling for shopping, found that 12% would stop
available to provide for some extra car trips. The rate of shopping in Chester if park and ride were unavailable,
trip suppression recorded assumes the current degree of and that a further 25% would come less often. Rather
congestion. The actual level might be expected to be more of his sample would stop visiting altogether, but
higher if congestion were to be more severe, although rather less of those who would continue to patronise that
elasticities affecting demand for other modes might tend city would cut down their rate of visiting.
to return the level of congestion towards a robust equi- Table 10 shows the change in frequency of visits
librium. made (by any mode) to York city centre by park and
The assumption that suppressed trips are relocated ride users since the service was established. Virtually no
elsewhere travelling the same distance is even more one had visited less frequently, but a large minority
problematic. A certain proportion of these trips would claimed to have visited more.
not be made at all or an acceptable, although not perhaps Using a similar question in Chester, Jones (1994)
ideal, trip-end may exist locally. For others, however, found that 14.5% of shoppers had visited this city less
Oxford could be considered the most local of an accept- until given the opportunity to park and ride, with evid-
able range of destinations, and greater travel will be ence of both redirection and generation of extra trips.
made to the next satisfactory place. The likelihood of Overall, the number of trips made by the sample would
this latter response should not, however, be exaggerated. be expected to increase by 13.4%. This effect seems to
It seems likely that park and ride has made Oxford be rather more modest than for York, despite consider-
unusually accessible, and that in some cases other loca- ing only shopping trips. This may reflect Chester already
tions which are in fact closer might prove to be being extremely popular as a shopping destination
acceptable alternatives if the total cost of the travel within the Cheshire-Merseyside-Greater Manchester
opportunities from a particular origin were to be region. It may also be an effect of phrasing the question
increased. to refer to behaviour pre rather than posf the introduc-
Finally, it would be desirable to make some allow- tion of park and ride.
ance for the use of cars left at home by those switching
to public transport. However, given the mean range of
Discussion and conclusions
trips made to park and ride sites (40 km return trip), this
travel would have to be considerable to offset the saving Given the restricted range of policy tools available to
by the primary driver using a bus all the way. local authority planners and engineers, park and ride has
naturally been a leading component of ‘balanced’ poli-
cies. Local authority powers and finance are constrained,
Economic effects on city centre and some authorities feel that park and ride systems can
It was noted above (Table 6) that respondents travelling give (relatively) quick and easy benefits, compared with
for shopping were much more likely to indicate they the long haul to implement some other radical policies.
would not visit Oxford without the chance to park and Park and ride offers real benefits to motorists. Some
ride, and also that fewer were prepared to consider alter- people value savings in parking costs, effort, and some-
times travel time. Others enjoy an environment which is

Table 9. Effect of losing park and ride on shopping trips to Oxford Table 10. Visits by park and ride users to York city centre since intro-
(%) duced (%)

Day surveyed F S 6 Day Surveyed F S 6

Shop in Oxford similarly to now 36.8 34.4 36.3 More frequently 39.7 48.4 41.8
Shop in Oxford less and other places similarly 20.1 18.4 19.9 Similar amount 52.1 39.7 48.5
Shop in Oxford less but other places more 34.8 34.3 34.6 Less 0.8 0 0.6
Stop shopping in Oxford altogether 6.6 11.6 7.6 Moved to area since opening (June 1990) 5.8 7.4 5.8
Other 1.7 1.3 1.6 Don’t know 1.6 4.6 3.2

21
Park and ride: could it inwease car trajjic?: G Parkhurst

more pleasant than driving in congested traffic, perhaps people who would otherwise use the bus, walk or cycle
in an unfamiliar area. Some users have been given the to start using park and ride instead?
chance to consider living in places that would otherwise Three policy levers are worthy of consideration. First,
create difficult travel problems. Central area parking there is already widespread support for the idea of
policies based on restraint have been implemented with- further discouragement of car use, by more intense cen-
out deterring visitors. Indeed, in economic terms, more tral area parking control, pedestrianisation, area-wide
people have been able to visit the two city centres, and traffic calming, and perhaps by road pricing, all
land that might have been reserved for parking has been designed to ensure that the personal benefits of car use
available for other uses. to the user are not outweighed by the social and environ-
Provided park and ride services are well used, they mental costs to everybody else. The argument here
have the potential to increase the space efficiency of the would support that, and raises the question of its exten-
urban transport system in the area within the park and sion to areas outside the city centre.
ride provision, and this is the area usually most seriously For example, the current practice for the construction
affected by congestion. If lower levels of traffic are of bus priorities in urban areas focuses on where provi-
identified in the urban area, and similar levels in the hin- sion is possible rather than where need is greatest. At
terland, then the overall system is more efficient. If we many junctions and bottlenecks, the only way to achieve
find higher levels in the hinterland, that may not contra- further priority on main routes will be through appor-
dict overall policy, but it is necessary to examine tioning road space on a ‘time-share’ basis rather than
whether these increases can be justified by the savings in physical segregation. Consequently there is a growing
the city. interest in the use of traffic light priorities and queue
However, what this research has shown is rather more relocation arrangements. These incorporate variable tim-
worrying. Not only does the ‘energy audit’ approach of ing which gives the option to manipulate the flow of
vehicle-kilometres travelled seem to strongly suggest general traffic. As car trips are diverted, reductions in
significant increases in travel, but there is apparently no traffic volume are not offset by further growth.
‘decongestion dividend’ in the urban areas studied Secondly, it is necessary to improve the attractiveness
against which this can be offset. This means that the of public transport outside the centre, so that it can com-
success of park and ride has to be justified in terms of pete on the long-haul part of the journey, not only the
politics and economics; that it makes the ‘package’ town centre part. The provision of bus priority measures
approach an acceptable one and that it facilitates visitor on major routes outside urban areas would enable buses
growth without encouraging traffic growth. to offer a more competitive overall journey time over
What seems to be missing amongst the various the route to the edge of the city. For some users, this
strands of policy is a need to consider the region as a might be expected to restore the advantage of travelling
whole, including the impact on traffic growth, trip by bus for the whole journey. In addition these measures
generation and attraction, residential location and would benefit all inter-urban bus travellers, some of
switching from modes other than the car. Jones (1994) which may not currently possess the opportunity to park
has argued that models used to predict the cost-benefits and ride.
of park and ride could be made more robust by includ- Thirdly, it would be desirable to find a way of using
ing some of these effects, and by avoiding the the pricing structure of the park and ride sites to encour-
assumption that travel demand, both on the part of users age this effect. A strategy which sought to impose some
and non-users, is inelastic. He implies, however, that or all of the costs of car park provision on users would
some schemes would not be affirmed on that basis. deter some current marginal users. However, this is
For two reasons the ‘balanced’ strategy has, to some complex, since it is not practical (and would in any case
extent, been undermined. First, park and ride actually be very inequitable) to charge different prices to differ-
increases the share of road space available for that por- ent users, and because trips would divert to different
tion of traffic which does not intend to park centrally, or modes depending upon a number of factors. But, a more
has access to private spaces. detailed study should perhaps consider whether some
Secondly, rather than reducing economic inequalities, additional charge, applied at the same time as an
the inducements to users mean that further anomalies improvement in public transport services as suggested
have resulted. The cost of providing the necessary facil- above, would have the desired effect.
ities has often been met by public resources rather than Charging users of the site would also assist in the cost
user-charges. Consequently, new car trips to the edge of of site provision and a charge levied on the car would
the city have been encouraged and some trips are now favour higher car occupancy. Charging for the bus and
made by car which were previously made by other site could be administered in a number of ways. One
modes. possibility would be to include a permit-to-travel for the
The question that arises, therefore, is whether it is driver with the parking ticket, with other passengers
possible to keep hold of all the positive effects, while paying on the bus.
using other policy instruments to offset any negative If these levers could be applied in such a way that the
effects. In particular, is it possible to attract the maxi- predominant mode switch was from park and ride to
mum number of people who would otherwise use a car, bus, then a useful by-product would be that the spare
but with the least possible encouragement to divert capacity at sites could contribute to stimulating and

22
 Park and ride: could it increase car trajjic?: G Parkhurst

accommodating demand for park and ride services. Oxford Bus Company and Thames Transit are acknowl-
Extra peak public transport services might have to be edged. The tabulated data presented are results from
provided if some commuters switched to this mode, studies of park and ride carried out by Gordon Stokes
although capacity may currently exist within the system. and the author in 1994. An earlier version of the paper
However, additional peak services would tend to be was presented to the 22nd PTRC European Transport
profitable and to enhance the overall service frequency. Forum, 12-16 September 1994, University of Warwick,
At best, for the case-study of Oxford, this could free UK. The conclusions are those of the author.
up to 800 spaces per day, a number equivalent to provid-
ing a proposed fifth site, although it is unlikely that these
References
would all be divertible in practice. The construction
costs saved by not building extra capacity would typi- Bixby, B (1978) ‘A study of park and ride use and the influence of
parking availability on patronage: Oxford’ Working Paper 35,
cally be of the order of &l loo/space plus land
School of Planning, Oxford Brookes University
acquisition costs. Collins, S, K Dykesman, S Cornwell, S Marsh and I Robinson (1987)
Hence, the intention would be to create a hierarchy of ‘The park and ride experience: Thornhill-Seacourt one year on’
costs which favoured public transport by charging for Unpublished Student Project, School of Planning, Oxford Brookes
the benefits of car access to park and ride sites. At the University
Cooper, B (1993) ‘An investigation into the behaviour and attitudes of
same time, the relative advantages of park and ride over users of peripheral and city centre parking in York’ [Link]. Thesis,
driving to the centre need to be increased, so that more Transport Studies Group, University of Westminster
of the motorists who currently drive to the centre are Devonald, S, P Dickinson, R Morgan, D Murray, C Smyth, A Thomas,
intercepted. Such a scenario is not contingent upon, but C Villar and P Wright (1978) Park and Ride Study: O.$ord School
of Planning, Oxford Brookes University
might be easier to pursue, if measures became available
Heggie, I G and D Papoulias (1975) ‘Forecasting the number and char-
in the future to control private parking or road use. acteristics of park-and-ride users in Oxford’ Working Paper 5,
It is necessary to emphasise that such policies cannot Transport Studies Unit, University of Oxford
be implemented by local authorities alone, especially in Heggie, I G and D Papoulias (1976) ‘Operational performance of
the context of the trunk roads programme, bus deregula- park-and-ride: objectives and achievements in Oxford’ Working
Paper 23, Transport Studies Unit, University of Oxford
tion, rail privatisation, constraints on public expenditure, Huntley, P (1993) ‘Park and ride: strategic shift or cheap overflow?’
and the need to obtain Department of Transport Local Transpor-t Today, 22 July 12-l 3
approval for such innovative policies. There would need Jones, R (1994) ‘The effect of park and ride on users’ travel
to be encouragement from central government. behaviour’ [Link]. Thesis, Department of Engineering, University of
Newcastle-upon-Tyne
Finally, we should note that in practice, at least, park
Newson, N ( 1993) Tmw Cenrre Truflc Munugmenr: The Key Role of
and ride by rail typically differs from bus-based systems Park und Ride Oxford City Council
in a number of respects. Consequently, caution must be Papoulias, D and I G Heggie (1976) ‘A comparative evaluation of
observed in generalising results and conclusions beyond forecast and use of park and ride in Oxford’ Working Paper 22,
the scope of this study. Transport Studies Unit, University of Oxford
Parkhurst, G P (1993) ‘A comparison of policies aimed at controlling
car use in the historic cities of Oxford and York’ Paper to 2lst
PTRC Summer Annual Meeting, UMIST, Manchester, UK
Acknowledgements Parkhurst, G P and G Stokes (1994) Pork und Ride in Oxford und
This paper reports research carried out as part of the York; Repour of Surveys (1994) Transport Studies Unit, University
of Oxford, Ref 797
Transport and the Environment programme of the
Stokes, G (1994) ‘The effects of park and ride on travel demand’
Economic and Social Research Council, with additional forthcoming Working Paper, Transport Studies Unit, University of
support from the Rees Jeffreys Road Fund. The coopera- Oxford
tion, assistance and advice of York City Council, White, C J (1977) ‘Park and ride in Oxford: and a survey to investigate
Oxford City Council, Oxfordshire County Council, the possibility of closing the Botley Road Site’ Student Project,

23

View publication stats