Land Use Policy 26 (2009) 792–798

Contents lists available at ScienceDirect

Land Use Policy

journal homepage: www.elsevier.com/locate/landusepol

Attitudes towards compact city living: Towards a greater understanding of

residential behaviour

Peter Howley∗

Rural Economy Research Centre (RERC), Teagasc, Athenry, Co. Galway, Republic of Ireland

a r t i c l e i n f o

Article history:

Received 11 April 2008

Received in revised form 30 July 2008

Accepted 13 October 2008

Keywords:

Compact city policy

Residential mobility

a b s t r a c t

Policy prescription in most Western societies has increasingly favoured urban intensification policies in

order to ensure a more sustainable development pattern. In particular, it is now widely felt that residential

decisions concerning where to live profoundly affect, among other things, environmental pollution,

resource use and land and habitat loss. Using the central area of Dublin city as a case study, this paper

focuses specifically on garnering a better understanding of the residential behaviour of residents who

have moved into new relatively high-density residential environments. This is a group who have made

the choice to move into a relatively high-density urban area and hence it will be revealing to assess

the motives, preferences and future intentions of this residential population. Findings suggest that the

ultimate residential preference of the majority of residents in these areas is for lower density locations

which call into question the long term success of urban intensification efforts. Results from a logit model

of residential mobility indicate that stage in their life cycle, satisfaction with both the dwelling and the

neighbourhood emerge as significant predictors of respondents intended future mobility patterns.

© 2008 Elsevier Ltd. All rights reserved.

Introduction

Recent years have witnessed an increasing emphasis placed on

planning systems in most advanced capitalist societies to develop

a more sustainable development pattern, resulting in policies to

increase residential densities. It is now widely felt that residential

decisions concerning where to live profoundly affect, among other

things, environmental pollution, resource use and land and habitat

loss (Gillham, 2002; Haughton and Hunter, 2003). Much of the

focus of interest in sustainable cities, for instance, relates to the

apparently relentless increase in the demand for car travel and the

contribution that higher residential densities can make in reducing

the negative environmental impacts associated with car dependency

(Newman and Kenworthy, 1989; Banister, 1997; Banister et

al., 1997). The empirical evidence underpinning compact city policy

is, however, mixed and often contradictory (see Neuman, 2005

for a recent review). That said, despite conflicting views, European,

national and local policy and discussion documents have increasingly

advocated a high-density mixed-use urban form as a means

to ensuring amore sustainable development pattern. As evident by

rapidly sprawling development patterns, however, individual residential

preferences appear to be at variance with this policy agenda.

∗ Tel.: +353 091 845 295.

E-mail address: peter.howley@teagasc.ie.

In light of these residential preferences many commentators now

question whether the process of urban sprawl can be reversed

(Breheny, 1992, 1995, 1997; Heath, 2001; Senior et al., 2004). Specifically,

Breheny (1997) has suggested that the compaction case can

be subjected to three types of tests: veracity, feasibility and acceptability.

He argues that the acceptability of urban compaction to

the public is the most neglected test of the three, yet it may be

the point on which the whole issue turns (Breheny, 1997). Breheny

describes how people have voted with their feet and as such, residential

preferences are substantially out of alignment with recent

policy objectives which emphasis the need for urban compaction.

Williams et al. (1996) emphasise the importance of public acceptability

and attractiveness in building more compact urban areas.

They contend that it is essential that firstly, urban intensification

brings about some of the advantages put forward in the compact

city debate, such as better services and facilities, better public

transport, and a more vibrant cultural life and secondly, that these

benefits are not outweighed by the negative impacts of compact

city living such as congestion and overcrowding.

Despite the prevalent trend towards urban sprawl, many cities

in recent times (as a result of increasing policy emphasis towards

higher residential densities) have attracted large numbers of residents

back into the urban core. Questions remain, however, relating

to the long term viability and stability of these newresidential environments.

As McEldowney et al. (2005) assert, urban planners and

designers must demonstrate to residents that high-density areas

0264-8377/$ –

 

P. Howley / Land Use Policy 26 (2009) 792–798 793

can be an attractive residential location throughout all stages of

their life cycle. This study focuses specifically on garnering a better

understanding of the residential preferences and behaviour of

residents who have moved into new (built since 1996) relatively

high-density apartment developments in the central area of Dublin

city. This is a group who have made the choice to move into a relatively

compact urban area and hence it will be revealing to assess

the motives, preferences and future intentions of this residential

population. Firstly, this paper by way of context will analyse and

discuss the various explanatory factors put forward by urban theorists

to explain sprawling development patterns evident in the

majority of major urban areas within western societies. Secondly,

this paper will discuss the recent repopulation of many inner city

areas. Thirdly, this paper using results froma household survey will

provide an analysis of the residential preferences and aspirations

of residents in new relatively compact urban environments in the

central area of Dublin city. Finally, this paper will conclude with a

discussion of its major findings and implications for urban policy.

Residential mobility

For the purposes of this paper, it is appropriate to distinguish

between residential mobility and residential migration as well

as between involuntary and voluntary moves. Firstly, residential

mobility is defined as intra-urban moves i.e. relatively short distance

moves within the same region. On the other hand, migration

involves a shift from one labour market to another and hence constitutes

a move from one region to another. There is usually quite

a different set of factors behind residential mobility and migration.

More often than not, employment considerations lie behind

migration decisions whereas as will be described later a variety

of factors such as household and quality of life issues usually lie

behind residential mobility. This paper will focus on residential

mobility which as described by Pacione (2001) accounts for the

vast majority of moves made by individuals within advanced capitalist

societies. Secondly, it is important to distinguish between

voluntary and involuntary moves as involuntary moves account

for a significant proportion of the total number of moves made

by urban residents. In Rossi’s classic study on residential mobility

in the Philadelphia region, discussed later, he demonstrated

that almost a quarter of moves were involuntary including demolition

or change of use of housing properties or eviction of tenants

(Rossi, 1955). Knox and Pinch (2006) describe additional categories

of forced moves, typically arising from the formation and dissolution

of households, through marriage, divorce, death, ill health or

retirement. This paper will focus on explaining the underlying factors

behind voluntary moves as they account for the vast majority

of moves made by individuals and in addition offer the most scope

for policy prescription.

Factors behind urban sprawl

Emerging from the economics literature as a central explanation

for urban sprawl is the monocentric urban model. This model

was developed initially in the 1960s by Alonso (1964), Mills (1967)

and Muth (1969). It has proven to be a useful and popular model as

it provides a framework for empirically measuring and comparing

the degree of centralisation across cities and time periods (Mills and

Lubuele, 1997; Anas et al., 1998, 2000). The key observation from

this model is that “commuting cost differences within an urban

area must be balanced by differences in the price of living space”

(Brueckner, 1987, p. 822). In this model, households located centrally

are the most expensive and therefore individuals live in small

housing units. Consequently, the highest residential densities are

typically found in central areas of the city. Households that live further

from the central area of the city entail higher commuting costs

but are awarded by lower housing costs. The central premise of this

model is that if either incomes rise or transport costs drop then residential

densities will fall as individuals will move further from the

central city.

While economic factors such as rising incomes and falling transport

costs undoubtedly have a role in explaining decentralisation

trends, it would seem likely that increasingly suburbanisation can

be explained in terms of perceived differences in quality of life

between relatively high-density urban areas and lower density

locations outside the city. The movement of people from relatively

built up urban areas to lower density areas is often described as

the flight from blight phenomenon (Mieszkowski and Mills, 1993;

Gaffikin and Morrissey, 1999). Problems such as crime, congestion

and pollution all of which are more prevalent in central cities than

elsewhere, serve to induce central city residents to migrate to the

suburbs. In the past, individuals were often faced with the choice

between an attractive social and natural environment on the one

hand, and economic well-being on the other and therefore, for the

most part, people were constrained to living in cities (Dillman,

1979). Improvements in transport and technology, however, have

minimised perceived distances, allowing the home and workplace

to be more distantly located from each other (Barcus, 2004). Hence,

individuals are now both willing and able to commute long distances

for both employment and facilities in order to satisfy their

residential preferences for what they perceive to be lower density

locations that offer a higher quality of life.

The role of the changing spatial distribution of employment

opportunities is seen as one further significant factor behind urban

sprawl. The outward shift of industries to more peripheral areas is

as a result of changing comparative costs which now increasingly

favour non-central locations (Pacione, 1990). The relative cost of

locating in the outskirts of central areas has fallen for a number

of reasons such as: technological innovation; international competition;

physical plant obsolescence; escalating local production

costs; growth of industries which require a high land to output

ratio; reduction in transportation and communication costs; and

shifting consumption patterns (Cross, 1990; Pacione, 1990). Furthermore,

with modern methods of communication, the need for

face-to-face communication is not as important nowas itwas in the

past (Mills, 1995). Whereas curtailing sprawl is nowa stated objective

of most governments inWestern countries, past policies can be

identified which have had the effect of encouraging the sprawling

development patterns that are observable today. For instance, government

subsidies for road improvements and highway provision

and the neglect of public transport have made the car the mode of

choice for the average citizen (Duany et al., 2000). The planning system

through the mechanism of zoning has further contributed to

urban decentralisation as traditional zoning applications result in

the separation of land-use activities, thus facilitating sprawl (Hall,

1992; Moss, 1997; Duany et al., 2000).

The structure of households is one further factor held to be a

significant predictor of residential preferences for lower density

living. Perhaps the most frequently cited work that focuses on residential

mobility is Peter Rossi’s: Why Families Move (1955). Rossi

focused onmobility patterns in four Philadelphia census tracts and

analysed the causal factors that influence a household’s decision to

move. The study concludedthat family life cycle stagewas themajor

characteristic found to differentiatemobile fromstable households.

Rossi cited space deficits as themost important reasonwhy housing

needs changed through the course of a family’s life cycle as during

the child-rearing stage of a family’s life cycle, household size

and hence density in that household will increase. Therefore, Rossi

viewed the major function of mobility to be the “process by which

front matter © 2008 Elsevier Ltd. All rights reserved.doi:10.1016/j.landusepol.2008.10.004

794 P. Howley / Land Use Policy 26 (2009) 792–798

families adjust their housing to the housing needs that are generated

by the shifts in family composition that accompany life-cycle

changes” (Rossi, 1955, p. 9). The importance of family life cycle stage

as an important determining factor behind residential mobility has

been outlined in further studies such as Chevan (1971), McAuley

and Nutty (1982, 1985) and Clark and Dieleman (1996).

New residential populations within the central city

Despite the dominant trend towards urban sprawl, in recent

times many cities have witnessed a considerable population influx

into its central area. An analysis of census data indicates that

between 1991 and 2002 the population of the central area of Dublin

city increased by 36% (almost 28,000 people). This is in stark contrast

to the decades preceding this period of urban regeneration as

population in the central area of Dublin city fell from 268,851 in

1926 to 76,558 in 1991 (Drudy and Punch, 2000). This was similar

to many other cities throughout Europe and, in particular, North

America at this time, as large numbers of individuals chose to

leave the central area of cities to live in suburban or rural areas

(Mieszkowski and Mills, 1993). There has also been a considerable

population influx into many city centres in the U.K. Bromley et al.

(2007) reports that the population in the central area of Birmingham,

Swansea, Cardiff and Bristol increased by 8%, 13%, 60% and

66% respectively between 1991 and 2002. Madden et al. (2001)

found that the population in Liverpool’s city centre increased by

almost 300% between 1991 and 1999. Additionally, Seo (2002) outlines

how the central city areas of both Glasgow and Manchester

have also attracted large numbers of new residents since the mid

1980s.

The influx of new residents into central city areas has resulted

in some significant demographic and social changes. For example,

the number of individuals aged 20–29 living in the central area of

Dublin city more than doubled from a figure of 16,031 in 1991 to

34,374 in 2002. This is a complete turnaround from the situation

pre-1990 as large numbers of relatively young adults were leaving

the inner city for outlying areas. The numbers of individuals,

for example, aged between 20 and 29 fell by a quarter between

1981 and 1991 alone. Within the U.K, Bromley et al. (2007) reports

that in 2001 as much as 41% and 39% respectively of the city centre

populations of Liverpool and Manchester were aged between

20 and 29. Correspondingly, Nathan and Urwin (2005) report that

between 1991 and 2001 the number of people between 45 and 60

in Manchester more than halved. In terms of household structure

new city centre residents tend to live in small sized non-familial

households. In Dublin the percentage of households within the central

area of the city containing children decreased from a figure of

21.4% in 1991 to just 12.7% in 2002. Likewise, Nathan and Urwin

(2005) report that in Liverpool and Dundee, the proportion of family

households in the city centre is approximately half the city average

and in Manchester, it is around a sixth. One further characteristic

feature of this population increase is the resulting upward social

changes in these central city areas. For instance, thenumber of individuals

belonging to thetwo highest social classes (professional and

managerial occupations) more than doubled within Dublin’s inner

city between 1991 and 2002. Similarly, Madden et al. (2001) and

Davidson and Lees (2005) found that new residents in Liverpool’s

and London’s city centre respectively were also weighted towards

managerial and professional occupations. The upward social and

physical changes accruing from the population influx into inner

urban areas have been termed as the gentrification process by,

among others, Davidson and Lees (2005) and Smith and Butler

(2007). In effect whatwas until recently derelict land in the central

city are nowprestigious residential areas housing a newpopulation

Table 1

How likely are you to move residence in the next 5 years?

Frequency Percent

Very unlikely 20 7.5

Unlikely 20 7.4

Likely 63 23.3

Very likely 146 54

Do not know 21 7.8

Total 270 100

living in small-sized non-familial households with the residents

themselves typically much younger and affluent than members

of more established communities. In addition to British and Irish

cities the influx of new young urban professionals back into the

central city has been widely reported in many other European and

North American cities such as Toronto (Bourne, 1991), Amsterdam

(Karsten, 2003) and New York (Lees, 2003).

Research design

The case study for this paper is Dublin’s central city, which since

the early 1990s has attracted large numbers of residents back into

the urban core. Originally, the emphasis on promoting urban and in

particular inner urban living had been motivated by a policy objective

of physical renewal and the reuse of derelict land. However, in

more recent times, the focus on central city renewal can be increasingly

viewed as a central tool in managing a growing number of

households and in delivering more sustainable patterns of urban

development. In relation to the overall research methodology, a

list of apartment developments built in the central area of Dublin

city between 1996 and 2006 were identified through fieldwork.1 A

total of 50 apartment developments were subsequently randomly

chosen to be surveyed. A questionnaire was then distributed by

post to 1050 of these apartment units in the summer of 2006 with

a follow up reminder letter distributed 1 week later. A total of

270 completed questionnaires were returned. The survey gathered

information relating to the personal characteristics of respondents,

their residential preferences and intended futuremobility patterns

and their satisfaction with both dwelling unit and various features

of the neighbourhood.

Results

Residential preferences for lower density living

Residents living in these new residential environments in the

central area of the city are predominantly young, affluent and

at the early stage of their life cycle. For instance, 80% of survey

respondents are between the ages of 20 and 39 and almost 77%

of respondents report a primary degree or higher as their highest

level of education achieved. In terms of household structure, 29%,

31% and 29% respectively of households surveyed consist of either a

single person living alone, a couple without children or non-family

households (single individuals living together). Eleven percent of

households contained children (ether a couple with children or a

single parent with children). In relation to future mobility patterns,

respondents were asked how likely or unlikely it is that they will

move residence in the next 5 years and the results are presented in

1 The author would wish to thank Sinead Kelly and Andrew MacLaran from the

Centre of Urban and Regional Studies, Trinity College, Dublin for kindly providing

addresses of new apartment developments built in the central area of Dublin city,

which provided the basis for the sampling frame in this study.

P. Howley / Land Use Policy 26 (2009) 792–798 795

Table 2

Future dwelling type.

Frequency Percent

Detached 118 45.2

Semi-detached 41 15.7

Terraced 33 12.6

Apartment 64 24.5

Do not know 5 1.9

Total 261 100

Table 1. In the next 5 years over half of respondents (54%) feel that it

is very likely that they willmove property. Combining the figures for

likely and very likely in both categories means that just over 77.3%

of respondents feel that it is either likely or very likely that they

will move residence in the next 5 years. In terms of preferences for

various types of accommodation, just under half (45.2%) of respondents

who are all currently living in an apartment feel they will be

living in a detached house in the future (see Table 2). A further 12.6%

and 15.7% of respondents believe they will be living in a terraced or

semi-detached house respectively. One third of respondents envisage

themselves as living in either the open countryside, a village,

small town or a large town in the future (see Table 3). Thirty-nine

percent of respondents still see themselves as living in an Irish city

in the future, but given that the vast majority of respondents do not

envisage living in an apartment then it seems likely that for a large

number of these respondents this location will be in lower density

locations outside the central area of the city. To sum up, evidence

would suggest that for the majority of residents in these areas their

residential preferences appear to be weighted towards ultimately

living in lower density areas outside the central area of the city.

Benefits and limitations of urban living

To provide a more detailed understanding of the views of individuals

living in these newresidential environments towards urban

living, respondents were asked what they felt was the main benefit

and limitation of living in urban areas through open ended

questions (Tables 4 and 5). Firstly, in relation to respondents perceptions

of urban living it can be seen that accessibility and related

issues are cited as the most common benefit of urban living. That

is, 65% of respondents stated that either accessibility, access to

services and facilities, access to amenities or access to work was

the main benefit of urban living. One further significant response

was employment opportunities as this was responsible for almost

13% of all responses. Various social aspects of urban living such as

social life and cultural activities were reported by 8.3% of residents

as the main benefit of urban living. Travel related considerations

such as less commuting and better transport/public transportwere

reported by 2.9% and 3.7% of respondents respectively as the main

benefit of urban living.

As regards the perceived limitations of urban living, the expense

of living in urban areas such as the cost of living and cost of hous-

Table 3

What type of area do you think you will be living in, in the future?

Frequency Percent

The open countryside 17 6.3

A village 12 4.5

A small town 26 9.7

A large town 34 12.6

A city 106 39.4

Another country 41 15.2

Do not know 33 12.3

Total 269 100

Table 4

Main benefit of urban living.

Frequency Percent

Near everything/accessibility 63 26

Access to services and facilities 44 18.2

Access to amenities 29 12

Access to work 21 8.7

Employment opportunities 31 12.8

Social life 14 5.8

Cultural activities 6 2.5

Transport/public transport 9 3.7

Less commuting 7 2.9

Other 18 7.4

Total 242 100

Table 5

Main limitation of urban living.

Frequency Percent

Cost of living/cost of housing 50 22.8

Lack of space/too crowded 40 18.3

Traffic congestion/poor public transport 37 16.9

Noise 23 10.5

Pollution 14 6.4

Crime 10 4.6

Unsuitable to bring up children 9 4.1

Lack of community spirit/community involvement 8 3.6

Pace of life 6 2.7

Poor housing/apartment design 3 1.4

Poor quality environment 2 .9

Other 17 7.8

Total 219 100

ing seemed of greatest concern to respondents, as 22.8% of residents

feel that this is the main limitation of urban living. Space related

considerations such as lack of space/open space or too crowded

were cited by 18.3% of respondents. Issues surrounding transport is

one further prominent limitation cited by respondents with some

16.9% of respondents stating traffic congestion/poor public transport

as the main limitation of urban living. Noise and pollution

accounted for 10.5% and 6.4% of the total number of responses

respectively. Four and a half, 4.1% and 3.6% of respondents reported

that crime, unsuitability to bring up children or lack of community

spirit was the main limitation of urban living.

Logistic regression model

Selection of variables

A logistic regression model was formulated to provide a deeper

understanding of what were the major factors behind the majority

of respondents future intentions to relocate in the short tomedium

term (see Greene, 1997; Long, 1997; Long and Freese, 2006 for a

more detailed description of this type of estimation). More precisely,

this model was designed in order to look at the ability of

individuals’ stage in their life course, income, commuting time to

work, satisfaction with both dwelling unit and various attributes

of the neighbourhood, in predicting their intended future mobility

patterns. The dependent variable chosen for the analysis was

respondents’ opinion on how likely or unlikely it is that they will

move residence in the next 5 years (see Table 1 as discussed earlier).

In particular, this variable was split into two categories, namely

respondentswho feel it is very likely that they will move residence

and respondents who feel it is less than very likely that they will

move residence in the next 5 years. Given the ordinal nature of the

dependent variable an alternative to the binary logistic regression

model usedherewould have been to use ordered/logistic regression

as this would have avoided the arbitrary grouping of the depen

796 P. Howley / Land Use Policy 26 (2009) 792–798

Table 6

Description of variables in logistic model.

Dependent variable

Residential mobility: Compares the effect of feeling it is very likely to

move residence in the next 5 years (56% of respondents) with those

that feel it is less than very likely to move residence (44% of

respondents).

Explanatory variables

Age: Compares the effect of being under 30 (41% of respondents) on

an individual’s probability of moving in the next 5 years with that of

being 30 or over (59% of respondents).

Household size: Compares the effect of being in a multi occupancy

dwelling (61% of respondents) on an individual’s probability of moving

with that of being in a single person household (29% of respondents).

Income: Compares the effect of having an income greater than

D 50,000 (36% of respondents) on an individual’s probability of moving

in the next 5 years with that of having less than D 50,000 (64% of

respondents).

Satisfaction with apartment: Compares the effect of being less than

very satisfied (either quite satisfied, somewhat unsatisfied or very

unsatisfied) with apartment (69% of respondents) on an individual’s

probability of moving in the next 5 years with that of being very

satisfied with their apartment (31% of respondents).

Time to work: Compares the effect of spending 30 min or more

travelling to work (39% of respondents) on an individual’s probability

of moving in the next 5 years with that of spending less than 30 min

(61% of respondents).

Neighbours: Compares the effect of speaking to neighbours less than

twice a month (51% of respondents) on an individual’s probability of

moving in the next 5 years with that of speaking to neighbours more

often (49%).

Pollution: Compares the effect of perceiving pollution to be a very big

or fairly big problem in the area (55% of respondents) on an

individual’s probability of moving in the next 5 years with that of

perceiving pollution to be less than a fairly big problem (minor

problem, not a problem or no opinion) (45% of respondents).

dent variable into two categories.Abinary logistic regression model

was ultimately chosen, however, for two reasons; firstly the split

between very likely and less than very likely divides the population

into two roughly equal groups (54% and 46%, respectively)

and secondly this study aimed to gather information relating to the

respondents most likely to move residence and it was felt that a

large share of respondents who report that it is likely that theywill

move residence may not actually do so.

For ease of use in the model it was necessary to convert the

coding of each variable into binary form. For instance, the variable

‘satisfaction with apartment’ which asked respondents whether

they were very satisfied, quite satisfied, somewhat unsatisfied or

very unsatisfied with their apartmentwas split into two categories

namely those who felt very satisfied and those who felt less than

very satisfied. As there are only two categories to each variable, the

odds ratio for one category only is shown as the other categorywill

have an opposite effect of equal significance. The odds ratios for the

coefficients of the explanatory variables allows us to see the likelihood

that this variable category is associated with being very likely

to move residence in the next 5 years rather than less than very

likely. The odds ratio is a way of comparing whether the probability

of a certain event is the same for two groups. For example, the

odds ratio of 2.24 for satisfaction with apartment (see Table 7) indicates

that the group of individuals who are less than very satisfied

with their apartment are over twice as likely to feel that it is very

likely that they will move residence in the next 5 years than the

group of individuals who are very satisfied with their apartment.

An odds ratio of 1 herewould imply that both groups of individuals

are equally likely to feel that it is very likely rather than less than

very likely that they will move residence in the next 5 years. Table 6

provides more detail on the classifications used for each variable.

The variables usedwere firstly the age of the respondent (‘age’), and

whether the respondent lived in a single person household (‘house-

Table 7

Logistic model of residential mobility.

Independent variables Odds ratio Significance

Age* 1.888409 0.06

Household size* 1.862219 0.063

Satisfaction with apartment** 2.243099 0.009

Neighbours** 2.273754 0.005

Pollution* 1.629243 0.097

Income 1.027484 0.934

Commute time 1.06508 0.837

* Significant at the 90% level.

** Significant at the 99% level.

hold size’) both of which were included to test the importance of

life-cycle stage in respondents futuremobility intentions. Our a priori

expectation here would be that relatively younger individuals

and those living in larger householdswould have a higher probability

of moving residence in the next 5 years. In addition, an income

variable was included in the analysis as numerous studies have

highlighted the importance of income as a significant factor behind

residential mobility. In relation to the neighbourhood, our a priori

expectation is that thosewhoaremost dissatisfiedwith their neighbourhood

would express a greater desire to move residence in the

next 5 years.Two variableswere chosen to be included in this model

to represent different features of the neighbourhood, namely the

level of interaction with neighbours (‘neighbours’) chosen to represent

overall community spirit or social features and perception

of the level of pollution (‘pollution’) chosen as a proxy for environmental

quality. Similarly to satisfaction with the neighbourhood,

respondents satisfaction with theirdwelling unit (‘satisfaction with

apartment’) was one further variable selected for analysis as it was

felt that dissatisfaction with their apartment could prompt individuals

to move in order to find more a suitable dwelling that

could improve their quality of life. Finally, it was thought useful

to examine whether accessibility to work (‘commuting time’) was

an important factor behind respondents intended future mobility

decisions in these areas.

Model results

The variables ‘age’, ‘household structure’, ‘satisfaction with

apartment’, ‘neighbours’ and ‘pollution’, were significant at the

90% significance level or better (see Table 7). There was no statistically

significant difference found in the probabilities of moving

residence in the next 5 years between respondents earning more

than D 50,000 per annum with those earning less than D 50,000

per annum suggesting that income is not a significant barrier for

this residential population when determining their futuremobility

behaviour. Additionally, it appears commuting time towork is not a

decisive factor for this residential population as therewas no difference

found in the probabilities of moving residence between those

who spend 30 min or less commuting towork and those that spend

over 30 min. The variable with the largest statistically significant

odds ratio was ‘neighbours’. Those that speak to their neighbours

less than twice a month were 2.3 times more likely than respondentswho

speak to their neighbours more often to report that they

were very likely rather than less than very likely tomove residence

in the next 5 years. Respondents’ perception on the level of pollution

in the area is one further important neighbourhood feature to

influence residents’ future mobility intentions. Respondents who

perceive pollution to be a very big or fairly big problem in the area

are 1.6 times more likely to feel that they are very likely rather than

less than very likely to move residence in the next 5 years. Individuals

who are less than very satisfied with their apartment were

2.2 times more likely to report that they were very likely rather

than less than very likely tomove residence in the next 5 years. The

P. Howley / Land Use Policy 26 (2009) 792–798 797

age of respondents was one further important predictor of respondents’

future mobility intentions. Respondents who are 29 years

of age or younger were almost twice (1.9) as likely as respondents

who are over 29 years of age to feel that they are very likely rather

than less than very likely to move residence in the next 5 years.

Finally, ‘household size’ also has an important influence on individuals’

future mobility intentions. Individuals who live in multi

person households are almost twice (1.9) as likely than respondents

living in single person households to feel that they are very

likely rather than less than very likely tomove residence in the next

5 years.

Discussion

Compact city policies are designed tomeet the primary sustainability

aims of reducing car dependence and hence pollution, and

minimising the loss of open countryside and habitats to development

(Haughton and Hunter, 2003; Gillham, 2002). The onus on the

planning systems of most advanced capitalist societies to develop

a more sustainable urban development pattern has resulted in an

ever-increasing emphasis on policies to increase residential densities.

For example, the European Spatial Development Perspective

(ESDP) (CEC, 1999) strongly encouragesmember states and regional

authorities to pursue the concept of the ‘compact city’ (or the city

of short distances) as a method to control the physical expansion

of cities, integrating land-use and transport more effectively,

and reducing the physical separation of daily activities. Residential

preferences appear to be at variance with this policy agenda

as an ever-increasing share of individuals are residing in lower

density locations outside the central area of cities. This is by no

means an Irish phenomenon, as declining residential densities in

the central area of cities coupled with population expansion on

the urban fringe has been symptomatic of most of the world’s

advanced capitalist societies for the greater part of the last century

(Mieszkowski and Mills, 1993). Despite the general trend towards

urban sprawl, in recent times many cities have been successful in

attracting large numbers of residents back into the central area.

While this is likely to be a good starting point in urban consolidation

efforts, questions still remain relating to the liveability implications

of such developments. For instance, this paper demonstrates

that the residential preferences of individuals living in new relatively

high-density residential environments in the central area

of Dublin city are weighted towards (ultimately) living in lower

density areas. This raises a number of questions relating to developing

more “sustainable” patterns of urban development as if these

residential preferences manifest themselves into actual behaviour

then this will result in further developmental pressures on rural

and suburban areas. While policymakers stress the need for compaction,

current residential preferences if left unchecked will lead

to further sprawling development patterns. In this context, policymakers

need to consider not just the environmental impacts of the

residential communities they are building but also the wider liveability

implications of high-density living. Respondents perceive

the main benefit of urban living to be greater accessibility levels

and better social life, whereas the main limitation relate to cost of

living, lack of space, transport related issues, noise and pollution.

This paper formulated a logistic regression model in order to

develop a deeper understanding of the factors behind respondents’

future mobility intentions to locate in lower density areas.

Significant factors to emerge from the analysis included firstly,

family life cycle stage, as younger respondents and those living

in multi occupancy dwellings have a much higher probability of

moving in the next 5 years than relatively older respondents and

those living by themselves. Overall satisfaction with dwelling unit

and aspects of the neighbourhood such as neighbourly interaction

(proxy for social features of the neighbourhood) and perceptions

on the level of pollution (proxy for overall environmental quality)

in the area were further features to emerge as significant influences

on residential mobility. This would suggest a significant role

for urban planners and designers as improvements in the design

of the dwelling unit and the neighbourhood as well as providing

residential environments that are suitable throughout all stages of

an individuals’ life cycle can improve the stability and quality of life

experienced in these areas.

Acknowledgements

The author would wish to thank the Environmental Protection

Agency’s ERDTI Programme (funded from the National Development

Plan) for funding and supporting this research project.

Additionally, the authorwould wish to thank Mark Scott and Declan

Redmond from the School of Geography, Planning and Environmental

Policy and Emma Dillon from the Rural Economy Research

Centre, Teagasc for helpful comments and suggestions.

References

Alonso,W., 1964. Location and Land Use. Harvard University Press, Cambridge.

Anas, A., Arnott, R., Small, K.A., 1998. Urban spatial structure. Journal of Economic

Literature 36, 1426–1464.

Anas, A., Arnott, R., Small, K.A., 2000. The panexponential monocentric model. Journal

of urban Economics 47, 165–179.

Banister, D., 1997.Reducing the need to travel. Environment and Planning B: Planning

and Design 24, 437–449.

Banister, D., Watson, S., Wood, C., 1997. Sustainable cities: transport, energy, and

urban form. Environment and Planning B: Planning and Design 24, 125–143.

Barcus, H.R., 2004. Urban-Rural migration in the USA: an analysis of residential

satisfaction. Regional Studies 38, 643–657.

Bourne, L.S., 1991. The Roepke lecture in economic geography recycling urban systems

and metropolitan areas: a geographical agenda for the 1990s and beyond.

Economic Geography 67, 185–209.

Breheny, M., 1992. The contradictions of the compact city: a review. In: Breheny,

M. (Ed.), Sustainable Development and Urban Form. Pion Limited, London, pp.

138–159.

Breheny, M.,1995. Thecompact city and transport energyconsumption. Transactions

of the Institute of British Geographers, New Series 20, 81–101.

Breheny, M., 1997. Urban compaction: feasible and acceptable? Cities 14, 209–217.

Bromley, R.D.F., Tallon, A.R., Roberts, A.J., 2007. New populations in the British city

centre: evidence of social change from the census and household surveys. Geoforum

38, 138–154.

Brueckner, J.K., 1987. The structure of urban equilibria: a unified treatment of the

Muth Mills Model. In: Mills, E.S. (Ed.), Handbook of Urban and Regional Economics,

vol. 2. North Holland/Elsevier Publishers, Amsterdam, pp. 822–845.

Commission of the European Communities, 1999. European Spatial Development

Perspective: Towards Balanced and Sustainable Development of the Territory of

the EU. Office for Official Publications of the European Communities, Luxembourg.

Chevan, A., 1971. Family growth, household density, and moving. Demography 8,

451–458.

Clark,W.A.V., Dieleman, F.M., 1996. Households and Housing: Choice and Outcomes

in the Housing Market. CUPR Press, Rutgers University, New Jersey.

Cross, D.F.W., 1990. Counterurbanization in England andWales. Avebury Publishers,

Aldershot.

Davidson, M., Lees, L., 2005. New-build gentrification and London’s riverside renaissance.

Environment and Planning A 37, 1165–1190.

Dillman, D.A., 1979. Residential preferences, quality of life, and the population

turnaround. American Journal of Agricultural Economics 61, 960–966.

Drudy, P.J., Punch, M., 2000. Economic restructuring, urban change and regeneration:

the case of Dublin. Journal of the Statistical and Social Inquiry Society of Ireland

29, 155–212.

Duany, A., Plater-Zyberk, E., Speck, J., 2000. Suburban Nation. North Point Press, New

York.

Gaffikin, F., Morrissey, M., 1999. Urban regeneration: the new policy agenda. In:

Gaffikin, F., Morrissey, M. (Eds.), City Visions. Pluto Press, London, pp. 116–151.

Gillham, O., 2002. The Limitless City: A Primer on the Urban Sprawl Debate. Island

Press,Washington, DC.

Greene,W.H., 1997. Econometric Analysis. Prentice Hall, New York.

Hall, P., 1992. Urban and Regional Planning. Routledge, London.

Haughton, G., Hunter, C., 2003. Sustainable Cities. Routledge, London.

Heath, T., 2001. Revitalizing cities, attitudes towards city-centre living in the United

Kingdom. Journal of Planning Education and Research 20, 464–475.

798 P. Howley / Land Use Policy 26 (2009) 792–798

Karsten, L., 2003. Family gentrifers: challenging the city as a place simultaneously

to build a career and to raise children. Urban Studies 40, 2573–2584.

Knox, P., Pinch, S., 2006. Urban Social Geography: An Introduction, 5th ed. Prentice

Hall, New York.

Lees, L., 2003. Super-gentrification: the case of Brooklyn Heights, New York City.

Urban Studies 40, 2487–2509.

Long, J.S., 1997. Regression Models for Categorical and Limited Dependent Variables.

Sage Publications, Thousands Oaks, CA.

Long, J.S., Freese, J., 2006. Regression Models for Categorical Dependent Variables

using Stata. Stata Press, Texas.

Madden, M., Popplewell, V., Wray, I. 2001. City centre living as the springboard for

regeneration? Some lessons from Liverpool. University of Liverpool Department

of Civic Design, Liverpool.Working Paper 59.

McAuley,W.J.,Nutty, C., 1982. Residential preferences andmoving behavior: a family

life-cycle analysis. Journal of Marriage and the Family 44, 301–309.

McAuley,W.J., Nutty, C.J., 1985. Residential satisfaction, community integration, and

risk across the family life cycle. Journal of Marriage and the Family 47, 125–130.

McEldowney, M., Ryley, T., Scott, M., Smyth, A., 2005. Integrating land-use planning

and transportation in Belfast: a newpolicy agenda for sustainable development?

Journal of Environmental Planning and management 48, 507–526.

Mieszkowski, P., Mills, E.S., 1993. The causes of metropolitan suburbanization. Journal

of Economic Perspectives 7, 135–147.

Mills, E.S., 1967. An aggregative model of resource allocation in ametropolitan area.

American Economic Review 57, 197–210.

Mills, E.S., 1995. Crisis and recovery in office markets. Journal of Real Estate Finance

and Economics 10, 49–62.

Mills, E.S., Lubuele, L.S., 1997. Inner cities. Journal of Economic Literature 35,

727–756.

Moss, M., 1997. Reinventing the central city as a place to live and work. Housing

Policy Debate 8, 471–490.

Muth, R., 1969. Cities and Housing. University of Chicago Press, Chicago.

Nathan, M., Urwin, C., 2005. City People: City Centre Living in the UK. Centre for

Cities, London.

Neuman, M., 2005. The compact city fallacy. Journal of Planning Education and

Research 25, 11–26.

Newman, P., Kenworthy, J., 1989. Cities and Automobile Dependence: An International

Sourcebook. Gower Publishing Company, Aldershot, UK.

Pacione, M., 1990. Urban Problems: An Applied Urban Analysis. Routledge, London.

Pacione, M., 2001. Urban Geography. A Global Perspective. Routledge, New York.

Rossi, P.H., 1955. Why Families Move. Macmillan, Aldershot, UK.

Senior, M.L.,Webster, C.J., Blank, N.E., 2004. Residential preferences versus sustainable

cities: quantitative and qualitative evidence from a survey of relocating

owner-occupiers. Town Planning Review 75, 337–358.

Seo, J.K., 2002. Reurbanisation in regenerated areas of Manchester and Glasgow:

new residents and the problems of sustainability. Cities 19, 113–

121.

Smith, D., Butler, T., 2007. Conceptualising the socio-spatial diversity of gentrification:

‘to boldly go’ into contemporary gentrified spaces, the final frontier.

Environment and Planning A 39, 2–9.

Williams, K., Burton, E., Jenks, M., 1996. Achieving the compact city through intensification:

an acceptable option? In: Jenks, M., Burton, E., Williams, K. (Eds.), The

Compact City: A Sustainable Urban Form? E&FN Spon, London.