Journal of Retailing and Consumer Services 76 (2024) 103591

Contents lists available at ScienceDirect

Journal of Retailing and Consumer Services

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

Retail rings of influence: Rethinking retail catchment analysis in the light of

purchasing power
Chung Yim Yiu a, *, Quan Le Truong b, Jiaying Kou c, Ka Shing Cheung a
a
Department of Property, The University of Auckland, Auckland, New Zealand
b
Insight Chapter, Air New Zealand, New Zealand
c
Australian Urban Research Infrastructure Network (AURIN) and Faculty of Architecture, Building & Planning, The University of Melbourne, Australia

A R T I C L E I N F O A B S T R A C T

Handling Editor: Prof. H. Timmermans Catchment areas are vital indicators of purchasing power that significantly influence the retail sales of shopping
malls. However, conventional models, typically provided by mall owners, often assume a fixed isochrone,
Keywords: neglecting the sensitivity of the catchment size on mall’s positioning strategies and sizes. This study introduces a
Shopping centres novel, flexible approach, leveraging the population and household incomes of administrative units as a proxy for
Catchment area
purchasing power, thereby enabling a sensitivity analysis on adjustable catchment sizes. We develop a Python
Species-energy relationship
programme to geofence these units based on an adjustable isochrone radius. This pioneering research also applies
Niche specialisation
Isochrone principles from the species-energy relationship and niche specialisation theory to build a theoretical model of
catchment areas, empirically testing it using Australian data. Our findings confirm that while purchasing power
within varying catchment sizes exerts a positive influence on retail sales, the impact is less pronounced for larger
malls due to their more generalised niche in attracting customers far away from the malls. This study sheds new
light on the dynamic interplay between catchment area characteristics and mall performance, offering valuable
insights for future research and retail strategy planning.

1. Introduction position in a hierarchy of other retail centres.

In theory, the purchasing power and demographics of households
A retail catchment is defined as the geographic area that encom­ within a catchment area determine the retail sales volume of shopping
passes the primary patrons of a store or retail centre (Dolega et al., centres, as they indicate the strength of retail demand. However, past
2016). The concept of the retail catchment has inspired extensive aca­ studies have generally assumed a fixed radial distance — for example, a
demic research and commercial studies centred on predicting retail store 30-min drive — from a shopping centre as its catchment area. Such an
sales (Dennis et al., 2002) and determining optimal locations for new assumption can be misleading, as the size of a catchment area depends
stores and shopping centres (Joseph and Kuby, 2011). The existing on the type and size of the shopping centre. Moreover, empirical evi­
literature delves into retail catchments for single stores or shopping dence has been lacking to confirm catchment sizes. This study is one
centres, including a review of the Gravity Models on the supply side attempt to conduct a sensitivity test on catchment sizes to evaluate the
factors (Joseph and Kuby, 2011), an entropy-maximising model incor­ effect of purchasing power on malls of varying sizes. In defining a retail
porating the demand side factors (Birkin et al., 2010), and uncertainty catchment, a common approach is to establish a certain threshold for the
analysis on the catchment areas (Rasouli and Timmermans, 2013). geographic extent, representing the proportion of customers likely to
While the general concept of a retail catchment encompasses three patronise a specific store or retail centre. These threshold values, also
major components: demand factors, supply factors, and consumer in­ known as primary, secondary, or tertiary catchments, represent the
teractions (Birkin et al., 2010), a network of retail centres requires a geographic distribution of at least 70%, 80%, and 95% of a particular
more thorough examination of additional crucial dimensions and con­ centre’s shoppers, respectively (Lloyd and Cheshire, 2017). Conse­
straints (Dennis et al., 2002). One such factor is the purchasing power of quently, shopping centres are often classified into a hierarchy, as illus­
the household demographics within a retail centre, considering its trated by the Property Council of Australia’s classification in Table 1,

* Corresponding author. University of Auckland Business School, New Zealand.

E-mail addresses: [email protected] (C.Y. Yiu), [email protected] (Q.L. Truong), [email protected] (J. Kou), william.cheung@
auckland.ac.nz (K.S. Cheung).

https://doi.org/10.1016/j.jretconser.2023.103591
Received 9 July 2023; Received in revised form 1 September 2023; Accepted 3 October 2023
Available online 7 October 2023
0969-6989/© 2023 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
C.Y. Yiu et al. Journal of Retailing and Consumer Services 76 (2024) 103591

Table 1 ensuing discussion. In Section 5, we encapsulate the findings and im­

Shopping centre classifications by the property council of Australia. plications of the study.
Classifications Descriptions Total Gross Lettable
Area (GLA) in sqm. 2. Literature review
Super regional The largest shopping centres Exceeds 85,000
centre typically incorporate two full-line Retail catchment is a significant area of interest for both scholars and
department stores, one or more full- practitioners, as it impacts the market share and competitive advantage
line discount department stores, two of a retail centre (Drezner and Drezner, 2002). Traditionally, the central
supermarkets and over 250
speciality shops
place theory has modelled the functional roles of differently sized cen­
Major regional typically incorporating at least one 50,000–85,000 tres (Christaller, 1933). The characteristics of the catchment area are
centre full-line department store, one or primarily shaped by three factors: demand, supply, and consumer
more full-line discount department interaction (Birkin et al., 2010). The demand side is dictated by popu­
stores, one or more supermarkets
lation size and distance, regarded as the determinants of a catchment
and around 150 speciality shops
Regional centre typically incorporating at least one 30,000–50,000 area’s retail sales. In contrast, the characteristics of a shopping mall or
full-line department store, one full- store significantly influence its catchment area on the supply side. The
line discount department store, one share of customers a retail outlet attracts from a region is represented by
or more supermarkets and around the entropy-maximising model, which is inversely proportional to the
100 or more speciality shops
Sub-regional A medium-sized shopping centre, 10,000–30,000
distance and directly proportional to the outlet’s attractiveness.
centre typically incorporating at least one Historically, shopping malls have been classified hierarchically
full-line department store, a major based on location and accessibility, as these are viewed as primary de­
discount department store, a major terminants of their catchment areas (Dennis et al., 2002; Teller and
supermarket and around 40 or more
Reutterer, 2008). Guy (1998) categorised shopping centres based on the
speciality shops
Neighbourhood A local shopping centre comprising a Less than 10,000 size of their catchment area, with tiers ranging from a central area
centre supermarket and up to 35 speciality covering an entire town to a local centre encompassing immediate
shops neighbouring centres with a population of approximately 500-5000.
City centre Retail premises within an arcade or Exceeds 1000 Accessibility has been then linked to catchment areas in earlier
mall development owned by one
company, firm or person and
research. For example, Tanwar et al. (2011) included shoppers within a
promoted as an entity within a major 3 km radius of the shopping mall as part of the catchment area, dividing
CBD this 3 km area further into primary, secondary, and tertiary trade areas
Source: Australia Property Council, adapted from Green (2019).
based on distance from the mall. However, Guy (1998) argued that this
definition of the catchment area is vague, as its size also depends on the
type and size of the mall. Several studies confirmed that location of
which ranges from the smallest city centre of about 1000 sqm to the
shopping malls significantly influences customers’ visitation decisions
super regional centre of more than 85,000 sqm in total gross lettable
(Thomas and Pathak, 2012). Ertekin et al. (2008) also suggested that a
area. Super regional centres or metropolitan malls can attract shoppers
mall’s accessibility and location dictate its size and positioning strategy,
from a wide area, whereas city or district malls primarily attract nearby
as well as the size of its catchment area.
households. Thus, a mall’s catchment area is typically defined as a cir­
For instance, inner-city shopping malls, usually major regional
cular area surrounding the mall, measured in terms of travel distance or
(metropolitan) malls, tend to have larger catchment areas than those in
time.
peripheral areas due to their increased accessibility to shoppers. Other
Building on Christaller’s (1933) central place theory, it is widely
studies also provide evidence that the size of the mall and the size of the
accepted that shopping malls of various sizes accommodate different
catchment area are strongly correlated with accessibility (Cervero and
tenant mixes and facilities. These variables determine the array of goods
Kang, 2011; Deng and Nelson, 2010; Higgins and Kanaroglou, 2018;
and services—from low-order to high-order—that the malls can offer.
Lam and Chau, 2012; Rao and Pafka, 2021). However, these studies did
The type of goods and services then dictate the extent of the catchment
not consider supply-side factors, particularly the positioning and char­
area (threshold) from which the malls can draw their customers. In other
acteristics of shopping malls, in shaping the catchment area. This study
words, the catchment range depends on the type of shopping mall,
seeks to address this gap in catchment area literature.
which, in turn, depends on the purchasing power within the catchment
range.
2.1. Estimating retail catchment
Xu et al. (2022) propose a model akin to the Island-Species-Area
Energy Relationship (ISAER) in island biogeography theory, wherein
The conventional industry approach to retail catchment measure­
the purchasing power of a shopping mall’s catchment area symbolises
ment involves delineating fixed buffer rings around a store or creating
the habitat energy level that sustains the biodiversity of the habitat.
polygons based on the distance or travel time customers are typically
Geographical range size is a key concept in ecology and biogeography
willing to commit to reaching a particular centre. This can be guided by
that has been the focus of intensive study in recent decades. Scholars
factors such as walking distance or travel time (Segal, 1999; Taniar and
have employed niche positions and specialisation to account for varia­
Rahayu, 2015; Tanwar et al., 2011; Adhinugraha et al., 2022). However,
tions in geographical range size (Gregory and Gaston, 2000; Mason
Guy (2002) posited that stores of a larger scale often command wider
et al., 2008). A meta-analysis by Slatyer et al. (2013) discovered that
catchment areas. Such stores, bolstered by a wider variety of offerings
species with wider niche breadth tend to have a larger range size, while
and greater customer appeal, expand the breadth of their retail ag­
species with specialised niches and smaller body sizes can prosper
glomeration’s catchment area. Bilková et al. (2016) attributed the suc­
within a smaller habitat range and/or at lower habitat energy levels
cess of their top-performing local mall to its broader range of services
(Böhm et al., 2017). As such, this paper posits that catchment range size
and products relative to competing shopping centres, enabling it to draw
hinges on the niche breadth of shopping malls.
customers from a more extensive catchment area. Teller (2008) main­
The structure of this study is as follows. In Section 2, we examine the
tained that shopping centres boasting major anchor tenants typically
literature on the catchment areas of shopping malls. In Section 3, we
have significantly wider catchment areas compared to those without
discuss the development of the hypothesis, research design, and data
such attractions.
pertinent to this study, while in Section 4, we present the results and the
Traditionally, purchasing power is regarded as a demand-side factor

2
C.Y. Yiu et al. Journal of Retailing and Consumer Services 76 (2024) 103591

within a fixed catchment area, with the assumed purchasing power 3. Research design and data
within this area perceived as determining the demand for retail services.
However, this model tends to overlook the supply side. For instance, 3.1. An ecological perspective of the catchment area
consumers might be willing to travel considerable distances to secure
higher-order goods and services, while they may be less inclined to do so There have been very few theoretical studies on the relationship
for less expensive items. Therefore, retail performance isn’t solely between catchment areas and retail sales of shopping malls. The pur­
determined by purchasing power but also by whether the size and type chasing power of the catchment area is either ignored (Eaton and Lipsey,
of the mall can accommodate different tiers of goods and services. This, 1982) or taken for granted that it represents one of the demand side
in turn, influences the catchment area size, alongside other factors like determinants to the sales of shopping malls. Most previous studies
accessibility and location. Elements such as size, type, accessibility, and conceptually assumed that the catchment range size is known and fixed
location are intertwined in shopping mall positioning, given that without theoretical justifications or empirical evidence. This study is
metropolitan malls are typically large and sited in highly accessible one of the earliest attempts to theorise the relationship between catch­
areas. This is necessitated by the need for a certain customer base ment area and retail sales with reference to the species-energy theory
threshold to sustain the mall’s size and high rent costs. and the niche specialisation theory from ecology. Carrara and Vàzquez
While demographic factors like population size and household in­ (2010) generalised Wright’s (1983) species-energy relationship by
comes within catchment areas significantly influence shoppers’ behav­ incorporating both total energy and energy variation that determine
iour and retail performance (Kuruvilla and Joshi, 2010), the extent to total species richness as in Equation (1).
which malls can attract customers largely depends on their size and ( )z
E
positioning strategy (Rosouli and Timmermans, 2013). This relationship S=k (1)
eτ
has seen scant empirical examination, likely due to difficulty pinpoint­
ing the catchment size. This study seeks to fill this gap by conducting a Taking natural logarithm, it becomes:
sensitivity analysis on the impact of purchasing power across various
catchment sizes on the retail performance of different sized shopping
ln(S) = ln(k) + zln(E) − zln(eτ ) (2)
malls.
where the species richness S and the ratio of total energy available to
In this study, we leverage the traditional approach to pinpoint the
energy variability eEτ is linked by a power law, with k and z being the
purchasing power within catchment areas. Initially, the size of the
catchment area is assumed to be fixed based on the distance from the constants. Mason et al. (2008) further contended that the species
mall. The purchasing power within the area is estimated by the total richness-energy relations are shaped by the niche information. In our
income of the population living within this zone (Birkin et al., 2010). empirical models, we consider retail sales amount as analogous to spe­
Several previous studies have confirmed a positive correlation between cies richness S and the total disposable income, which is the product of
the purchasing power of catchment areas and the retail performance of the total number of households Pi,t and the median household income Ii,t ,
shopping malls. For instance, population density has been shown to within the catchment area of mall i at time t as the total energy available,
determine potential retail performance (Morrison and Abrahamse, Eit . Other supply factors from the characteristics of shopping malls are
1996; Li and Liu, 2012). included as niche information. Time-fixed effects are also included to
Beyond examining population size, Lam and Chau (2012) considered accommodate the energy variability effect , eit , especially for testing the
the influence of the income levels of nearby households on the overall pre- and post-COVID effects. Mathematically,
performance of shopping malls. However, they did not find a significant ln(Sit ) = α0 + zT ln(Eit ) + zτ ln(eit ) + βQit + γLi +θMi + πTi + εit (3)
correlation. They postulated that the accessibility of shopping malls
might affect the size of catchment areas. However, they didn’t factor in The niche variables are categorised into a four-factor model, viz.,
the sizes and types of the malls, which could be determined by the mall’s mall type and scale factors Qit , location factors Li , leasing-management
accessibility and influence the size of the catchment areas. factors Mi , and tenant mix factors Ti . The mall scale is measured by
This study introduces a new supply-demand framework to examine the leased area, Qit,A = ai (1 − Vi,t ), where ai is the gross lettable area,
the effect of purchasing power within catchment areas of malls on retail and Vi,t is the vacancy rate of mall i at year t.
performance. We aim to balance the often-overlooked supply-side fac­ To control the leasing-management factors, we use a dataset with all
tors, such as the size and type of malls, with the traditionally focused shopping malls being managed by the same company group in Australia,
demand factors, like population density and income levels. By doing so, the Scentre Group, from 2014 to 2021. Thus, it assumes uniform
we can gain a more nuanced understanding of the dynamics that shape competition among shopping malls in the neighbourhoods. To further
retail performance and catchment area size, shedding light on the investigate the purchasing power effects on different scale malls, we
intricate interplay of factors such as mall size, location, and product categorise the sampled shopping centres into three mall types based on
offerings with purchasing power. This will ultimately provide more their ranges of gross lettable area: small malls, Qi,S (GLA ≤75,000 sqm),
refined insights for strategic planning and decision-making in the retail large, Qi,L (GLA >123,000 sqm) and medium, Qi,M (75,000 sqm < GLA≤
sector. 123,000 sqm), which almost divides the sample into three equal
numbers of samples. Tenant mix factors include the proportion of
2.1.1. Catchment studies on biodiversity foreign anchor tenants in the lettable area, Ti,a ; the proportion of su­
The concept of catchment has been intensively applied in studying permarkets in the lettable area, Ti,s ; the proportion of cinema in the
biodiversity, as the sustainability of ecosystems is shaped by catchment- lettable area, Ti,c ; and number of tenants per sqm lettable area, Ti,n ..
scale properties and processes (Kärnä et al., 2019). These studies Our model specifications basically align with Huff’s (1964) Gravity
emphasise that the health and resilience of ecosystems are intricately Model, which considered shopping mall size, no. of shops and distance
tied to the broader catchment area from which they draw resources, to customers as the major factors of mall attractiveness. Finn and Lou­
nutrients, and energy (Beier and Brost, 2010). The notion of biodiversity viere (1996) and Teller and Reutterer (2008) further suggested
has been adapted to develop a tenant-mix model in retailing (Yiu and including anchor stores and tenant mix. El-Adly (2007) also found
Xu, 2009) and Xu et al. (2022) further formulated the entertainment, mall essence, and convenience as determinants of mall
Island-Species-Area Energy Relationship (ISAER) model to investigate attractiveness. Our model is derived from ecological theory which has
the relationship between catchment attributes and retail sales. incorporated similar factors such as mall size, no. of shops, catchment
demographics, carparking space, anchor tenants, shops for entertain­
ment, supermarket, tenant mix, etc. The novelty of our model, however,

3
C.Y. Yiu et al. Journal of Retailing and Consumer Services 76 (2024) 103591

is to allow a flexible catchment size to conduct sensitivity analysis. The fixed isochrone of 30-min car driving distance. Further, GIS information
fundamental insight of the Gravity Model is that shoppers may not of each shopping centre, including latitude and longitude, is retrieved
choose to shop at the nearest stores but choose locations in proportion to from their website.
the attractiveness of the malls and in inverse proportion to their dis­ However, it does not allow sensitivity analysis on the catchment size,
tances. For example, Anselmsson (2016) showed positive sales return on as the raw data of purchasing power, such as number of households and
improvements in access, among others, but Calvo-Porral and median household incomes, are not readily available. This study
Lévy-Mangín (2018) found that the convenience of shopping malls is not therefore retrieves the total number of households, the median weekly
a significant factor in pulling customers. A sensitivity analysis of household income, and the total household income data from the 2021
catchment size enables a study on the spatial interactive results between censuses of the Australia Bureau of Statistics (AURIN, 2023) to estimate
accessibility and attractiveness. the purchasing power of the catchment area of each mall. The common
demarcation of household income data is based on some administrative
3.2. An adjustable isochrone model for measuring purchasing power of area units, such as Statistical Area Level 1 (SA1) in Australia which is the
catchment area smallest area unit. This study tests the shopping malls’ species-energy
hypothesis by two shoppers’ energy levels metrics. The first one is
To develop a more thorough understanding of catchment area dy­ based on the data provided by the mall owners, and the second one is
namics, we perform a sensitivity analysis to assess the purchasing power based on the census household data of the SA1s intersected with an
of catchment areas of varying sizes using an isochrone model. Tradi­ adjustable isochrone of 1-km, 5-km, 10-km, 20-km, 30-km from each
tional methods typically use information provided by mall owners, mall.
which represents the purchasing power of households within a fixed 30- The market vacancy data is provided by Vicinity (2021) on the
min car driving isochrone from the mall. However, the lack of public following five states (New South Wales (NSW), Victoria (VIC),
accessibility to these raw data sources hinders the possibility of con­ Queensland (QLD), South Australia (SA), Western Australia (WA) and
ducting sensitivity analysis for different catchment sizes. In response, Mirvac Group (2014–2021) for the Australia Capital Territory (ACT)
our study introduces an adjustable isochrone model that enables a more State. It is calculated as the average vacancy of all reporting shopping
comprehensive sensitivity analysis of the catchment area’s purchasing centres in the same state in the same year (Mirvac, 2021; Vicinity,
power’s impact on retail sales. 2021). Table 2 describes shopping centres in line with each state at the
This model operates on the basis of isochrones of various distances end of 2021. Appendix 1 shows the description of these shopping centres
from the mall. The choice of this approach is largely driven by the from 2014 to 2021.
availability of median total family income data which is commonly The variables employed include total retail sales (Sit ), and gross
derived from census records based on administrative area units, such as lettable area (Qi,A ) of each mall i in year t. According to the Shopping
census tracts in the U.S. and Statistical Area (SA) units in Australia. Centre Council of Australia, total sales are the annual sales from January
Through geofencing approach, we identify administrative area units to December, including trade from all tenants (SCCA, 2010). The gross
intersecting with the designated isochrone circle centred at the mall. lettable area includes all tenants’ trading areas at the shopping centres.
This study contemplates five different isochrone radii representing In addition, it specifies anchor tenants as major tenants who occupy at
catchment areas, namely, 1-km, 5-km, 10-km, 20-km, and 30-km from least 400 sqm of the lettable area. This study employs two common types
the mall. As an illustration, Fig. 1 presents the geofenced SA1s within the of anchor tenants: supermarkets and cinemas. Foreign anchor tenants
5-km isochrone area of Airport West, highlighting a total of 338 SA1s are the major tenants who meet two conditions: (1) the lettable area is
intersecting with the isochrone. more than 400 sqm and (2) they are founded overseas instead of in
The primary advantage of this model lies in its flexibility; the size of Australia. The foreign anchor variable represents the percentage of the
the isochrone can be freely adjusted while ensuring that each adminis­ total lettable area of foreign anchor tenants in the mall.
trative area unit maintains a similar population size. This adjustable To represent tenant mix, two variables are employed. First, the total
feature enhances the precision of our sensitivity analysis by allowing for tenant per sqm of lettable area (Ti ) is determined by dividing the total
variations in catchment area sizes. However, the irregular shape of each number of tenants in a mall by the total lettable area of the mall. Second,
administrative unit can distort the size of the catchment area and it also estimates non-linear effect of the number of tenants per sqm by
include households beyond the designated isochrone. including the square terms of total tenant per sqm of lettable area.
In this study, Python geospatial packages are used for geofencing. Shopping centres are categorised into three types in terms of their gross
The complete Python codes are available on GitHub at https://github.co lettable area: small (Qi,A ≤ 62,000 sqm), large (Qi,A > 98,000 sqm), and
m/AURIN-OFFICE/shopping_centre_buffer. We consider the total num­ medium (62,000 sqm < Qi,A ≤ 98,000 sqm) to test the moderating effect
ber of households, the mean of the weekly median total household in­ of mall size on the purchasing power effect on retail sales.
comes, and the total household incomes of SA1s in the intersected area
(isochrone) as three proxies of the purchasing power of the mall’s 3.3.1. Study area and data of purchasing power of isochrones
catchment area. The descriptive statistics of the total number of This study covers significant number of shopping malls in Australia.1
households and the mean of weekly median total household incomes of Major shopping malls at the six states (NSW, VIC, QLD, SA, WA, ACT)
all SA1s in the isochrone area are estimated. On average, 31, 377, 1117, are included.2 The population of Melbourne and Sydney represented
3199, and 5064 SA1s are identified in the 1-km, 5-km, 10-km, 20-km, 59.3% of all capital cities in 2021.
30-km intersected areas, respectively. We retrieved the demographic data from Census of Population and
Housing data (AURIN, 2023). Data of the number of households and the
3.3. Data median income dataset are collected from “SA1-G02 Selected Medians

This study examines the determinants of retail sales based on a pool

of data from 37 shopping centres in Australia with a total gross lettable 1
Australia is a highly urbanised country. According to the World Bank, the
area of about 3.6 million sqm. The sample represents 13.6% total gross
urban population in Australia reached 86% of the total population in 2021.
lettable area of shopping centres across Australia, according to Baker https://data.worldbank.org/indicator/SP.URB.TOTL.IN.ZS?locations=AU.
(2018). The data on shopping centres are provided by Scentre Group in 2
It covers major malls at the capital cities, such as Melbourne and Sydney,
their annual reports of property compendium from 2014 to 2021 which experienced the largest population growth among Australia’s capital
(Scentre, 2014–2021). Accordingly, it defines and only provides total cities between 1971 and 2021. https://www.abs.gov.au/articles/50-years-capit
household income data on the catchment areas of shopping centres by a al-city-population-change.

4
C.Y. Yiu et al. Journal of Retailing and Consumer Services 76 (2024) 103591

Fig. 1. SA1s within the Airport West’s 5-km isochrone area are identified by the geofencing programme and coloured in blue. Source: created by the authors.

Table 3 shows the summary statistics of the variables, and Appendix

Table 2
1 provides the breakdown analysis of the shopping centres based on the
Data descriptions of the shopping centres by state in 2021.
three size categories across the six states from 2014 to 2021 and their
No State Number of shopping centres Total GLA (‘000 sqm) % GLA location maps.
1 NSW 15 1409 39.0%
2 VIC 7 739 20.5% 4. Results and discussion
3 QLD 6 674 18.7%
4 SA 3 307 8.5%
5 WA 4 315 8.7%
Table 4 shows the results of ISAER models. The baseline model
6 ACT 2 168 4.6% (ISAER E30min Model) confirms the positive impact of purchasing
Total 37 3611 100% power within the 30-min catchment area on shopping mall sales.4 One
per cent rise in the purchasing power in the catchment increases 4.1% of
the retail sales amount of a mall. Most of the other coefficients are sig­
and Averages-Census” 2021. We chose the highest granular data without
nificant and of the expected signs. For example, the leased area is
restriction by the Statistical Level 1 area (SA1). Based on the Australian
positively associated with the sales amount, and all three types of special
Statistical Geography Standard (ASGS) Edition 3, SA1s generally have a
tenants are found to impose positive effects, yet their statistical signifi­
population of 200–800 people and an average population of about 400.3
cances are marginal. The effect of the number of tenants per sqm lettable
Based on the ABS description of SA1, there were 61,845 spatial SA1
area on sales amount is increasing at a decreasing rate, with an optimal
regions in 2021. The total number of households, the mean of the weekly
of about 4.5 shops per 1000 sqm. The negative pandemic effect is also
median total household incomes, and the total household incomes of all
detected as the coefficient of the year 2021 dummy is the only negative
the SA1s intercepted in the isochrone are estimated by our program to
proxy the purchasing power of the catchment area.
Summary Statistics.
4
The results of a reliability and validity tests are provided in Table A3 in the
Appendix. The data are found to be reliable as the catchment effect is found to
3
https://www.abs.gov.au/statistics/standards/australian-statistical-geogra be robust in using different factors of attractiveness. The data are also found to
phy-standard-asgs-edition-3/jul2021-jun2026/main-structure-and-greater-ca be valid, as the catchment effect is found to be equally strong and significant
pital-city-statistical-areas/statistical-area-level-1. before the COVID-19 pandemic.

5
C.Y. Yiu et al. Journal of Retailing and Consumer Services 76 (2024) 103591

Table 3 Table 3 (continued )

Data descriptions and summary statistics of the variables. Var. Descriptions Mean SD Min Max
Var. Descriptions Mean SD Min Max
Ti,s Proportion of supermarket 0.10 0.05 0.00 0.27
ln(Sit ) Natural log of total retail sales 6.15 0.58 3.01 7.15 tenants’ size (in ratio)
of mall i at time t (in A$) Ti,c Proportion of cinema tenants’ 0.06 0.03 0.00 0.15
ln(E30 min Natural log of the purchasing 3.41 0.66 1.95 5.00 size (in ratio)
i,2021 )
power of the catchment area for Ti,n Number of tenants per 1000 3.26 0.64 0.97 5.19
mall i in 2021, which equals sqm of total gross lettable area
total household incomes in
Sources: Mall characteristics data are retrieved from Scentre Group (2014–2021)
billions within the 30-min
isochrone from the mall. Data (https://www.scentregroup.com/getmedia/5f8a54e7-0ffa-4b84-9e3d-0d5b3ce
provided by the mall owners. 60b2e/2021-Property-Compendium.pdf); data of total household incomes
ln(H5− km Natural log of the purchasing 10.82 0.53 9.12 12.06 within 30-min driving isochrone Ei,202130min are also retrieved from Scentre
i,2016 )
power of the catchment area for Group (2021); only 5-km isochrone data for sensitivity analyses are shown here,
mall i in 2016, which equals including total number of households, mean of median household incomes, and
total number of households total household incomes are sourced from the AURIN (2023) at
within the 5-km isochrone from (https://researchdata.edu.au/sa1-g02-selected-census-2016/1983179).
the mall. Data estimated by the
authors using census data.
ln(H5− km
i,2021 )
Natural log of the purchasing 10.92 0.53 9.19 12.15
power of the catchment area for one in the year dummies, though insignificant. More interestingly, the
mall i in 2021, which equals positive catchment purchasing power effect can also be confirmed by
total number of households
within the 5-km isochrone from
using number of households, mean of the median household incomes
the mall. Data estimated by the and total household incomes within the 5-km isochrone from the malls
authors using census data. in ISAER H5km, M5km and E5km Models respectively. The magnitudes
ln(M5− km
i,2016 )
Natural log of the purchasing 7.59 0.21 7.22 8.02 of the catchment purchasing power effects of the models are commen­
power of the catchment area for
surate. It supports the following sensitivity analyses using the proxies.
mall i in 2016, which equals
mean of the median household
incomes in AUD within the 5-
4.1. Sensitivity analysis on ISAER models
km isochrone from the mall.
Data estimated by the authors
using census data. Applying the adjustable catchment area analysis model, the sensi­
ln(M5− km
i,2021 )
Natural log of the purchasing 7.63 0.19 7.33 8.04 tivity tests result in Table 5 show that the purchasing power effects are
power of the catchment area for sensitive to the catchment area size, and level off at about 20 km from
mall i in 2021, which equals
the mall, which comes close to the industry practice of using 30-min
mean of the median household
incomes in AUD within the 5- driving distance as the catchment area. However, the number of
km isochrone from the mall. households proxy imposes a stronger short-range effect, while the me­
Data estimated by the authors dian household income proxy has a longer-range effect on retail sales.
using census data.
ln(E5− km
i,2016 )
Natural log of the purchasing 18.25 0.62 16.37 19.76
power of the catchment area for 4.2. Robustness tests
mall i in 2016, which equals
total household incomes in
millions within the 5-km
Table 6 shows the results of the DID models for testing the pur­
isochrone from the mall. Data chasing power effects on different-sized malls, using the 20-km iso­
estimated by the authors using chrone. First, all models’ results confirm the positive effect of the
census data. catchment’s purchasing power on retail sales for small-sized malls.
ln(E5− km Natural log of the purchasing 18.52 0.62 16.61 20.00
i,2021 ) Second, all models show that the larger the mall size, the higher the
power of the catchment area for
mall i in 2021, which equals retail sales, ceteris paribus, unless when the purchasing power exceeds
total household incomes in the threshold. For example, in ISAER-SIZE H20km Model, only when the
millions within the 5-km number of households within the isochrone exceeds 1.23 million
isochrone from the mall. Data 4.600

estimated by the authors using (e− 1(− 0.328) ), which is far beyond the maximum number in the sample,
census data. will the retail sales of small-sized malls equal that of large-sized malls.
ln(eit ) Dummy variable of the year t 2014–2021 The results reflect that small-sized malls rely more heavily on shoppers
that the sales amount of mall i is from the neighbourhood in the vicinity, while large-sized malls can
measured
Qi,A Total gross lettable area of mall 86.72 40.45 10.93 179.0
attract shoppers farther away from the malls. The findings agree with
i (in ’000 sqm.) shopping mall management theory that major and super-regional malls
Vi,t Market vacancy rate of mall i at 0.01 0.01 0.00 0.07 are usually larger in size and can attract shoppers far away from the local
time t (in ratio) neighbourhoods, and even cross-border tourists (Li et al., 2018).
Qi,S Dummy variable of the small- 124 small malls (33.70%)
sized malls, GLA ≤62,000 sqm
Qi,M Dummy variable of the 122 medium malls (33.15%) 4.3. Sensitivity analysis on ISAER-SIZE models
medium-sized malls, 62,000 <
GLA ≤98,000 sqm
Qi,L Dummy variable of the large- 122 large malls (33.15%) After taking into account the mall size effect, the sensitivity tests
sized malls, GLA >98,000 sqm result in Table 7 show that the catchment’s purchasing power effect for
Qi,C Number of car-parking spaces 3420 1642 0.00 7200 small-sized malls peaked at about 5- to 10-km from the malls, with a 1
Li Dummy variables of the six six states of Australia per cent increase in total household incomes can increase 0.28 per cent
states where mall i is located
in retail sales. The results show that the catchment size for small-sized
Ti,a Proportion of foreign anchor 0.12 0.10 0.00 0.44
tenants’ size (in ratio) malls is smaller, which agrees with the shopping mall management
theory that city and neighbourhood shopping centres are more reliant on
shoppers from local neighbourhoods in the vicinity.

6
C.Y. Yiu et al. Journal of Retailing and Consumer Services 76 (2024) 103591

Table 4 Table 5
The results of ISAER Models. The results of sensitivity analysis tests.
Variables ISAER ISAER H5km ISAER M5km ISAER E5km Panel 5.1 ISAER H1km ISAER H10km ISAER H20km ISAER H30km
E30min Model Model Model Model Model Model Model
Model
ln(Hx− km
i,2021 )
0.131 (5.98) 0.090 (4.67) 0.071 (3.99) 0.073 (4.03)
Constant 3.749 (7.33) 3.266 (6.17) 2.145 (3.27) 2.807 (5.00) *** *** *** **
*** *** *** ***
ISAER M1km ISAER M10km ISAER M20km ISAER M30km
ln(E30 min
i,2021 )
0.041 (2.50)
Panel 5.2 Model Model Model Model
**
ln(H5− km
i,2021 )
0.074 (3.43) ln(Mx− km 0.116 (1.80) 0.337 (4.14) 0.415 (3.63) 0.419 (3.01)
i,2021 )
*** *** *** *** ***
ln(M5− km
i,2021 )
0.265 (3.97)
ISAER E1km ISAER E10km ISAER E20km ISAER E30km
***
Panel 5.3 Model Model Model Model
ln(E5− km
i,2021 )
0.074 (3.99)
** ln(Ex− km 0.116 (5.89) 0.084 (4.98) 0.064 (4.04) 0.063 (3.93)
i,2021 )
Qit,A 0.012 (13.08) 0.012 (12.62) 0.012 (14.12) 0.012 (12.61) *** *** *** ***
*** *** *** *** Dep. ln(Sit ) ln(Sit ) ln(Sit ) ln(Sit )
Qi,c − 0.00004 − 0.00003 − 0.00004 − 0.00003 Variable
(− 2.42)** (− 1.92)* (− 2.74)*** (− 1.81)* No. of Obs 279 279 279 279
Ti,a 0.166 (0.68) 0.215 (0.90) 0.249 (1.04) 0.250 (1.05)
Ti,s 0.113 (0.34) 0.097 (0.30) 0.003 (0.01) 0.055 (0.17) Notes: ***, **, * means that the coefficient is significant at the 1%, 5%, 10%
Ti,c − 0.073 0.157 (0.34) 0.681 (1.46) 0.225 (0.49) levels. The figures in the parentheses are the t-statistics. For clarity, the results of
(− 0.15) the other variables are similar to the ISAER Model and are not shown here.
Ti,n 0.612 (2.24) 0.527 (1.95)* 0.330 (1.20) 0.456 (1.69)* Panels 5.1, 5.2 & 5.3 refer to ISAER H, M, E models. Superscript x in Hi,2021 x− km
,
** x− km x− km
Mi,2021 & Ei,2021 refers to 1 km, 10 km, 20 km or 30 km according to the column
(Ti,n )2 − 0.068 − 0.059 − 0.019 − 0.047
(− 1.69)* (− 1.48) (− 0.47) (− 1.19) header.
eit Fixed (year Fixed (year Fixed (year Fixed (year
dummies) dummies) dummies) dummies)
Li Fixed (state Fixed (state Fixed (state Fixed (state
dummies) dummies) dummies) dummies)
Dep. ln(Sit ) ln(Sit ) ln(Sit ) ln(Sit ) Table 6
Variable The results of ISAER-SIZE tests.
No. of Obs 276 279 279 279 Variables ISAER-SIZE ISAER-SIZE M20km ISAER-SIZE
Adj.R- 0.85 0.85 0.86 0.86 H20km Model Model E20km Model
Squared
Constant 1.3795 (2.20)** − 12.066 (− 4.37) 0.057 (0.08)
Notes: ***, **, * means that the coefficient is significant at the 1%, 5%, 10% ***
levels. The figures in the parentheses are the t-statistics. ln(X20− km
i,2021 )
0.208 (6.49)*** 2.117 (5.85)*** 0.190 (6.49)***
ln(X20− km
i,2021 ) ×
− 0.188 (− 4.92) − 1.671 (− 4.28)*** − 0.166 (− 4.80)
5. Conclusions Qi,M *** ***
ln(X20− km
i,2021 ) ×
− 0.328 (− 5.64) − 2.336 (− 6.20)*** − 0.307 (− 6.23)
Qi,L *** ***
Many previous studies concerning the catchment areas of shopping
Qi,M 2.636 (5.33)*** 12.897 (4.35)*** 3.622 (5.09)***
malls have utilised rigidly defined parameters, such as a 30-min driving Qi,L 4.600 (5.93)*** 18.068 (6.29)*** 6.675 (6.45)***
isochrone. However, these specifications can be somewhat arbitrary, Dep. Variable ln(Sit ) ln(Sit ) ln(Sit )
and their validity is difficult to verify. Moreover, these studies often do No. of Obs 279 279 279
not take into account the unique characteristics and positioning of in­ Adj.R-Squared 0.88 0.89 0.89
dividual malls, such as their size, which can impact the geographical Notes: ***, **, * means that the coefficient is significant at the 1%, 5%, 10%
extent of their reach. levels. The figures in the parentheses are the t-statistics. For clarity, the results of
This research project aimed to deepen our understanding of catch­ other variables are similar to the ISAER Model and are not shown here. Xi,2021 20− km

ment areas for a national set of retail malls by conducting a sensitivity 20− km
refers to either Hi,2021 20− km
, Mi,2021 20− km
orEi,2021 according to the column header.
analysis of purchasing power within various catchment sizes. Utilising
principles from species-energy theory and niche specialisation theory
within the field of ecology, we developed an empirical model to examine
demographics. Our sensitivity analysis also showed that the effect of
the relationship between retail sales and the purchasing power of
purchasing power begins to level off at about a 20-km distance from the
catchment areas. This was done while controlling for various factors,
malls, regardless of which purchasing power proxy was measured. This
including mall type, scale, location, leasing-management factors, and
finding aligns with industry norms, which typically measure catchment
tenant mix.
areas as a 30-min driving distance from urban areas. In conclusion, this
A crucial focus of this study was the introduction of an adjustable
study emphasises the importance of considering the purchasing power
isochrone model, allowing for the sensitivity analysis of different
and demographic characteristics within catchment areas for a more
catchment sizes ranging from 1-km to 30-km radii from the malls. This
accurate assessment of shopping mall performance.
model used administrative area units intersected with designated iso­
The findings of this retail catchment study hold significant mana­
chrones to estimate three proxies of purchasing power: (1) number of
gerial, social, and practical implications for both shopping mall opera­
households, (2) median of household incomes, and (3) total household
tors and urban planners. From a managerial standpoint, the identified
incomes within the isochrone.
catchment-scale effects on purchasing power provide a nuanced un­
Our findings provide novel insights into the relationship between
derstanding of how different types of malls can leverage their catchment
catchment areas and retail sales across malls of various sizes. Notably,
areas to optimise sales strategies. This understanding equips retailers
we found that the purchasing power within catchment areas has a sig­
with the knowledge to tailor their tenant mix, marketing approaches,
nificant, positive influence on retail sales, particularly for smaller malls.
and overall business strategies according to the unique characteristics of
This aligns with theories in shopping centre management, which suggest
their catchment area (Li et al., 2023). This targeted approach enhances
that larger malls are typically less dependent on local shopper
customer satisfaction and potentially leads to increased footfall and

7
C.Y. Yiu et al. Journal of Retailing and Consumer Services 76 (2024) 103591

Table 7 transportation infrastructure, urban density, local preferences, and so­

The results of ISAER-SIZE tests. cial status (Beiró et al., 2018). This underscores the importance of
Variables ISAER-SIZE ISAER-SIZE ISAER-SIZE ISAER-SIZE integrating a more comprehensive array of variables to capture the so­
E1km Model E5km Model E10km Model E30km Model cietal norms of catchment areas across different countries. So far, there
Constant 0.573 (0.71) − 0.990 − 1.349 0.740 (1.08) have been very few cross-country studies on the variations of retail
(− 0.84) (− 1.37) catchment.
ln(Ex− km
i,2021 )
0.207 (5.84) 0.285 (4.92) 0.282 (6.41) 0.152 (5.93) In light of these limitations, future research endeavours could
*** *** *** *** consider incorporating real-time mobility data and applying geofencing
ln(Ex− km − 0.128 − 0.239 − 0.234 − 0.131
i,2021 ) ×
(− 3.14)*** (− 3.70)*** (− 4.74)*** (− 4.13)***
techniques to identify the actual catchment sizes of malls (Álvarez and
Qi,M
− 0.162 − 0.295 − 0.323 − 0.296
Arredondo, 2019; Luo et al., 2014; Ho et al., 2020). They enable the
ln(Ex− km
i,2021 ) ×
Qi,L (− 3.54)*** (− 4.21)*** (− 5.30)*** (− 6.41)*** analysis of customers’ revealed preferences on retailing location choices
Qi,M 2.201 (3.43) 4.615 (3.88) 4.760 (4.95) 2.967 (4.46) (Streed et al., 2015). Further studies on the variations of retail catch­
*** *** *** *** ment across cities (Ballantyne et al., 2023) or countries to test the
Qi,L 2.805 (3.78) 5.736 (4.41) 6.595 (5.49) 6.608 (6.65) generalisability of the findings can also uncover new dimensions of
*** *** *** ***
catchment influence and devise innovative strategies that resonate with
Dep. ln(Sit ) ln(Sit ) ln(Sit ) ln(Sit )
Variable the ever-changing patterns of consumer behaviour, ensuring the sus­
No. of Obs 279 279 279 279 tained vitality of modern retail practices.
Adj.R- 0.88 0.88 0.89 0.89
Squared
Author contributions
Notes: ***, **, * means that the coefficient is significant at the 1%, 5%, 10%
levels. The figures in the parentheses are the t-statistics. For clarity, the results of Conceptualisation, CY.Y. and QL.T.; methodology, CY.Y. and J.K.;
other variables are similar to the ISAER Model and are not shown here. Super­ formal analysis, CY.Y. and KS.C.; investigation, CY.Y., QL.T., J.K.; re­
x− km
script x in Ei,2021 refers to either 1 km, 5 km, 10 km or 30 km according to the sources, KS.C.; data curation, QL.T., J.K.; writing—original draft prep­
column header. aration, CY.Y. and QL.T.; writing—review and editing, CY.Y. and KS.C.;
supervision, CY.Y.; project administration, CY.Y.; funding acquisition,
KS.C. All authors have read and agreed to the published version of the
sales, which can help better evaluate the feasibility of future venture. On manuscript.
a social level, the study’s emphasis on the dynamics between catchment
areas and retail sales highlights the interconnectedness between com­ Funding
mercial activities and local community needs. This notion of "local
community needs" also relates to the retail gentrification that encom­ No funding involved.
passes the preferences, demands, and necessities of the residents who
form the backbone of the neighbourhood’s identity and vitality (Cheung Disclosure statement
and Yiu, 2023). Understanding the nuances of catchment areas allows
for more community-centred retail planning, potentially addressing The authors declare no conflict of interest.
local needs and aspirations more effectively. From a practical perspec­
tive, the adjustable isochrone model introduced in the study presents a
Declaration of competing interest
valuable tool for urban planners and policymakers. The model’s ability
to conduct sensitivity analysis on different catchment sizes offers a
The authors declare that they have no known competing financial
practical approach to assess and validate catchment area assumptions.
interests or personal relationships that could have appeared to influence
This can guide urban development decisions on the locations and sizes of
the work reported in this paper.
shopping malls.
However, this study has several limitations that warrant consider­
Data availability
ation in the interpretation of the study’s findings. First, the census data
for population and household incomes can only be available once every
all data are publicly available
five years, which introduces a constraint in matching retail performance
in different years. Given the rapidly evolving socio-economic landscape,
Acknowledgements:
especially during the pandemic, the use of this census data may not fully
capture recent changes in purchasing power and demographic shifts
We thank the Australian Urban Research Infrastructure Network
within catchment areas. Additionally, the study’s dataset is mainly
(AURIN) for the censuses data, and Gary Yihan Guan for his assistance in
derived from a single shopping mall owner for better controlling the
reviewing the literature.
ownership and management factors, but it potentially limits the gen­
eralisability of the results, and has to assume uniformity in the compe­
tition and cooperation among shopping malls in the area. It further Appendix A. Supplementary data
limits the exploration of how the geolocation of shopping mall clusters
interacts with the size distribution and related catchment areas. Vari­ Supplementary data to this article can be found online at https://doi.
ability in retail centre ownership practices and competition-cooperation org/10.1016/j.jretconser.2023.103591.
among different owners may not be fully accounted for.
Furthermore, while innovative, the adjustable isochrone model is not References
without limitations. The model’s assumptions about isochrone boundary
Adhinugraha, K., Taniar, D., Phan, T., Beare, R., 2022. Predicting travel time within
intersections with administrative units may lead to distortions in catchment area using time travel oronoi diagram (TTVD) and crowdsource map
catchment area size estimates, potentially affecting the accuracy of features. Inf. Process. Manag. 59 (3), 102922.
purchasing power proxies. Moreover, the model’s focus on geographical Álvarez, M., Arredondo, A., 2019. Calculating catchment areas with human mobility
data. Data Sci. CARTO Blog, Nov. 21 https://carto.com/blog/calculating-catch
catchment areas may overlook socio-cultural factors that influence ment-human-mobility-data/.
consumer behaviour and retail patronage decisions, such as Anselmsson, J., 2016. Effects of shopping centre re-investments and improvements on
sales and visit growth. J. Retailing Consum. Serv. 32, 139–150.

8
C.Y. Yiu et al. Journal of Retailing and Consumer Services 76 (2024) 103591

AURIN, 2023. This Research Used the NCRIS-Enabled Australian Urban Research Li, Y., Liu, L., 2012. Assessing the impact of retail location on store performance: a
Infrastructure Network (AURIN) E-Infrastructure to Access Census Datasets on 10 comparison of Wal-Mart and Kmart stores in Cincinnati. Appl. Geogr. 32 (2),
March 2023. Available at: https://data.aurin.org.au/. 591–600.
Baker, M., 2018. Australia in the global shopping centre industry. Available at: https Lloyd, A., Cheshire, J., 2017. Deriving retail centre locations and catchments from geo-
://scca.org.au/wp-content/uploads/2020/01/AUSTR1.pdf. (Accessed 20 October tagged Twitter data. Comput. Environ. Urban Syst. 61, 108–118.
2022). Luo, X., Andrews, M., Fang, Z., Phang, C.W., 2014. Mobile targeting. Manag. Sci. 60 (7),
Ballantyne, P., Singleton, A., Dolega, L., Macdonald, J., 2023. Integrating the who, what 1738–1756. https://doi.org/10.1287/mnsc.2013.1836.
and where of US retail center geographies. Ann. Assoc. Am. Geogr. 113 (2), 488–510. Mason, N.W.H., Lanoiselee, P.I.C., Mouillot, D., Argillier, C., 2008. Evidence that niche
Beier, P., Brost, B., 2010. Use of land facets to plan for climate change: conserving the specialisation explains species-energy relationships in lake fish communities.
arenas, not the actors. Conserv. Biol. 24 (3), 701–710. J. Anim. Ecol. 77 (2), 285–296.
Beiró, M.G., Bravo, L., Caro, D., Cattuto, C., Ferres, L., Graells-Garrido, E., 2018. Mirvac, 2021. Results and Announcements available at: https://www.mirvac.com/inves
Shopping mall attraction and social mixing at a city scale. EPJ Data Science 7, 28. tor-centre/results-and-announcements/analyst-toolkit. (Accessed 4 October 2022).
https://doi.org/10.1140/epjds/s13688-018-0157-5. Morrison, P.A., Abrahamse, A.F., 1996. Applying demographic analysis to store site
Bilková, K., Krizan, F., Barlík, P., 2016. Consumers preferences of shopping centers in selection. Popul. Res. Pol. Rev. 15 (5), 479–489.
bratislava (Slovakia). Human Geographies 10 (1), 23. Rao, F., Pafka, E., 2021. Shopping morphologies of urban transit station areas: a
Birkin, M., Clarke, G., Clarke, M., 2010. Refining and operationalizing entropy- comparative study of central city station catchments in Toronto, San Francisco, and
maximizing models for business applications. Geogr. Anal. 42 (4), 422–445. Melbourne. J. Transport Geogr. 96, 103156.
Böhm, M., Kemp, R., Williams, R., Davidson, A.D., Garcia, A., McMillan, K.M., Rasouli, S., Timmermans, H., 2013. Assessment of model uncertainty in destinations and
Bramhall, H.R., Collen, B., 2017. Rapoport’s rule and determinants of species range travel forecasts of models of complex spatial shopping behaviour. J. Retailing
size in snakes. Divers. Distrib. 23, 1472–1481. https://doi.org/10.1111/ddi.12632. Consum. Serv. 20 (2), 139–146.
Calvo-Porral, C., Lévy-Mangín, J.P., 2018. Pull factors of the shopping malls: an SCCA, 2010. Sales reporting guidelines. SCCA. available at: https://scca.org.au/indust
empirical study. Int. J. Retail Distrib. Manag. 46 (2), 110–124. ry-information/industry-data/.
Carrara, R., Vàzquez, D.P., 2010. The species-energy theory: a role for energy variability. Scentre, 2014. Property Compendium. S. G. Limited available at: https://www.scentre
Ecography 33 (5), 942–948. https://doi.org/10.1111/j.1600-0587.2009.05756.x. group.com/getmedia/396300fa-e045-49cd-9362-0215507afa23/SCG-Propert
Cervero, R., Kang, C.D., 2011. Bus rapid transit impacts on land uses and land values in y-Compendium.pdf. (Accessed 4 October 2022).
Seoul, Korea. Transport Pol. 18 (1), 102–116. Scentre, 2015. Property Compendium. S. G. Limited available at: https://www.scentregr
Cheung, K.S., Yiu, C.Y., 2023. Unfolding touristification in retail landscapes: evidence oup.com/getmedia/69d4dbac-45dc-4597-9706-a6e08b2a3bf8/SCG-Prop-Comp-20
from rent gaps on high street retail. Tourism Geogr. 25 (4), 1224–1250. 16-25.pdf. (Accessed 4 October 2022).
Christaller, W., 1933. Central Places in Southern Germany, 1966. Englewood Cliffs, N.J. Scentre, 2016. Property Compendium. S. G. Limited available at: https://www.scentre
Deng, T., Nelson, J.D., 2010. The impact of bus rapid transit on land development: a case group.com/getmedia/c694c814-1353-42d4-86a2-817eb35e230a/Property-Compe
study of Beijing, China. Int. J. Hum. Soc. Sci. 4 (6), 1169–1179. ndium_21-Feb-17.pdf. (Accessed 4 October 2022).
Dennis, C., Marsland, D., Cockett, T., 2002. Central place practice: shopping centre Scentre, 2017. Property Compendium. S. G. Limited available at: https://www.
attractiveness measures, hinterland boundaries and the UK retail hierarchy. scentregroup.com/getmedia/24ca4fa6-1595-408e-b818-a87f2b24cd6d/2017-12-
J. Retailing Consum. Serv. 9 (4), 185–199. Property-Compendium-(inc-NZ)-updated-links.pdf. (Accessed 4 October 2022).
Dolega, L., Pavlis, M., Singleton, A., 2016. Estimating attractiveness, hierarchy and Scentre, 2018. Property Compendium. S. G. Limited available at: https://www.scentregro
catchment area extents for a national set of retail centre agglomerations. J. Retailing up.com/getmedia/fb3e2490-0909-43a0-99ca-693538730643/SCG-Property-Co
Consum. Serv. 28, 78–90. mpendium-2018_1.pdf. (Accessed 4 October 2022).
Drezner, T., Drezner, Z., 2002. Validating the gravity-based competitive location model Scentre, 2019. Property Compendium. S. G. Limited available at: https://www.scentregro
using inferred attractiveness. Ann. Oper. Res. 111, 227–237. up.com/getmedia/46b6eb2c-fa2b-438e-a48a-557348235804/Property-Compendi
Eaton, C., Lipsey, R.G., 1982. An economic theory of central places. Econ. J. 92 (365), um-FY2019.pdf. (Accessed 4 October 2022).
56–72. Scentre, 2020. Property Compendium. S. G. Limited available at: https://www.scentre
El-Adly, M.I., 2007. Shopping malls attractiveness: a segmentation approach. Int. J. group.com/getmedia/ab8986a7-0e8a-4554-a633-a026bddeb93f/Property-Compe
Retail Distrib. Manag. 35 (11), 936–950. ndium-2020-Full-Year.pdf. (Accessed 4 October 2022).
Ertekin, O., Dokmeci, V., Unlukara, T., Ozus, E., 2008. Spatial distribution of shopping Scentre, 2021. Property Compendium. S. G. Limited available at: https://www.scentre
malls and analysis of their trade areas in istanbul. Eur. Plann. Stud. 16 (1), 143–156. group.com/getmedia/5f8a54e7-0ffa-4b84-9e3d-0d5b3ce60b2e/2021-Prope
Finn, A., Louviere, J., 1996. Shopping centre image, consideration and choice: anchor rty-Compendium.pdf. (Accessed 4 October 2022).
store contribution. J. Bus. Res. 35 (3), 241–251. Segal, D.B., 1999. Retail trade area analysis: concepts and new approaches. J. Database
Green, J., 2019. The importance of catchment in retail. Cromwell Property Group Res. Market. 6, 267–278.
Insight. Dec. 9. https://www.cromwellpropertygroup.com/research-and-insight/r Slatyer, R.A., Hirst, M., Sexton, J.P., 2013. Niche breadth predicts geographical range
esearch/the-importance-of-catchment-in-retail. size: a general ecological pattern. Ecol. Lett. 16, 1104–1114. https://doi.org/
Gregory, R.D., Gaston, K.J., 2000. Explanations of commonness and rarity in British 10.1111/ele.12140.
breeding birds: separating resource use and resource availability. Oikos 88, Streed, O.J., Cliquet, G., Kagan, A., 2015. Optimizing geofencing for location-based
515–526. services: a new application of spatial marketing. In: Kubacki, K. (Ed.), Ideas In
Guy, C., 2002. Retail location analysis. In: Applied Geography. Routledge, pp. 478–490. Marketing: Finding the New and Polishing the Old. Developments In Marketing Science:
Guy, C., 1998. Classifications of retail stores and shopping centres: some methodological Proceedings of the Academy of Marketing Science. Springer, Cham. https://doi.org/
issues. Geojournal 45 (4), 255–264. 10.1007/978-3-319-10951-0_73.
Higgins, C., Kanaroglou, P., 2018. Rapid transit, transit-oriented development, and the Taniar, D., Rahayu, W., 2015. A taxonomy for region queries in spatial databases.
contextual sensitivity of land value uplift in toronto. Urban Stud. 55 (10), J. Comput. Syst. Sci. 81 (8), 1508–1531.
2197–2225. Tanwar, S., Kaushik, N., Kaushik, V.K., 2011. Retail malls: new mantra for success. Sri
Huff, D., 1964. Defining and estimating a trade area. J. Market. 28 (3), 34–38. Krishna Int. Res. Educ. Consortium 2, 103–113.
Ho, Y.-J., Dewan, S., Ho, Y.-C., 2020. Distance and local competition in mobile Teller, C., 2008. Shopping streets versus shopping malls–determinants of agglomeration
geofencing. Inf. Syst. Res. 31 (4), 1421–1442. https://doi.org/10.1287/ format attractiveness from the consumers’ point of view. Int. Rev. Retail Distrib.
isre.2020.0953. Consum. Res. 18 (4), 381–403.
Joseph, L., Kuby, M., 2011. Gravity modeling and its impacts on location analysis. In: Teller, C., Reutterer, T., 2008. The evolving concept of retail attractiveness: what makes
Foundations of Location Analysis. Springer, New York, pp. 423–443. retail agglomerations attractive when customers shop at them? J. Retailing Consum.
Kärnä, O.-M., Heino, J., Laamanen, T., et al., 2019. Does catchment geodiversity foster Serv. 15 (3), 127–143.
stream biodiversity? Landsc. Ecol. 34, 2469–2485. Thomas, S., Pathak, B., 2012. A study of consumer behavior approach towards shopping
Kuruvilla, S.J., Joshi, N., 2010. Influence of demographics, psychographs, shopping mall attractiveness with special reference to the city of ahmedabad. Res. J. Soc. Sci.
orientation, mall shopping attitude and purchase patterns on mall patronage in Manag. 2 (5), 146–154.
India. J. Retailing Consum. Serv. 17, 259–269. Vicinity, 2021. Financial results, vicinity. Available online at: https://www.vicinity.com.
Lam, B.Y., Chau, K.W., 2012. Explaining the variations in the pedestrian flow values of au/investors/financial-results. (Accessed 4 October 2022).
shopping centres. Facilities 30 (3/4), 164–176. Wright, D.H., 1983. Species-energy theory: an extension of species-area theory. Island
Li, L.H., Cheung, K.S., Han, S.Y., 2018. The impacts of cross-border tourists on local retail Ecol. 41 (3), 496–506.
property market: an empirical analysis of Hong Kong. J. Property Res. 35 (3), Xu, Y., Yiu, C.Y., Cheung, K.S., 2022. Retail tenant mix effect on shopping mall’s
252–270. performance. Market. Intell. Plann. 40 (2), 273–287.
Li, L.H., Cheung, K.S., Tse, W.S., 2023. Understanding the shoppers’ perception in retail Yiu, C.Y., Xu, Y., 2009. A tenant-mix model for shopping malls. Eur. J. Market. 46 (3/4),
shopping malls: a self-determination theory perspective. J. Strat. Market. 31 (1), 524–541.
58–73.