Content uploaded by David Paul Hastings
Author content
All content in this area was uploaded by David Paul Hastings on Mar 22, 2018
Content may be subject to copyright.
EXPLORING THE
COLLABORATIVE ACTIVITIES
OF HOME-BASED BUSINESSES
IN OECD COUNTRIES
David P. Hastings, Muhammad Naveed Anwar & Gobinda Chowdhury
Northumbria University, 2 Ellison Place, Newcastle upon Tyne, United Kingdom,
NE1 8ST
EXPLORING THE COLLABORATIVE
ACTIVITIES OF HOME-BASED
BUSINESSES IN OECD COUNTRIES
David P. Hastings
Northumbria University, 2 Ellison Place, Newcastle upon Tyne, United Kingdom,
NE1 8ST
Email: d.p.hastings@northumbria.ac.uk
Muhammad Naveed Anwar
Northumbria University, 2 Ellison Place, Newcastle upon Tyne, United Kingdom,
NE1 8ST
Email: naveed.anwar@northumbria.ac.uk
Gobinda Chowdhury
Northumbria University, 2 Ellison Place, Newcastle upon Tyne, United Kingdom,
NE1 8ST
Email: gobinda.chowdhury@northumbria.ac.uk
Abstract
Collaboration is frequently cited as a driver for sustainable success, and yet despite over half of all
small businesses in OECD countries being run from the home, within the existing literature little
attention is paid to how these businesses work with others. This article therefore presents a quantitative
study into the collaborative behaviours exhibited by home-based businesses located within OECD
countries. Based on a large, cross sectional data set collected by the Global Entrepreneurship Monitor,
this exploratory study outlines the extent of collaboration among home-based businesses, the nature of
their collaborative activities and the relationships which exist between the different behaviours that are
exhibited.
The study finds that collaboration is a widespread occurrence among home-based businesses, with
over 75% of home-based businesses collaborating in some way. Furthermore, home-based business
collaboration is diverse in its nature and is present across all industries. Moreover, it is found that
collaboration among home-based businesses is distinct enough from the current findings of
collaboration among SMEs that it warrants further investigation.
Keywords: Home-based business, business collaboration, Global Entrepreneurship
Monitor, OECD countries
1.0 Introduction
1.1 Background to the Research
In order to facilitate expansion and to attain competitiveness in a market, small
businesses frequently develop cooperative and collaborative relationships with other
organizations (Casals, 2011). The benefits offered by such relationships are numerous,
extending from a reduction in transaction costs through to the acquisition of hitherto
unavailable resources and the sharing of knowledge between businesses (Camarinha-
Matos & Abreu, 2007). One particular sector - home-based businesses - is able to gain
considerable benefits from these forms of collaborative relationships owing to the
scarcity of available financial, physical and knowledge based resources. By utilizing
data analytics techniques this study will offer insight into the extent of these
collaborations, the form which the take and the patterns in which they occur.
A home-based business (HBB), while often included as a form of small to medium
enterprise (SME) can be more specifically defined as “any business entity engaged in
selling products or services…operated by a self-employed person…that uses
residential property as a base from which the operation is run” (Mason, Carter &
Tagg, 2011, p.12). In this study, the term HBB is inclusive of mobile businesses and
businesses based from but not operated at the home, in line with the definition used in
contemporaneous research (Clark & Douglas, 2014). Further to this, collaboration in
the domain of SMEs and HBBs does not always rely upon formalized agreements and
may instead involve word of mouth agreements and tacit commitments (Johannisson,
1987). Thus the term “collaboration”, when used in this study, is inclusive of all
working relationships between organizations as indicated within the data.
In the UK, over 50% of SMEs are also HBBs, a sector with an annual turnover of over
£300bn, and which contributes around £40bn per year to local economies (Enterprise
Nation, 2014). Moreover, this trend is not exclusive to the UK, with studies indicating
that over 50% of small businesses are based from the home across most OECD
countries (Mason, 2010). Despite this, in many countries there is a lack of policy level
support for HBBs, with some in the literature arguing that research into the “real
world” of HBBs – including the extent of their collaborative activities – is required
for them to be perceived as important economic actors engaged in joint enterprise, and
to engender positive action among policy makers (Mason, Carter & Tagg, 2011;
Mason, 2010). Moreover, existing findings within the literature suggest that most
small businesses are reluctant to engage in collaborative activities (Casals, 2011). This
study is concerned with collaborative propensity among home-based businesses, and
will adopt a quantitative, data driven approach to providing evidence which is able to
support or deny this claim, providing evidence showing the extent of collaboration
among HBBs.
1.2 Aims of the Research
The aims of this study are as follows:
A1: To determine the extent of collaboration among HBBs.
A2: To determine the differences in collaborative behaviours across industry
sectors.
A3: To explore patterns of common associations between collaborative
behaviours exhibited by HBBs.
Collectively the insight provided via the above aims will provide an overview of the
current collaborative environment in which HBBs inhabit, in addition to illustrating
the areas in which collaboration is most required, thus providing direction for future
work.
2.0 Methodology
2.1 Research Structure
The study utilized a number of methods to assess the collaborative behaviours of
HBBs, presented as follows in the sequence which they were performed. Firstly,
summary statistics by frequency were used to develop an understanding of the degree
of collaboration exhibited by HBBs. Next, individual analysis was performed by
industry sector, using descriptive statistics to outline collaborative trends across a
range of industries. Lastly, association analysis was performed to detect trends and
frequent associations between the collaborative behaviours.
2.2 Data Overview
The data used for this research was provided by the Global Entrepreneurship Monitor
(Global Entrepreneurship Monitor, 2016), henceforth referred to as GEM. The 2012
release of the data was used due to the presence of year-specific questions concerning
the collaborative activities of the surveyed businesses, not found in prior or
subsequent releases of the dataset. The rationale behind the choice of using the GEM
dataset was twofold: firstly, the unique composition (among publically available
repositories) of the dataset which allows for the concurrent study of individual,
organizational and environmental variables, and secondly, as it provides access to
standard, consistent data relating to businesses from almost all OECD countries, thus
increasing the applicability of the findings produced.
2.3 Data Preparation
To identify usable cases for the study the original dataset was condensed on the basis
of three main conditions: firstly, the presence of values indicating that the business
was home-based. Secondly, the presence of data indicating the collaborative activities
of the business, and thirdly, the location of the business indicated as being within
OECD country, so that relative parity in terms of national economic conditions could
be assured. The total number of cases post data reduction was 3891, from a total of 20
countries (further detail can be found in Appendix A).
Variable Name Represented behaviour(s) Possible value
CollabProduce Production of goods or services with other
businesses or organizations
1 = Yes, 0 = No
CollabProcure Procurement of goods or materials with other
businesses or organizations
1 = Yes, 0 = No
CollabSellMarket Selling and/or marketing of goods or services
with other businesses or organizations
1 = Yes, 0 = No
CollabCreate Creating new goods or services with other
businesses or organizations
1 = Yes, 0 = No
CollabEffective Working with other businesses or organizations
to make the business more effective
1 = Yes, 0 = No
Table 1. Variables present in the GEM 2012 dataset representing collaborative
behaviour.
The dataset includes data on five different forms of collaborative activity, indicated
through the values contained in five variables, as shown in Table 1. The data in each
is represented by a Boolean value denoting a business’s participation in an activity.
While the behaviours identified within the data are not exhaustive, the scope of this
study is defined by the boundaries of the dataset, and is deemed satisfactory for the
purpose of identifying the general disposition towards collaboration demonstrated by
HBBs.
In the original dataset collaborative activities were split over ten variables, with each
behaviour represented by two variables – one for start-up businesses and another for
established businesses. As each pair of variables includes only one value, each pair
were consolidated into a single variable for analysis. An example of the data structure
and the associated interpretation of the variables can be seen in Table 2.
CollabPro
duce
CollabPr
ocure
Collab
SellMa
rket
CollabC
reate
CollabEff
ective
Interpretation
0 0 0 0 0 No collaborative activity
indicated
1 0 0 1 0 Some collaborative activity
indicated
1 1 1 1
1 All collaborative activities
indicated
Table 2. Example of variables within the GEM 2012 dataset.
Post data reduction there still remained a small quantity of missing values in the
collaboration variables (<5% for each). Imputation was therefore required to best
preserve the size of the dataset, with the use of the expectation-maximization (EM)
algorithm being chosen as the method best suited to the task, due to the ability of
algorithm to preserve the relationships between variables (Schaffer, 1997).
3.0 Findings and Discussion
3.1 Extent of Collaboration Among HBBs
To assess the extent of collaborative behaviours among HBBs, basic summary
statistics were produced, as seen in Table 3. The most prominent finding from the
summary statistics is the overall engagement in any collaborative behaviour by HBBs,
with over 75% of businesses (75.6%, shown in Table 3) collaborating in some way.
This is in contrast to the previous studies which indicated that the majority of HBBs
are indisposed to collaboration, due to the barriers impeding successful inter-firm
cooperation such as a lack of suitable partners, a lack of the required investment or the
fear of knowledge over-sharing (Casals, 2011),
Collaborative activity Percentage of HBBs engaged in activity (%)
Any collaborative activity 75.6
Production 49.8
Procurement 42.1
Selling/Marketing 43.6
Creation 26.7
Making business more effective 38.0
Table 3. Summary statistics of collaborative behaviours among HBBs.
Regarding the forms of collaboration engaged in, it can be seen that the most common
is working with others to produce goods or services, and the least common is working
with others to create new goods or services (as given in Table 3). This indicates that
collaboration among HBBs is primarily of a practical nature – utilizing it as a tool to
access resources not held internally or to derive transaction cost benefits via resource
pooling – as opposed to joint initiatives and ground-up collaborative product
development.
3.2 Analysis of Collaborative Propensity by Industry
An industry based analysis was performed in order to explore the nature of
collaborative activities among HBBs operating within various sectors. A double digit
International Standard Industrial Classification (ISIC) code (United Nations, 2014)
recorded within the GEM data was used as the industry identifier, with a range of
twelve industries being identified within the data, as seen in Table 4. Across each
industry two tests were performed: a breakdown of collaborative propensity by
percentage of industry total, and a collaborative activity breakdown illustrating the
ratios of industry members exhibiting or not-exhibiting each behaviour. A summary
of the results can be seen in Table 4.
Industry % of
businesse
s showing
no
collaborat
ive
behaviour
s
% of
businesses
showing
one or
more
collaborativ
e
behaviours
%
Deviatio
n from
aggregat
ed
industry
mean*
Most
common
collaborative
behaviour
(% engaged)
Least
common
collaborative
behaviour (%
engaged)
Agriculture,
Forestry, Fishing
22.8 77.2 1.6 Procurement
(49.9)
Creation
(20.9)
Mining,
Construction
18.0 82 6.4 Procurement
(56.8)
Creation
(26.0)
Manufacturing 19.0 81 5.4 Procurement
(53.8)
Creation
(27.8)
Utilization,
Transport
30.2 69.8 -5.8
Production
(51.2)
Creation
(21.8)
Wholesale trade 20.1 79.9 4.3 Procurement
(54.3)
Creation
(26.7)
Retail trade,
Hotels, Restaurants
24.6 75.4 -0.2 Procurement
(51.4)
Creation
(24.1)
Information,
Communication
22.8 77.2 1.6 Production
(55.4)
Procurement
(35.3)
Financial
intermediation,
Real Estate
25.7 74.3 -1.3 Selling/Mark
eting (55.8)
Procurement
(25.7)
Professional
services
25.7 74.3 -1.3 Production
(56.3)
Procurement
(31.8)
Administrative
services
34.9 65.1 -10.5 Production
(42.1)
Creation
(27.2)
Government,
Health, Education,
Social services
24.7 75.3 -0.3 Production
(47.1)
Creation
(27.8)
Personal/
Consumer service
24.4 75.6 0 Production
(57.3)
Creation
(31.7)
Table 4. Summary of collaborative behaviours across industries. (*Non-weighted mean
of the percentage of collaborative businesses across industries)
Across all industries, at least 65% of HBBs engaged in at least some form of
collaboration (65.1% being the lowest value, shown in Table 4) with the mean across
industries being 75.6%, calculated from the data shown in Table 4. The most common
form of collaboration (by frequency) across all industries is working with other
businesses to produce goods or services. The least common form of collaboration (by
frequency) is working with other businesses to create new goods or services. While
the majority of industries achieve similar collaborative propensities relative to the
mean, those outside of the standard deviation from the mean (which is calculated to be
4.5) include “Mining, Construction” and “Manufacturing” – both of which show a
higher than average inclination toward collaborative activity, in addition to
“Utilization, Transport” and “Administrative Services”, both of which demonstrate a
lower than average inclination toward collaborative activity.
Of note is the lack of focus on collaborative creation of new goods or services, which
runs as a counterpoint to the theory that working together to achieve innovation and
generate new products is the primary purpose of collaborative activity among SMEs.
(Casals, 2011; Narula, 2004).
3.3 Association Pattern Analysis
To explore the relationships which exist between the multiple forms of collaboration,
two key areas were investigated; the associations between the varying activities and
the likelihood of their common occurrences. To achieve an understanding of the
regularity of certain combinations of collaborative behaviours, a frequency pattern
(FP) tree was compiled, a method commonly used for the identification of frequently
occurring itemsets within a dataset (Han & Kamber, 2006). illustrating the number of
incidences of behaviours one to five (as shown in Table 1) occurring together, up to a
total of three concurrent behaviours. The minimum support cost was set at one fifth of
the number of cases, 778. Table 5 details the frequently grouped item sets which
achieved that threshold.
The measures of support and confidence were utilized as a method of identifying the
most prominent relationships within a dataset. Support can be seen as measure of
frequency, indicating the proportion of cases exhibiting a particular combination of
behaviours. Confidence designates the amount of times a statement of association can
be seen to be correct. From the data it can therefore be seen that the activities of
“Production” and “Selling/Marketing” jointly occur in 35% of all cases, yet based on
the presence of one of these activities it can be predicted with a 61.2% confidence that
the other will also be present in a given case.
Combination Support Confidence
Production, Selling/Marketing 0.35 61.2%
Production, Procurement 0.28 55.8%
Production, Making business more effective 0.26 52.8%
Selling/Marketing, Making business more
effective
0.26 58.7%
Procurement, Selling/Marketing 0.24 57.6%
Selling/Marketing, Creation 0.22 50.9%
Procurement, Making business more effective 0.22 52.0%
Production, Creation 0.22 43.8%
Table 5. The most numerous collaborative combinations ranked by support.
The association analysis identified that in addition to “Production” being the most
prevalent form of collaboration among HBBs when taken in isolation, it is
additionally the behaviour most likely to occur in combination with others. The
overall spread of behaviours however is diverse, with only four behavioural
combinations occurring in over 25% of cases. The following phase involved
determining the probabilities of a behaviour occurring based on the presence of one or
more other behaviours. Table 6 displays the behaviours most likely to occur in
conjunction with others.
Behaviours (Dependent | Independent(s) Conditional Probability
Making business more effective | (Selling/Marketing & Creation) 0.74
Making business more effective | (Production & Creation) 0.73
Making business more effective | Creation 0.69
Selling / Marketing | (Production & Procurement) 0.67
Making business more effective | (Procurement &
Selling/Marketing)
0.66
Making business more effective | (Production &
Selling/Marketing)
0.64
Selling/Marketing | Production 0.61
Creation | (Production & Selling/Marketing) 0.61
Table 6. Most probable incidences of behaviours occurring in combination.
The figures shown in Table 6 help to illustrate a number of trends shown in the data.
One combination of behaviours which is of interest is “Making business more
effective” and “Creation”, which in isolation are the two behaviours least likely to
occur (see Table 3) but possess a high probability (0.69) of occurring in tandem.
Another key trend revealed via the probability analysis is the prominence of “Making
business more effective”, with 5 of the 8 most probable behavioural combinations
including this behaviour, which when compared with the base rate of occurrence 38%
(shown in Table 3) indicates the increased likelihood of this behaviour to occur in
conjunction with other behaviours as opposed to in isolation. One explanatory
hypothesis for this phenomenon is that HBBs with existing willingness to collaborate
in areas such as joint purchasing and outsourced production of goods are more also
more open to receiving outside assistance in improving their internal business
processes.
4.0 Conclusions
The study has shown that collaboration among HBBs is widespread, with over 75%
exhibiting one or more collaborative behaviours, with collaborative production,
procurement and selling/marketing being the most frequent forms of collaboration
among HBBs. Equally, this study has shown that the collaborative behaviours of
HBBs vary considerably, with even the least frequently occurring behaviour –
collaborative creation – being exhibited by over 26% of HBBs.
Furthermore, collaboration is a practice not limited to a small selection of industries
and is instead commonplace across all industry sectors, with all industry’s possessing
at least a 65% rate of collaboration. The most collaboratively inclined industries were
shown to be the mining/construction and manufacturing industries, both of which
possessed collaboration rate in excess of 80%. Additionally, this study has provided
insights into the nature of collaboration in HBBs, illustrating which behaviours are
likely to occur in combination with others. This analysis has highlighted a number of
trends within the data, including the increased likelihood of collaboration to make a
business more effective occurring in conjunction with other behaviours, and the close
relationship displayed between the behaviours of collaborative production and
collaborative selling/marketing.
Of particular note is that a number of the findings generated by this study - concerning
both the extent of and the nature of HBB collaboration - are far enough removed from
those existing in the current literature on SME collaboration to reinforce the theory
that HBBs operate in a different manner to SMEs and must therefore be considered as
a separate entity (Clark & Douglas, 2014). By addressing the subject of HBB
collaboration from a data analytics perspective, the findings illustrate the reliance
shown by HBBs on collaborative activities, and are able to clearly demonstrate that
HBBs located within OECD countries are actors heavily engaged in joint enterprise
and inter-organizational cooperation.
5.0 Further research
The future research will comprise a more involved analysis of the areas covered in
this study, including studying HBB collaboration on the basis of intensity and
business maturity. Following this, classification of businesses into like groups on the
basis of their collaborative activity will be performed by means of cluster analysis,
with the aim of using the identified clusters to develop an understanding of common
factors which exist between collaboratively inclined HBBs.
Appendix A
Table 7 displays a breakdown of the composition of businesses utilized in the study
by country of origin. Businesses from a total of 20 OECD countries were used in the
study, a number limited by valid cases in dataset post data reduction, as detailed in
section 2.3.
Country Number of valid cases Percentage of total cases (%)
Spain 997 25.6
Netherlands 383 9.8
Poland 211 5.4
Estonia 209 5.4
Austria 202 5.2
Latvia 197 5.1
Hungary 175 4.5
Sweden 174 4.5
United Kingdom 171 4.4
Germany 171 4.4
Finland 171 4.4
Ireland 169 4.3
Slovakia 137 3.5
Slovenia 118 3
Denmark 98 2.5
Belgium 94 2.4
Israel 74 1.9
Italy 68 1.7
Greece 46 1.2
Portugal 26 0.7
Table 7. Breakdown of valid cases by country
References
Camarinha-Matos, L.M., Abreu, A. (2007) Performance indicators for collaborative
networks based on collaboration benefits. Production planning and
control, 18(7), 592-609.
Casals, F. E. (2011) The SME Co-operation Framework: a Multi-Method Secondary
Research Approach to SME Collaboration, 2010 International Conference on
E-Business, Management and Economics, IPEDR vol. 3, IACSIT Press, Hong
Kong.
Clark, D. N., Douglas, H. (2014) Micro-enterprise growth: Lessons from home-based
business in New Zealand. Small Enterprise Research, 21(1), 82-98.
Enteprise Nation Home Business Report (2014) https://en-production-
assets.s3.amazonaws.com/2014/10/22/08/47/31/599/Home_Business_Survey.
pdf, last accessed 2016/12/05.
Global Entrepreneurship Monitor (2016) http:/www.gemconsortium.org, last accessed
2017/06/20
Han, J, Kamber, M. (2006) Data Mining: Concepts and Techniques (2nd Ed.), Morgan
Kaufmann.
Johannisson, B. (1987) Beyond process and structure: social exchange
networks. International Studies of Management & Organization, 17(1), 3-23.
Mason, C. (2010) Home based business: Challenging their Cinderella Status, Small
Enterprise Research, 17(2), 104-111.
Mason, C.M., Carter, S., Tagg, S. (2011) Invisible businesses: The characteristics of
home-based businesses in the United Kingdom. Regional Studies, 45(5), 625-
639.
Narula, R. (2004) R&D collaboration by SMEs: new opportunities and limitations in
the face of globalisation. Technovation, 24(2), 153-161.
Schafer, J. L. (1997) Analysis of incomplete multivariate data. CRC press.
United Nations Statistics Division (2014) https://unstats.un.org/unsd/cr/registry/isic-
4.asp, last accessed 2017/09/07.
