816255 JIN Journal of Information TechnologyAsatiani et al.

Research Paper
JIT
Journal of Information Technology

Uncovering the nature of the 2019, Vol. 34(1) 39­–58

© Association for Information
Technology Trust 2019
relationship between outsourcing
motivations and the degree of Article reuse guidelines:
sagepub.com/journals-permissions

outsourcing: An empirical study on

https://doi.org/10.1177/0268396218816255
DOI: 10.1177/0268396218816255
Journals.sagepub.com/jinf

Finnish small and medium-sized

enterprises

Aleksandre Asatiani1, Esko Penttinen2 and Ashish Kumar2

Abstract
Prior literature has identified several outsourcing motivations, such as cost reduction and access to expertise, and
deciphered the influence of these variables on outsourcing decisions. In another stream of outsourcing studies, researchers
have gauged the degree of outsourcing, unearthing how companies may choose to outsource a set or processes instead
of the whole business function. In this article, we draw on both of these streams of outsourcing research to study the
relationship between outsourcing motivations and the degree of outsourcing within a particular business function. We
probe the effect of nine motivation items on outsourcing decision through an empirical study using survey data gathered
from 337 small and medium-sized enterprises. We find that cost reduction, a focus on core competence and business/
process improvements are all associated with a higher degree of outsourcing, but interestingly, access to expertise is
negatively associated with the degree of outsourcing. This finding suggests that companies that outsource mainly to
acquire external expertise outsource only a limited number of processes within a specific business function. Our main
theoretical contribution lies in uncovering the dynamic nature of outsourcing motivations, meaning that as companies
outsource a larger degree of their business processes, some motivation items become more accentuated and others
fade in importance.

Keywords
Business process outsourcing, outsourcing motivation, degree of outsourcing, survey, accounting

Introduction outsourcing literature (Lacity, Khan, & Yan, 2016; Lacity,

Khan, Yan, & Willcocks, 2010; Lacity, Solomon, Yan, &
Business process outsourcing (BPO) is an act of delegation Willcocks, 2011) offer a systematic and holistic summary
of one or more information-intensive business processes to of evidence behind the effects of the most important moti-
a third-party provider (Borman, 2006; Luo, Zheng, & vation items on outsourcing decisions. These company-
Jayaraman, 2010). Companies commonly outsource pro- level analyses have improved our understanding of how
cesses in non-core business functions, such as finance and
accounting, call centres and human resources, to third-party
service providers for various reasons. The extant literature 1Aston University, UK
identifies a plethora of these outsourcing motivations, the 2Aalto University, Finland
most widely cited being access to expertise (Currie,
Michell, & Abanishe, 2008; Lam & Chua, 2009), cost Corresponding author:
Aleksandre Asatiani, Operations & Information Management
reduction (Borman, 2006; Saxena & Bharadwaj, 2009) and Department, Aston Business School, Aston University, Aston Triangle,
scalability (Redondo-Cano & Canet-Giner, 2010). To com- Birmingham B4 7ET, UK.
plement and synthesize these studies, recent reviews of Email: [email protected]
40 Journal of Information Technology 34(1)

companies differ in their motivations to outsource business Our results show that companies that outsource more tasks
processes. seek to focus on core competence, whereas companies that
The BPO market has been growing steadily in recent outsource only a limited number of tasks mainly seek exter-
years, reaching estimated US$322 billion by the end of nal competence and assurance.
2016 (Snowden & Fersht, 2016), as cloud computing and The remainder of the article is organized as follows.
other emerging technologies offer new opportunities to After this introduction, we proceed with a literature review
BPO providers to consolidate and grow their business on outsourcing motivations. In the third section, we build
(Singh & Tornbohm, 2016). Developments in the outsourc- our conceptual framework, which consists of a set of moti-
ing market have enabled greater flexibility in designing vation variables and the degree of outsourcing. In the fourth
outsourcing deals. For example, in accounting, cloud com- section, we present our empirical study. In the fifth section,
puting provides a platform where two parties (a client com- we report the aggregate-level findings of our empirical
pany and an outsourcing service provider) can jointly study, and, in the sixth section, the process group-level
access the data and workflow in real time. Endowed with findings. In the remaining sections, we discuss the findings
greater transparency and control through, this new breed of and avenues for further research.
accounting information systems (AIS) allows the outsourc-
ers to make outsourcing decisions on a task level instead of
Review of outsourcing motivations
outsourcing the whole business function (Asatiani, Apte,
Penttinen, Rönkkö, & Saarinen, 2014). For example, in The outsourcing research has focused on two main areas: the
accounting outsourcing, some may outsource a particular decision to outsource and outsourcing outcomes. The stream
payroll-related task (e.g. payroll calculations), while others of literature on the decision to outsource addresses questions
may choose to outsource the preparation and submission of of why outsource (motivations), what to outsource (what
financial statements. type of processes; transaction attributes) and how to out-
This emerging complexity and flexibility in outsourcing source (e.g. how much to outsource; how to implement con-
calls for a revised understanding of outsourcing motiva- trol mechanisms) (Dibbern, Goles, Hirschheim, & Jayatilaka,
tions, which requires us to delve deeper from a company- 2004; Lacity et al., 2011). The stream on outsourcing out-
level analysis into a task-level analysis. Tangential to our comes has sought to improve our understanding of how out-
research, Dibbern, Chin, and Heinzl (2012) challenged the sourcing impacts business performance, which factors
modular view of outsourcing, where an outsourcing deci- contribute to the perceived success of outsourcing arrange-
sion on one process is viewed as independent from deci- ments and how outsourcing impacts the quality of the rela-
sions on other processes. Dibbern and colleagues observed tionships between outsourcing parties (Chou,
a systemic influence on outsourcing motivations in infor- Techatassanasoontorn, & Hung, 2015; Lee & Kim, 1999).
mation systems (IS) outsourcing. We argue that a similar We focus on outsourcing decisions and, more specifi-
phenomenon can be observed in BPO. While business cally, on outsourcing motivations (why and what to out-
functions such as accounting may have a modular structure, source). The existing literature has identified a plethora of
motivations to outsource a particular process are not inde- motivations, ranging from cost reduction to business trans-
pendent of the context of other processes. Therefore, we formation (Mani, Barua, & Whinston, 2010). In a series of
theorize that motivations to outsource particular processes articles, Lacity et al. (2010), Lacity et al. (2011) and Lacity
within a business function are related to the degree of out- et al. (2016) systematically reviewed the body of literature
sourcing. Motivated by the recent developments in out- on outsourcing. Lacity et al. (2011) observed a total of 19
sourcing markets and our limited current understanding of main outsourcing motivations used in the literature.
the link between outsourcing motivations and the degree of However, when the authors analysed the findings on the
outsourcing, our main research question is as follows: effect of motivations on the decision to outsource, they
found that only a fraction of outsourcing motivations
Research Question. What is the relationship between moti- received consistent empirical support (Lacity et al., 2016;
vations to outsource and the degree of outsourcing? Lacity et al., 2011). These motivations were cost reduction,
access to skills/expertise, focus on core competence, busi-
To address this research question, we reviewed the exist- ness process improvements, scalability, rapid delivery and
ing literature on outsourcing motivations to build a concep- concern for security (negative effect). An exploratory, qual-
tual model with nine motivation items: cost reduction, itative pre-study on contextual BPO motivations conducted
focus on core competence, access to expertise, process by the authors of this article largely supports these findings
improvements, scalability, rapid delivery, ease of use, fear (Asatiani & Penttinen, 2016).
of losing control (negative) and concern for security (nega- Existing research has analysed BPO motivations in vari-
tive). We collected empirical data on the outsourcing of ous contexts, focusing on specific industries (e.g. Currie
accounting functions and the outsourcing motivations in et al., 2008; Lam & Chua, 2009), geographical locations
337 Finnish small and medium-sized enterprises (SMEs). (e.g. Martinez-Noya, Garcia-Canal, & Guillen, 2012) and
Asatiani et al. 41

business processes (e.g. Redondo-Cano & Canet-Giner, motivations and various outsourcing configurations (cited
2010). Many of the existing studies treat motivation as a above), we note a lack of a comprehensive understanding
static concept with a straight, linear effect on outsourcing. of how individual motivation variables interact with the
These studies test the relationship between outsourcing degree of outsourcing.
decisions and motivations without considering the degree
of outsourcing within a business function. However, other
Hypothesis development
studies tend to view an outsourcing decision as a choice
among only three options: selective outsourcing, total out- To study the relationship between motivations and the degree
sourcing and total insourcing (e.g. Dahlberg, Nyrhinen, & of outsourcing, we put forward nine hypotheses. We selected
Santonen, 2006; Lee, Miranda, & Kim, 2004). We believe a set of motivation variables for our analysis based on the
that such an oversimplification of the outsourcing arrange- reviews in existing literature and our pre-study (Asatiani &
ment prevents us from fully understanding how outsourcers Penttinen, 2016). We selected motivation variables that have
match their motivations with a particular outsourcing been extensively studied, with empirical results supporting
configuration. the effects of these motivations on outsourcing decisions. To
Among the few studies that address the influence of identify these motivation variables, we consulted the litera-
motivations on the degree of outsourcing, Gewald and ture reviews on outsourcing (Lacity et al., 2016; Lacity et al.,
Dibbern (2009) studied transaction process outsourcing by 2010; Lacity et al., 2011). These reviews provide an exten-
banks and compared motivations among organizations that sive analysis of the outsourcing motivations research and
had already outsourced, those that planned to outsource and present a summary of motivations used in the current
those that had decided against outsourcing. The authors literature.
found that organizations that had already made the decision The final set of variables included nine motivation
to outsource portrayed a more balanced recognition of items: (1) cost reduction, (2) focus on core competence, (3)
motivations than organizations that had decided against access to expertise, (4) process improvements, (5) scalabil-
BPO. Closer to the scope of this article, Redondo-Cano and ity, (6) rapid delivery, (7) ease of use, (8) fear of losing
Canet-Giner (2010) studied the outsourcing of R&D activi- control (negative) and (9) concern for security (negative).
ties in the agrochemical industry and measured both the Table 6 in Appendix A presents a summary of the selected
degree of outsourcing and motivations. They observed that outsourcing motivations in comparison with the existing
in cases of highly outsourced functions, companies were literature and the pre-study.
motivated by the lack of resources and production capabili- Cost reduction is one of the most cited and examined
ties, whereas in cases of functions with a lower degree of motivations for BPO (Lacity et al., 2016). Cost reduction
outsourcing, outsourcers were motivated by acquiring and describes the client company’s desire to reduce or control
maintaining knowledge that was important to the core of the costs of a business process (Borman, 2006). The ration-
their business while increasing the economic efficiency of ale behind cost reduction through outsourcing lies in econ-
non-core components of the function. omies of scale (Poppo & Zenger, 1998), where specialized
Weigelt (2009) studied the relationships among compa- market agents can minimize production costs by develop-
nies’ degree of outsourcing, integrative capabilities and ing production capacity and by aggregating demand across
market performance in the context of business process- several buyers. From this, we can assume that a BPO ser-
enhancing technologies. Weigelt’s findings suggest that a vice provider (a specialized market agent) possesses supe-
greater reliance on outsourcing (high degree of outsourcing rior production capabilities for non-core business processes
in a given function) leads to impediments in integrative compared with the outsourcer company. Therefore, we
capabilities and market performance. In a later study, while argue that companies that are highly motivated by cost
evaluating outcomes of outsourcing, Weigelt and Sarkar reduction will pursue a higher degree of outsourcing. Our
(2012) observed that companies outsourcing business pro- first hypothesis, therefore, reads as follows:
cesses that are reliant on emerging technologies have to
make the trade-off between efficiency and adaptability that Hypothesis 1. A higher level of motivation to reduce
occurs when the degree of outsourcing increases. The costs through BPO leads to a higher degree of
authors concluded that differing objectives (in this case, outsourcing.
efficiency and adaptability) require divergent governance
structures (Weigelt & Sarkar, 2012), and therefore, the Focus on core competence describes the client compa-
degree of outsourcing should be adjusted according to the ny’s desire to outsource non-core tasks in order to focus on
objectives of the company. These findings further indicate the core part of the business (Premuroso, Skantz, &
the need for a deeper understanding of how motivations Bhattacharya, 2012). An example of such a non-core task is
interact with outsourcing decisions regarding the degree of accounting, which lies outside the core business activities
outsourcing in a particular business function. Although a for the majority of companies. Outsourcing such processes
few studies have investigated the relationship between allows a company to reallocate freed-up resources to more
42 Journal of Information Technology 34(1)

productive and/or value-generating tasks (Gewald & BPO providers possess a large pool of specialized assets that
Dibbern, 2009). In contrast, for a third-party service pro- could be allocated to a task on an on-demand basis, whereas
vider, the outsourced tasks are a part of the core business client companies would have to invest time and resources in
(e.g. an accounting firm’s core competence is accounting). order to scale up the operations of any given business func-
Thus, the providers possess superior competence needed to tion. In contrast, scaling down involves the costly realloca-
complete said tasks. Therefore, we argue that companies tion and/or downsizing of human resources, which could be
seeking to focus on their core business will outsource more a painful process. Therefore, we hypothesize the following:
business processes:
Hypothesis 5. A higher level of motivation to scale a
Hypothesis 2. A higher level of motivation to focus on business process through BPO leads to a higher degree
core competence through BPO leads to a higher degree of outsourcing.
of outsourcing.
Rapid delivery describes the outsourcer’s motivation to
Access to expertise refers to an outsourcer’s desire to speed up the delivery of services by outsourcing its compo-
access a service provider’s expert knowledge that is not nents (Lacity & Willcocks, 2016). As BPO providers can
available internally (Lacity et al., 2016). Outsourcing to devote greater resources to completing a task, client com-
access expertise is justified when a company lacks inter- panies expect faster delivery compared with an in-house
nal expertise in a particular area and is unable or unwill- arrangement (Bandyopadhyay & Hall, 2009; Freytag,
ing to develop the needed skills in-house (Lam & Chua, Clarke, & Evald, 2012). Thus, we posit the following
2009). Developing specialized knowledge and assets hypothesis:
internally requires considerable investments (Jacobides,
2008). In contrast, BPO service providers are able to Hypothesis 6. A higher level of motivation to achieve
provide highly qualified assets (e.g. payroll and tax spe- rapid delivery through BPO leads to a higher degree of
cialists) by focusing on specialized resources and per- outsourcing.
forming a continuous development of production
capacity (Gewald & Dibbern, 2009). Therefore, we posit Ease of use refers to the outsourcer’s desire to simplify
the following hypothesis: the interaction with a business process and/or its output
through, for example, improved customer support or
Hypothesis 3. A higher level of motivation to access enhanced software. Improving the ease of use of a business
expertise through BPO leads to a higher degree of process through outsourcing may also decrease the level of
outsourcing. frustration associated with that specific business process
(Lacity et al., 2016; McKenna & Walker, 2008). It is
Process improvements describe the client company’s assumed that, in order to stay competitive, BPO providers
desire to engage a BPO service provider to improve and are incentivized to focus on delivering superior customer
develop a business process (Lacity et al., 2011). The moti- service to their clients. Therefore, outsourcers would expect
vation to outsource in order to reap process improvements a better service from external BPO providers compared to
is often associated with the desire to seek efficiency and their in-house equivalents. We included this motivation fac-
effectiveness gains (Buco et al., 2004; Gewald & Dibbern, tor in our conceptual framework based on our pre-study
2009). Arguably, third-party providers are able to better conducted within the same population of SMEs (Asatiani &
organize the outsourced processes due to asset specializa- Penttinen, 2016). We hypothesize the following:
tion, extensive experience and economies of scale.
Therefore, a higher degree of outsourcing could carry Hypothesis 7. A higher level of motivation to achieve
higher potential for efficiency gains (Weigelt & Sarkar, ease of use through BPO leads to a higher degree of
2012). Based on the above, we propose the following outsourcing.
hypothesis:
In addition to the above-presented motivations for out-
Hypothesis 4. A higher level of motivation to improve a sourcing, we include two motivation variables that deter
process through BPO leads to a higher degree of outsourcing. Here, our goal is to examine whether these
outsourcing. motivations restrict companies from outsourcing a larger
number of business processes—in other words, leading to a
Scalability describes the outsourcer’s desire to scale up/ lower degree of outsourcing.
down a process depending on the demand (Currie et al., Fear of losing control describes the concerns that the
2008; Lacity et al., 2016). Again, due to the economies of client company might have regarding maintaining control
scale, BPO providers are able to provide the level of scalabil- over the outsourced tasks (Lacity et al., 2016). Transferring
ity, which outsourcers normally cannot achieve internally. control over a number of tasks to a third-party provider has
Asatiani et al. 43

been identified as a significant risk factor that may result in well-developed accounting outsourcing market. Second,
subpar performance and disruption in internal processes of accounting practices offer a clean, well-defined, docu-
a company (Sanders, Locke, Moore, & Autry, 2007). mented and standardized environment to study. The bound-
McKeen and Smith (2011) found that the fear of losing aries of the processes are well defined, and the outcomes are
control is an important factor in guiding organizations to often standardized (e.g. financial statements or tax reports).
choose internally controlled shared service centres over The outsourcing decision makers can be clearly defined, and
outsourced ones. Moreover, Bhagwatwar, Hackney, and they are able to recognize these processes easily. All the
Desouza (2011) observed that the desire to regain control above allow the use of more controllable data collection
over tasks was one of the major factors contributing to methods and contribute to the content and face validity of
backsourcing. Building on the negative impact of fear of the responses. Third, the accounting function is well suited
losing control on outsourcing decisions, we thus hypothe- for modularization as it includes a variety of processes
size the following: related to sales, purchases, payroll, payments and reporting.
The work needed to accomplish these processes can be split
Hypothesis 8. A higher level of fear of losing control between a client company and an accounting firm (BPO
over the business process leads to a lower degree of provider). In fact, it is common for companies to selectively
outsourcing. choose a set of accounting processes to outsource, instead of
merely opting for full outsourcing.
Concern for security encompasses the client company’s The accounting function can be divided into five major
concerns regarding the privacy and safety of data and/or parts: sales recording, purchases recording, payroll pro-
intellectual property associated with the outsourced tasks cessing, preparation of interim and annual reports, and pay-
(Lacity et al., 2011). Security considerations constitute an ments (Everaert, Sarens, & Rommel, 2007, 2008). We
important part of the outsourcing decision and contribute to operationalize the accounting function through a set of 22
the subsequent success of the outsourcing project processes (Table 7, Appendix B) that could potentially be
(Nassimbeni, Sartor, & Dus, 2012). Concern for security is outsourced to an accounting firm. All of these processes fall
negatively related to a company’s desire to outsource under one of the five major parts of the accounting func-
(McIvor, Humphreys, McKittrick, & Wall, 2009). In their tion. Practitioners in the accounting industry in Finland use
review of the BPO literature, Lacity et al. (2011) proposed, the same (or similar) set of processes to make outsourcing
‘the more concern for security or intellectual property, the arrangements between a client and an accounting firm.
less likely a client firm chose outsourcing’. Following this Furthermore, the set of 22 processes is aligned with the
proposal, we posit the following hypothesis : standard contract terms between service providers and cli-
ent companies recommended by The Association of Finnish
Hypothesis 9. A higher level of security concerns over Accounting Firms. Therefore, the decision makers in com-
the business process leads to a lower degree of panies that have undertaken any degree of accounting out-
outsourcing. sourcing are familiar with this set of 22 processes.

Thus, we hypothesize that the nine motivation variables

Data collection
identified in earlier reviews (Table 6) affect the degree of
outsourcing within one business function. Figure 1 presents We collected the data in Finland, where accounting outsourc-
a summary of our hypotheses. ing is a €960 million industry, with almost 4300 service pro-
viders on the market (The Association of Finnish Accounting
Firms, 2017). There are 356,790 enterprises in Finland, the
Empirical study majority of which are micro-enterprises (89.2%) with less
We study BPO in the context of accounting, which is suita- than five employees, and non-micro SMEs (10.6%) with less
ble for this research for three reasons. First, in most coun- than 250 employees (Statistics Finland, 2017). The micro-
tries, regardless of the type of business or industry, enterprises and SMEs are the major consumers of accounting
companies are mandated by law to record their transactions outsourcing services in Finland. Accounting in Finland is
and conduct the associated accounting processes, such as highly regulated and standardized across all industries, yet
financial and tax reporting. Companies that employ people accounting outsourcing market is highly competitive. The
need to make salary calculations and payments, and again, majority of accounting outsourcing contracts operate based
they need to record those salary payments in their book- on some version of the service agreement (see Table 8 in
keeping. Thus, every company faces a decision whether to Appendix B) provided by The Association of Finnish
conduct these processes in-house or outsource them to Accounting Firms (2014). Further facilitating outsourcing of
accounting firms. Indeed, outsourcing is a common prac- accounting processes, there are currently 18 major, distinct
tice, especially among SMEs (BDO Finland, 2015). This AIS on the Finnish market (The Association of Finnish
proliferation of accounting outsourcing has led to a Accounting Firms, 2015).
44 Journal of Information Technology 34(1)

Figure 1. Summary of hypotheses.

We collected survey data on SMEs. We refer to the defi- not outsource any of the 22 accounting processes. We also
nition set by the European Commission, which states that excluded the companies that fell outside of the European
an SME is a company with no more than 250 employees Commission’s definition of an SME. The final sample used
and an annual turnover of less than €50 million (European for our analysis includes 337 responses. Table 1 presents
Commission, 2003). The study was conducted in coopera- the sample demographics summary.
tion with the Federation of Finnish Enterprises, the largest The survey included three parts: (1) background ques-
business federation in Finland, which unites more than tions, collecting the basic information about the SME and the
116,000 enterprises (The Federation of Finnish Enterprises, respondent’s role in it; (2) outsourcing questions, enquiring
2016). The Federation focuses particularly on SMEs. about the accounting outsourcing arrangements of the SME;
A representative of the Federation distributed the survey and (3) outsourcing motivation questions, addressing general
questionnaire through email. The email message included a and process-specific motivations for outsourcing. The
cover letter signed by both the representative of the English translation of the survey questions used for this study
Federation and the authors of this article, and it contained a is provided in Appendix C.
web link to the survey instrument. The cover letter included
an invitation to participate in the study, a brief explanation
of the purpose of the research and a statement regarding the
Measures
anonymity of the responses. Both the cover letter and the In this study, all motivation variables were measured on a
survey questionnaire were written in Finnish by a native seven-point Likert-type scale ranging from ‘No influence
speaker (one of the authors) and were reviewed by all three on the outsourcing decision’ to ‘Very strong influence on
authors and the representative of the Federation. the outsourcing decision’. The dependent variable, ‘degree
The survey was sent to a random sample of 5000 SMEs of outsourcing’, was derived based on the 22 survey items
selected from the Federation’s member database. The sur- corresponding to the accounting processes (Table 7) out-
vey was open for 4 weeks. In addition to sending the origi- sourced by a respondent. Therefore, the dependent variable
nal invitation to the survey, the representative of the represented the sum of outsourced processes and thus was
Federation distributed two follow-up messages during this measured on a scale of 1 to 22. As we are studying the
period. In total, 460 complete responses were returned, for effect of motivations on the degree of outsourcing, measur-
a response rate of 9.2%. As we were interested in studying ing the dependent variable based on the count of outsourced
the degree of outsourcing, we excluded companies that did processes was justified.
Asatiani et al. 45

Table 1. Sample demographics (n = 337). such problems (Kauffman, Techatassanasoontorn, & Wang,
2012). We adopt a negative binomial model that allows our
% dependent variable, degree of outsourcing, to be more dis-
Job of a respondent persed, which is often the case in the SME context.
CEO/director 82.78 We let CntOutProci be a random variable capturing the
Accountant 8 count of outsourced processes by the company i. Then, we
HR manager 1.7 specify the following functional form, f, to model the num-
Other 7.52 ber of outsourced processes
Industry
Agriculture 1.5  COSi , FOCi , EXPi , IMPi , SCLi ,
Manufacturing 43 CntOutProci = f 
Services 55  RPDi , EASi , CTLi , SECi ,
Size CLOUDi , EMPLi , SMEAGEi ,
(1)
1–20 61 MNCi , TURNOVERi ,
21–100 34
101–250 5 INDUSTRY_TYPEi )

Table 2. Descriptive statistics. where variables COS, FOC, EXP, IMP, SCL, RPD, EAS,
CTL and SEC refer to the company’s motivations, namely,
Variable Mean SD cost reduction, focus on core competences, access to exper-
Degree of outsourcing (count 1–22) 9.25 5.46 tise, business improvements, scalability, rapid delivery, ease
Cost reduction (Likert-type scale 1–7) 4.09 1.69 of use, fear of losing control and security concerns, respec-
Focus on core competence (Likert-type scale 5.80 1.41 tively. Variables CLOUD, EMPL, SMEAGE, MNC,
1–7) TURNOVER and INDUSTRY_TYPE are control variables in
Access to expertise (Likert-type scale 1–7) 6.07 1.11 the model. CLOUD is a dummy variable for the cloud plat-
Process improvements (Likert-type scale 1–7) 5.38 1.47 form that takes a value of 1 if the company has cloud access
Scalability (Likert-type scale 1–7) 4.88 1.67 and 0 otherwise. EMPL is the total number of employees in
Rapid delivery (Likert-type scale 1–7) 5.20 1.70 the company, SMEAGE captures the age of the company in
Ease of use (Likert-type scale 1–7) 5.68 1.36 years, MNC is a dummy for multinational company (which
Fear of losing control (Likert-type scale 1–7) 3.21 1.83 takes a value of 1 if it operates in more than one country and
Concerns for security (Likert-type scale 1–7) 2.68 1.64
0 otherwise) and TURNOVER captures the company’s
annual turnover. Finally, INDUSTRY_TYPE is a dummy
variable for the type of industry. Industry type is divided
Based on the literature review, we measured nine moti-
into three categories, primary, secondary and tertiary, where
vation variables: (1) cost reduction, (2) focus on core
the primary sector serves as the basis for comparison.
competences, (3) access to expertise, (4) process improve-
Due to the special nature of our dependent variable,
ments, (5) scalability, (6) rapid delivery, (7) ease of use,
CntOutProc, we cannot apply the ordinary least squares
(8) fear of losing control and (9) concern for security
method. Therefore, we view this process as a negative bino-
(Table 2). Correlation coefficients for all model variables
mial1 and model it as follows
are shown in Table 9 in Appendix D. In addition to the
main variables, we selected six control variables: (1) use
e −λi λ CntOutProci

of cloud-based IS for accounting, (2) company size by f ( CntOutProci ) = i

CntOutProci ! (2)
employees, (3) turnover, (4) age of the company, (5) mul-
tinational operations and (6) industry. for CntOutProci = 1, 2,  , n
In addition to measuring overall motivations against the
degree of outsourcing, we asked the respondents to indicate where λi is specified as follows
the most important outsourcing motivation for each process
that they had indicated to outsource. This was done through λ i = e( xi β + εi ) (3)
a drop-down menu of the seven positive motivation items.
where xi are variables specified in the function in Equation
Findings and analysis—aggregate (1), and β are the corresponding response parameters. We
write λi as λi = µiδ i . To obtain an unconditional distribu-
level tion, as in Equation (2), we specify the gamma distribution
The objective of our research is to study the degree of BPO. for δ i . The details of this specification are provided in
Thus, the topic area falls under the study of event frequen- Appendix E. We estimate the model using the maximum
cies, and a count data analysis is an apt method to analyse likelihood method.
46 Journal of Information Technology 34(1)

Table 3. Results of count regression.

Variable Estimate SE
Motivation (Intercept) 1.204*** 0.3208
variables Cost reduction COS 0.044** 0.020
Focus on core competence FOC 0.121*** 0.028
Access to expertise EXP −0.153*** 0.033
Business/process improvements IMP 0.074** 0.029
Scalability SCL 0.033 0.025
Rapid delivery RPD 0.022 0.026
Ease of use EAS −0.006 0.031
Fear of losing control CTL −0.037* 0.020
Concerns for security SEC −0.030 0.023
Control Use of cloud-based IS CLOUD 0.302*** 0.058
variables Number of employees EMPL 0.001 0.001
Age of the company SMEAGE −0.005*** 0.002
Multinational operations MNC −0.167** 0.066
Turnover TURNOVER 0.000 0.001
Secondary sector (services) INDUSTRY_2 0.548** 0.258
Tertiary sector (manufacturing) INDUSTRY_3 0.581** 0.256

AIC: Akaike information criterion.

Dispersion parameter for negative binomial (5.9364) family taken to be 1. Null deviance: 510.74 on 336 degrees of freedom. Residual deviance:
362.64 on 320 degrees of freedom. AIC: 1993.3.
*Significance values (*p < 0.05; **p < 0.01; ***p < 0.001).

Aggregate-level results company (negative). A potential explanation for the effect of

cloud-based IS could be the fact that cloud-based systems
The results of our model are summarized in Table 3. We make it easier to disseminate digital information between the
observe a significant relationship between the degree of outsourcer and the outsourcing service provider. Also, these
outsourcing and five motivation variables (Table 1). We systems allow the design of optimal arrangements to increase
find a significant relationship between motivation to reduce collaboration and transparency by providing a platform in
costs and an increase in the degree of outsourcing. Thus, the form of a cloud-based AIS endowing companies with
Hypothesis 1 is supported. As expected, a stronger motiva- tools to more efficiently (re)allocate the accounting pro-
tion to reduce and control costs leads companies to out- cesses between the client company and the accountant. The
source a larger number of tasks. In line with Hypothesis 2, finding that younger companies seem to outsource a larger
we observe a strong relationship between companies’ desire degree of processes probably stems from the need to focus
to focus on the core business and a higher degree of out- on the growth of the business in the nascent phases of the
sourcing. According to the results, the motivation to focus company resulting in a relatively strong need to outsource
on core competence appears to be the strongest driver of the non-core business processes.
degree of outsourcing. We observe a significant positive To aid in the interpretation of our results, we conducted
relationship between process improvement and a higher additional analysis of marginal effects and visualized its
degree of outsourcing. This result supports Hypothesis 4. results. We can get a better understanding of the model if
We also observe a significant negative relationship between we examine the marginal effects of the explanatory varia-
the degree of outsourcing and fear of losing control. Thus, bles. The marginal effects analysis calculates the unit
Hypothesis 8 is supported. The factor appears to play a change in the predicted number of outsourced processes in
negative role in determining the degree of outsourcing. response to a unit change in the explanatory variable by
We observe a statistically significant relationship holding all other variables at their means. This allows us to
between access to expertise and the degree of outsourcing. gauge the sensitivity of the firms with respect to their moti-
However, contrary to our expectations, access to expertise vations to outsource. The results indicate that the firms are
is associated with a lower degree of outsourcing. This more sensitive to motivation variables to focus on core
means that a stronger desire to access external expertise competence and to access expertise (see Table 4). A unit
leads companies to outsource fewer processes. Therefore, change in a firm’s motivation to focus on core competence
Hypothesis 3 is not supported. This is a surprising and leads to the outsourcing of one more process (1.05),
interesting result that warrants further discussion. whereas a one-unit increase in the motivation to access
Concerning the control variables, we note statistically sig- expertise leads to more than a one-unit decrease in out-
nificant effects of the use of cloud-based IS and the age of the sourced processes (–1.33). In practical terms, this means
Asatiani et al. 47

Table 4. Marginal effects. Salesi = α 0 + α1 IND _ COSi + α 2 IND _ FOCi

Variable Marginal effect SE + α 3 IND _ EXPi + α 4 IND _ IMPi
(4)
Cost reduction 0.385* 0.174 + α 5 IND _ SCLi + α 6 IND _ RPDi
Focus on core competence 1.052*** 0.246 + α 7 IND _ EASi + ε iSales
Access to expertise −1.327*** 0.287
Business/process improvements 0.645** 0.255
Scalability 0.282 0.195 Purchasei = θ 0 + θ1 IND _ COSi + θ 2 IND _ FOCi
Rapid delivery 0.195 0.227 + θ3 IND _ EXPi + θ 4 IND _ IMPi
Ease of use −0.049 0.268 (5)
+ θ5 IND _ SCLi + θ 6 IND _ RPDi
Fear of losing control −0.318* 0.178
Concerns for security −0.259 0.199 + θ 7 IND _ EASi + ε iPurchase

*Significance values (*p < 0.05; **p < 0.01; ***p < 0.001).
Payrolli = η0 + η1 IND _ COSi + η 2 IND _ FOCi
+ η3 IND _ EXPi + η 4 IND _ IMPi
that the motivations to focus on core business and access to (6)
expertise have a major role in determining the number of +η5 IND _ SCLi + η6 IND _ RPDi
outsourced processes. +η7 IND _ EASi + ε iPyaroll
Furthermore, to demonstrate the sensitivity of the varia-
bles, we have visualized the results of the regression model Reportingi = π 0 + π 1 IND _ COSi + π 2 IND _ FOCi
using simulation (see Figure 2). The simulation is carried
out for each of the motivation variables using parameters of + π 3 IND _ EXPi + π 4 IND _ IMPi
(7)
the model from Table 3 by fixing other variables at their + π 5 IND _ SCLi + π 6 IND _ RPDi
mean values. The visualization clearly demonstrates the size + π 7 IND _ EASi + ε iReporting
of the effect, confirming the conclusions drawn from the
marginal effects table (Table 4), discussed above.
Paymenti = ω0 + ω1 IND _ COSi + ω2 IND _ FOCi
+ ω3 IND _ EXPi + ω4 IND _ IMPi
Findings and analysis—process + ω5 IND _ SCLi + ω6 IND _ RPDi
(8)
group level
+ ω7 IND _ EASi + ε iPayment
In addition to probing the relationship between outsourcing
motivations and the degree of outsourcing, we looked into The error term Σ = (ε iSales , ε iPurchase , ε iPayroll , ε iReporting ,
the most important outsourcing motivations for each major
part of the accounting function. To group the accounting εiPayment ) is distributed normal, that is, Σ ~ N (0, σ 2 I ) .
processes into distinct parts, we used the categorization by We use Bayesian method to estimate the model. Along
Everaert et al. (2007): sales, purchases, payroll, reporting with data augmentation technique, we use Markov chain
and payments. In Table 10 in Appendix F, we have reported Monte Carlo (MCMC) method to run 50,000 iterations of
the responses to the main motivation items for each process which last 10,000 iterations are used for parameter
(P1–P22) and process group (sales, purchases, payroll, inference.
reporting and payments).
We use probit model to evaluate the relative importance of
each of the seven motivations on each of the outsourced pro-
Process group-level results
cesses that are grouped into five major categories. Let Salesi, The results of our model are summarized in Table 5. Our
Purchasei, Payrolli, Reportingi and Paymenti be the out- results show that with exception of reporting, all the stud-
sourced process for the company i for processes pertaining to ied outsourcing motivations were positively associated
sales, purchase, payroll, reporting and payment, respectively with outsourcing of accounting processes. However,
(please see Table 7 in Appendix B for categorization of 22 there are differences in the strengths of the effect. For
outsourced processes into the five major categories). Let example, while it appears that access to expertise is the
IND_COS, IND_FOC, IND_EXP, IND_IMP, IND_SCL, most important motivation factor to outsource payroll
IND_RPD and IND_EAS be the indicator variables pertaining and payments, focus on core competences seems to be the
to the importance of cost reduction, focus on core compe- main motivation for outsourcing transaction-related pro-
tence, access to expertise, business improvements, scalability, cesses. Concerning the processes within reporting, which
rapid delivery and ease of use, respectively, that takes a value were outsourced the most in our data set (see Table 10 in
of 1 if the company i stated that individual motivation to be Appendix F), the results suggest access to expertise and
the most important for their outsourcing decision, otherwise it focus on core competence are the main motivations for
is 0. Thus, we model the following outsourcing those processes.
48 Journal of Information Technology 34(1)

Figure 2. Simulated effect size based on the regression model.

Table 5. Importance of motivation on outsourced process.

Motivations Sales Purchases Payroll Reporting Payments

Intercept −4.5802*** −4.7674*** −3.6784*** 1.7096 −4.3246***
Cost reduction 6.0859*** 8.8442*** 7.0167*** 1.0875 8.0305***
Focus on core competence 6.8941*** 9.7257*** 9.0602*** 2.9299* 9.5469***
Access to expertise 6.2242*** 9.0857*** 9.4338*** 4.3107** 9.7882***
Process improvements 6.4545*** 9.2851*** 7.5277*** 1.7986 8.2367***
Scalability 5.8885*** 7.6005*** 4.5443** 0.6092 5.2962**
Rapid delivery 9.1035*** 9.08*** 6.3655*** 0.8768 7.5509***
Ease of use 6.0856*** 9.4482*** 8.5668*** 2.0923 9.1826***

*Significance values (*p ≤ 0.10; **p ≤ 0.05; ***p ≤ 0.01).

Discussion Sarkar (2012) outlined two main desired organizational out-

comes of outsourcing, which are broadly referred to as effi-
In this article, we set out to better understand the relation- ciency and adaptability. We observe a similar trend in our
ship between motivations to outsource and the degree of findings, where the dividing line among motivations is
outsourcing within a business function in the context of drawn between the desire for efficiency (related to utilization
SMEs. Our purpose was to critically analyse the outsourc- of supply-side economies of scale) and the desire to seek for
ing motivations cited in the existing literature and propose external expertise (drawing on the access to specialized
a conceptual basis for future studies. Next, we will present resources on the market). Our key contribution lies in uncov-
the implications of our research to theory and practice. ering the differences in the relationship between these groups
of motivations and the degree of outsourcing. We would like
to highlight two specific theoretical implications.
Implications for theory
Our first theoretical implication concerns the strength of
Even in the early days of outsourcing, researchers proposed various motivations and the degree of outsourcing. As
that while efficiency-focused motivations have a major role hypothesized, the three motivation variables associated with
in outsourcing decisions, strategy-focused motivations could efficiency (cost reduction, focus on core competence and
also play a key part in the decision to outsource particular business/process improvements) turned out to be positively
functions (McLellan, Marcolin, & Beamish, 1995; Gambal, related to the degree of outsourcing. These efficiency-related
Kotlarsky, & Asatiani, 2018). More recently, Weigelt and benefits of BPO emerge mainly as a result of the supply-side
Asatiani et al. 49

economies of scale and scope (Poppo & Zenger, 1998). outsourcing arrangement resulting from the decision. Our cen-
Therefore, for a company to be able to fully harness these tral claim is that the strength of various motivations defines the
benefits, a larger degree (scale) of the processes within the degree of outsourcing. However, these motivations may have
specific business function should be outsourced. Following a varying effect on the degree of outsourcing, where one set of
this reasoning, a sound theoretical argument drives the strong motivations facilitates a higher degree of outsourcing, another
positive relationship between efficiency-related motivations set has the opposite effect and a third set has no or little effect
and the degree of outsourcing. on the extent of outsourcing (see Figure 2).
Interestingly, our results suggest that not all motivation The implication of our approach is that one cannot study or
items work in the same way: Although a motivation varia- compare various outsourcing decisions without considering
ble may drive the overall decision to outsource, it may be the relationship dynamics between outsourcing motivations of
negatively associated with the degree of outsourcing. This the decision-making entity and the degree of outsourcing of a
was the specific case of the motivation variable access to given business function. While the majority of the existing
expertise in our study. While the motivation to access research (as demonstrated by Lacity et al., 2016; Lacity et al.,
expertise through outsourcing has been found in earlier lit- 2011) focuses on studying the role of motivation in making
erature to explain the overall decision to outsource (Lacity outsourcing decisions, very few studies analyse the relation-
et al., 2016), in our study companies that strongly sought to ship between the articulated motivations and the configuration
gain external expertise to the business function ended up (in our case, the degree of outsourcing) of the outsourcing con-
outsourcing fewer processes. We discuss this finding from tract. Our study demonstrates the potential knowledge gains
the perspective of company resources. By gaining access to from studying outsourcing arrangements from the perspective
expertise through outsourcing, companies seek to comple- of motivations and possibly other sourcing determinants, such
ment their resource based on the specific processes for as transaction attributes.
which they lack internal competence (i.e. cannot perform
themselves). Theoretically, compared to efficiency seeking
Implication for practice
through outsourcing, this type of resource complementing
is clearly targeted to a smaller number of processes. Thus, Uncovering the nature of the relationship between motiva-
based on the argumentation above, it seems reasonable to tions and the degree of outsourcing yields insightful impli-
find a negative relationship between access to expertise and cations for practitioners. Outsourcing service providers
the degree of outsourcing. should understand that their customer companies have dif-
Building on these observations, the second main theoreti- fering motivations to outsource their business processes and
cal implication from our results is a suggestion of systemic that the degree of outsourcing is one dimension that can be
influence on the interaction between outsourcing motiva- used to analyse these differences. According to our results,
tions and the degree of outsourcing. In other words, analys- companies seeking access to expertise outsource only a lim-
ing motivations on the scale of the whole function provides ited set of tasks, whereas those seeking efficiency gains out-
greater insights to the interaction between motivation factors source a larger set of tasks. Outsourcing service providers
and outsourcing, than merely studying the motivations could take this into account when designing their market
behind outsourcing each individual process. Our findings offerings. For example, when entering into negotiations to
suggest that when looking at outsourcing motivations of a renew the contract of an existing customer, the service pro-
given process or a group of processes within a larger busi- vider should analyse the past behaviour of the customer and
ness function, the relationship between motivations and deci- build arguments for continuing the contract accordingly.
sion to outsource remains more or less the same across the Outsourcing service providers can also develop service
spectrum of processes. Therefore, taking a modular view on packages tailored to different segments of the market and
outsourcing (i.e. analyse outsourcing motivation for each thus make their sales processes more efficient.
task or a small group of tasks within a function) could lead to Moreover, information system providers may use the
a conclusion that all outsourcing decisions are equally influ- findings of this study to build systems with features that
enced by a set of common outsourcing motivations. However, enable both efficiency and integration of expertise. For cus-
when looking at the bigger picture of outsourcing of a whole tomers that outsource a larger set of tasks, ensuring the effi-
business function (in our case, accounting), we observe the ciency of the outsourced processes is important, whereas
clear relationship between the strength of certain motivations easy integration of expertise is important for customer
and the degree of outsourcing. This calls for more systemic companies that outsource only a limited number of tasks.
analyses of outsourcing, echoing the observations made by Companies that are considering outsourcing should care-
Dibbern et al. (2012) on IS outsourcing. fully reflect upon their motivations and evaluate which of the
We argue that, contrary to earlier research (e.g. Dahlberg processes should be outsourced, in correspondence with
et al., 2006; Lee et al., 2004), outsourcing should not be viewed their objectives. Many cloud-based information systems
as a single decision with a fixed set of outcomes. Instead, we allow a more granular division of tasks between the out-
need to study this decision beyond the decision point and sourcing service provider and the customer company, which
focus on how the outsourcing decision determinants affect the enables more alternatives for organizing business processes.
50 Journal of Information Technology 34(1)

Limitations and opportunities for further enquiry on process group-level outsourcing revealed that
research while four of the process groups (sales, purchases, payroll,
payments) portrayed a balanced set of motivations across
We note three main limitations to our findings. First, we lim- the seven motivation items, outsourcing of processes in one
ited our empirical setting to the context of accounting. While process group (reporting) was motivated by access to
accounting offers a somewhat clean and controllable study expertise. Our empirical findings open avenues for further
setting, further research could explore the relationship studies probing the nature of the relationship between out-
between outsourcing motivations and the degree of outsourc- sourcing motivations and the degree of outsourcing.
ing in a broader context. Researchers could ask, for example:
Does the composition of the business function influence the Acknowledgments
degree of outsourcing and the associated motivation varia-
bles? Furthermore, empirically, we focused on SMEs, which We thank the Federation of Finnish Enterprises for the opportunity
to survey their members. We are grateful to the Senior Editor and the
have more limited access to resources and have fewer layers
two anonymous reviewers whose comments and support helped us
of hierarchy than larger firms. Clearly, we must assume the improve the manuscript. We also thank the HSE Support Foundation
outsourcing behaviour and motivations to be different in for partially funding this project. Our pre-study based on qualitative
larger firms compared to SMEs. Therefore, here, we restrict data was presented in a paper session at ECIS 2016 and we wish to
ourselves to making theoretical and practical knowledge thank the participants of that session for their feedback.
claims on SMEs’ outsourcing behaviour. Further research
could look into the differences between large and small firms Declaration of conflicting interests
in terms of their outsourcing motivations and degree of out- The author(s) declared no potential conflicts of interest with
sourcing. Second, in this study, we only focus on the degree respect to the research, authorship, and/or publication of this
of outsourcing omitting specific outsourcing configurations article.
and their interaction with outsourcing motivations. Further
studies could focus on identifying and characterizing com- Funding
mon outsourcing configurations within a business function
The author(s) disclosed receipt of the following financial support
and analyse the potential interdependencies between differ- for the research, authorship, and/or publication of this article: This
ent processes and motivations to outsource. Third, while the study was partially funded by the HSE Support Foundation.
motivation items used in this study have strong empirical
support from earlier research, due to the nature of our empiri- Note
cal study (survey), the items had to be simplified, condensed
1. We also compare the specified model with other count mod-
and measured on seven-point Likert-type scales. In reality,
els, such as the Poisson process. The negative binomial
the motivations driving an outsourcing decision might be distribution (NBD) model fits the data best. The Akaike
more complex calling for studies using data collection meth- information criterion (AIC) for the Poisson model is 2154.5,
ods that allow for more in-depth probing. Further research whereas the AIC for negative binomial model is 1991.3.
could, therefore, take a more qualitative stance and critically
review the results of our study. References
Asatiani, A., Apte, U., Penttinen, E., Rönkkö, M., & Saarinen, T.
Conclusion (2014). Outsourcing of disaggregated services in cloud-based
enterprise information systems. In 47th Hawaii International
We draw on two streams of outsourcing literature—out- Conference on System Sciences (HICSS), 2014 (pp. 1268–
sourcing motivations and degree of outsourcing—to probe 1277). IEEE.
the relationship between motivations and degree of out- Asatiani, A., & Penttinen, E. (2016). Profiling outsourcers of
sourcing. Relying on a rich data set with responses from professional services - an exploratory study on small and
337 companies, we find that cost reduction, a focus on core medium-sized enterprises. Proceedings of the 24th European
competence and business/process improvements are all Conference on Information Systems 2016–May: 1–16.
associated with a higher degree of outsourcing, but interest- The Association of Finnish Accounting Firms. (2014).
ingly, access to expertise is negatively associated with the Toimeksiantosopimus Tilitoimiston Palveluista KL2004.
degree of outsourcing. This finding suggests that compa- Retrieved from https://taloushallintoliitto.fi/sites/default/
nies that outsource to acquire external expertise are more files/122_KL2004_03062014.pdf
The Association of Finnish Accounting Firms. (2015). TAL-IT2015
selective. These companies outsource only a limited num-
Tilitoimistojen ohjelmistot. Taloushallintoliitto.fi. Retrieved
ber of processes within a specific business function. Our from https://taloushallintoliitto.fi/sites/default/files/doku-
main theoretical contribution lies in uncovering the mentit/page/fields/field_related_attachments/tal-it2015.pdf
dynamic nature of outsourcing motivations, meaning that The Association of Finnish Accounting Firms. (2017).
as companies outsource a larger degree of their business Tilitoimistoala Suomessa. Taloushallintoliitto.fi. Retrieved
processes, some motivation items become more accentu- from https://taloushallintoliitto.fi/tietoa-meista/tutkimuksia-
ated and others fade in importance. Furthermore, our ja-tietoa-alasta/tilitoimistoala-suomessa
Asatiani et al. 51

Bandyopadhyay, J., & Hall, L. A. (2009). Off-shoring of tax Gewald, H., & Dibbern, J. (2009). Risks and benefits of business pro-
preparation services by U.S. accounting firms: An empirical cess outsourcing: A study of transaction services in the German
study. Advances in Competitiveness Research, 17(1), 72–91. banking industry. Information & Management, 46, 249–257.
BDO Finland. (2015). Taloushallinnon Ulkoistaminen Jacobides, M. G. (2008). How capability differences, transaction
Keskisuurissa Yrityksissä. Author. Retrieved from https:// costs, and learning curves interact to shape vertical scope.
www.bdo.fi/fi-fi/nakemyksia/julkaisut/tutkimukset/taloush- Organization Science, 19, 306–326.
allinnon-ulkoistaminen-keskisuurissa-yrityk Kauffman, R. J., Techatassanasoontorn, A. A., & Wang, B. (2012).
Bhagwatwar, A., Hackney, R., & Desouza, K. C. (2011). Event history, spatial analysis and count data methods for
Considerations for information systems ‘backsourcing’: empirical research in information systems. Information
A framework for knowledge re-integration. Information Technology & Management, 13, 115–147.
Systems Management, 28, 165–173. Lacity, M., Khan, S., & Yan, A. (2016). Review of the empirical
Borman, M. (2006). Applying multiple perspectives to the BPO business services sourcing literature: An update and future
decision: A case study of call centres in Australia. Journal of directions. Journal of Information Technology, 31, 269–328.
Information Technology, 21, 99–115. Lacity, M., Khan, S., Yan, A., & Willcocks, L. (2010). A review
Buco, M. J., Chang, R. N., Luan, L. Z., Ward, C., Wolf, J. L., & Yu, of the IT outsourcing empirical literature and future research
P. S. (2004). Utility computing SLA management based upon directions. Journal of Information Technology, 25, 395–433.
business objectives. IBM Systems Journal, 43, 159–178. Lacity, M., Solomon, S., Yan, A., & Willcocks, L. (2011). Business
Carey, P., Subramaniam, N., & Ching, K. C. W. (2006). Internal process outsourcing studies: A critical review and research
audit outsourcing in Australia. Accounting & Finance, 46, directions. Journal of Information Technology, 26, 221–258.
11–30. Lacity, M., & Willcocks, L. (2016). Robotic process automation at
Chou, S. W., Techatassanasoontorn, A. A., & Hung, I. H. (2015). telefónica O2. MIS Quarterly Executive, 15(1), 21–35.
Understanding commitment in business process outsourcing Lam, W., & Chua, A. Y. K. (2009). An analysis of knowledge out-
relationships. Information & Management, 52, 30–43. sourcing at Eduware. Aslib Proceedings: New Information
Currie, W. L., Michell, V., & Abanishe, O. (2008). Knowledge Perspectives, 61, 424–435.
process outsourcing in financial services: The vendor per- Lee, J. N., & Kim, Y. (1999). Effect of partnership quality on IS
spective. European Management Journal, 26, 94–104. outsourcing success: Conceptual framework and empirical
Dahlberg, T., Nyrhinen, M., & Santonen, T. (2006). The suc- validation. Journal of Management Information Systems,
cess of selective and total outsourcing of firm-wide 15(4), 29–61.
IT-infrastructure: An empirical evaluation. In Proceedings Lee, J. N., Miranda, S. M., & Kim, Y. M. (2004). IT outsourcing
of the 14th European Conference on Information Systems strategies: Universalistic, contingency, and configurational
(pp. 1–13). explanations of success. Information Systems Research, 15,
Dibbern, J., Chin, W. W., & Heinzl, A. (2012). Systemic deter- 110–131.
minants of the information systems outsourcing decision: Luo, Y., Zheng, Q., & Jayaraman, V. (2010). Managing business
A comparative study of German and United States firms. process outsourcing. Organizational Dynamics, 39, 205–217.
Journal of the Association for Information Systems, 13, Mani, D., Barua, A., & Whinston, A. (2010). An empirical analy-
466–497. sis of the impact of information capabilities design on busi-
Dibbern, J., Goles, T., Hirschheim, R., & Jayatilaka, B. (2004). ness process outsourcing performance. MIS Quarterly, 34,
Information systems outsourcing: A survey and analysis 39–62.
of the literature. Database for Advances in Information Martinez-Noya, A., Garcia-Canal, E., & Guillen, M. F. (2012).
Systems, 35(4), 6–102. International R&D service outsourcing by technology-inten-
European Commission. (2003). Commission recommendation of sive firms: Whether and where? Journal of International
6 May 2003 concerning the definition of micro, small and Management, 18, 18–37.
medium-sized enterprises. Official Journal of the European McIvor, R., Humphreys, P., McKittrick, A., & Wall, T. (2009).
Union, 124, 36–41. Performance management and the outsourcing process: Lessons
Everaert, P., Sarens, G., & Rommel, J. (2007). Sourcing strategy from a financial services organisation. International Journal of
of Belgian SMEs: Empirical evidence for the accounting ser- Operations & Production Management, 29, 1025–1048.
vices. Production Planning & Control, 18, 716–725. McKeen, J. D., & Smith, H. A. (2011). Creating IT shared ser-
Everaert, P., Sarens, G., & Rommel, J. (2008). Using transaction vices. Communications of AIS, 29, 645–656.
cost economics to explain outsourcing of accounting. Small McKenna, D., & Walker, D. H. T. (2008). A study of out-sourcing ver-
Business Economics, 35, 93–112. sus in-sourcing tasks within a project value chain. International
The Federation of Finnish Enterprises. (2016). About the Journal of Managing Projects in Business, 1, 216–232.
Federation of Finnish enterprises. Yrittajat.fi. Retrieved McLellan, K., Marcolin, B. L., & Beamish, P. W. (1995). Financial
from https://www.yrittajat.fi/en and strategic motivations behind IS outsourcing. Journal of
Freytag, P. V., Clarke, A. H., & Evald, M. R. (2012). Reconsidering Information Technology, 10, 299–321.
outsourcing solutions. European Management Journal, 30, Nassimbeni, G., Sartor, M., & Dus, D. (2012). Security risks in
99–110. service offshoring and outsourcing. Industrial Management
Gambal, M. J., Kotlarsky, J., & Asatiani, A. (2018). Enabling & Data Systems, 112, 405–440.
Strategic Technological Innovations in IS Outsourcing Poppo, L., & Zenger, T. (1998). Testing alternative theories of the
Relationships: Towards an Innovation-molding Framework. firm: Transaction cost, knowledge-based, and measurement
In Proceedings of 39th the International Conference on explanations for make-or-buy decisions in information ser-
Information Systems 2018-December: 1–9. vices. Strategic Management Journal, 19, 853–877.
52 Journal of Information Technology 34(1)

Premuroso, R. F., Skantz, T. R., & Bhattacharya, S. (2012). Weigelt, C., & Sarkar, M. B. (2012). Performance implications
Disclosure of outsourcing in the annual report: Causes and of outsourcing for technological innovations: Managing the
market returns effects. International Journal of Accounting efficiency and adaptability trade-off. Strategic Management
Information Systems, 13, 382–402. Journal, 33, 189–216.
Redondo-Cano, A., & Canet-Giner, M. T. (2010). Outsourcing
agrochemical services: Economic or strategic logic? Service Author biographies
Business, 4, 237–252.
Sanders, N. R., Locke, A., Moore, C. B., & Autry, C. W. (2007). Aleksandre Asatiani is a Lecturer at the Department of Operations
A multidimensional framework for understanding outsourcing and Information Management at Aston Business School. He
arrangements. Journal of Supply Chain Management, 43(4), received his PhD in information systems from Aalto University
3–15. School of Business. His research focuses on outsourcing, virtual
Saxena, K. B. C., & Bharadwaj, S. S. (2009). Managing business organizations and technology-based innovation.
processes through outsourcing: A strategic partnering perspec-
tive. Business Process Management Journal, 15, 687–715. Esko Penttinen is a Professor of Practice at Aalto University
Singh, T., & Tornbohm, C. (2016). Predicts 2017: The future state of School of Business. He leads the Real-Time Economy Competence
business process services and outsourcing. Gartner. Retrieved Center and he is also Chairman of XBRL Finland. His research
from https://www.gartner.com/doc/3512952/predicts--future- interests include adoption of inter-organizational information
state-business
systems, focusing on application areas such as electronic financial
Snowden, J., & Fersht, P. (2016). The HFS market index—IT ser-
systems, and government reporting.
vices and BPO market size and forecast 2016–2020. Boston,
MA: HFS Research.
Statistics Finland. (2017). Yritykset. Tilastokeskus.fi. Retrieved Ashish Kumar is an Assistant Professor in the Department of
from https://www.stat.fi/tup/suoluk/suoluk_yritykset.html Marketing, Aalto University School of Business. He received his
Weigelt, C. (2009). The impact of outsourcing new technolo- PhD in marketing from the School of Management and an MA in
gies on integrative capabilities and performance. Strategic economics from the College of Arts and Sciences, respectively,
Management Journal, 30, 595–616. from the State University of New York at Buffalo.

Appendix A
Summary of motivations to outsource
Table 6. Motivation variables summary.

Motivations Current Pre-study Lacity, Solomon, Lacity, Khan, Lacity, Khan,Yan, References
study Yan, and Willcocks and Yan and Willcocks
(2011) (2016) (2010)
Cost reduction      Borman (2006); Martinez-Noya,
Garcia-Canal, and Guillen (2012)
Focus on core      Carey, Subramaniam, and Ching
competence (2006); Gewald and Dibbern
(2009); Premuroso, Skantz, and
Bhattacharya (2012)
Access to skills/      Currie, Michell, and Abanishe
expertise (2008); Gewald and Dibbern
(2009); Lam and Chua (2009)
Process      Buco et al. (2004); Gewald and
improvements Dibbern (2009); Weigelt and
Sarkar (2012)
Scalability      Currie et al. (2008); Redondo-
Cano and Canet-Giner (2010)
Rapid delivery      Bandyopadhyay and Hall (2009);
Freytag, Clarke, and Evald (2012)
Ease of use      McKenna and Walker (2008)
Fear of losing      Bhagwatwar, Hackney, and
control Desouza (2011); McKeen and
Smith (2011); Sanders, Locke,
Moore, and Autry (2007)
Concern for      McIvor, Humphreys, McKittrick,
security and Wall (2009); Nassimbeni,
Sartor, and Dus (2012)
Asatiani et al. 53

Appendix B
Accounting in Finland
Table 7. Typical processes within accounting function in Finnish SMEs.
Sales P1 Client register maintenance
P2 Product register maintenance
P3 Sending sales invoices
P4 Handling sales invoices
P5 Sending note of complaint
P6 Sales ledger maintenance
Purchases P7 Supplier register maintenance
P8 Receiving purchase invoices
P9 Handling purchase invoices
P10 Handling purchase, travel and other costs
P11 Purchases ledger maintenance
Payroll P12 Personnel register maintenance
P13 Basic payroll data maintenance
P14 Payroll calculations
Reporting P15 Preparation of balance sheet and income statement
P16 Preparation and sending of VAT
P17 Preparation and sending of annual salary reports
P18 Preparation and sending of annual pension insurance reports
Payments P19 Periodic VAT payments
P20 Salary payments
P21 Payments for purchases, travel and other expenses
P22 Monthly payroll tax payments

SMEs: small and medium-sized enterprises; VAT: value added tax.

Table 8. Typical terms of agreement between accounting service provider and a client company in Finland.

Item Termsa
Subject of agreement Performance of some variations of accounting processes outlined in Appendix B
Payment frequency Annual, quarterly or monthly
Contract length Until further notice. Two months period of notice (unless specified otherwise)
Pricing Hour-based
Task-based
Transaction-based
Service-based
Combination of above
aBased on a standard service agreement published by The Association of Finnish Accounting Firms (2014).

Appendix C □ Production
□ Other. What? ______________
Survey questions 2. What industry does your company operate in?
1. Which of the following describes your position in [Drop-down list of industries defined by
the company the best? Statistics Finland]
□ Top management 3. How many employees does your company have?
□ Finance and accounting ○ 1–5
□ HR ○ 6–10
□ Information management ○ 11–15
□ Administration ○ 16–20
□ Sales ○ 21–25
□ Marketing ○ 26–30
□ R&D ○ 31–35
54 Journal of Information Technology 34(1)

○ 36–40 4. Handling sales invoices

○ 41–45 5. Sending note of complaint
○ 46–50 6. Sales ledger maintenance
○ 51–75 7. Supplier register maintenance
○ 76–100 8. Receiving purchase invoices
○ 101–150 9. Handling purchase invoices
○ 151–250 10. Handling purchase, travel and other costs
○ 250+ 11. Purchases ledger maintenance
○ I am a sole trader 12. Personnel register maintenance
○ I do not know 13. Basic payroll data maintenance
4. What was your company’s last reported turnover (as 14. Payroll calculations
reported to the Finnish Patent and Registration 15. Preparation of balance sheet and income
Office and/or the Finnish Tax Authority)? statement
○ €0–€100,000 16. Preparation and sending of VAT
○ €100,001–€200,000 17. Preparation and sending of annual salary
○ €200,001–€300,000 reports
○ €300,001–€400,000 18. Preparation and sending of annual pension
○ €400,001–€500,000 insurance reports
○ €500,001–€1,000,000 19. Periodic VAT payments
○ €1,000,001–€2,000,000 20. Salary payments
○ €2,000,001–€5,000,000 21. Payments for purchases, travel and other
○ €5,000,001–€8,000,000 expenses
○ €8 000 001–€10,000,000 22. Monthly payroll tax payments
○ €10,000,001–€20,000,000 9. Please evaluate how the following factors affected
○ €20,000,001–€50,000,000 your decision to outsource on a scale of 1–7
○ €50,000,000+ (1 = had no influence, 7 = had a big influence)
○ I do not know – Cost reduction
5. Where are your customers located? – Ability to focus on core business
□ Finland – Expertise of an external accountant
□ The EU (other than Finland) – Opportunity to improve accounting process
□ The rest of Europe – Ability to quickly scale up/down accounting
□ Asia services based on the needs of our company
□ Africa – Faster delivery times by an external accountant
□ N. America than internal delivery
□ S. America – Ease of use
□ Australia – Fear of losing control over outsourced account-
6. Where does your company operate (have Branches)? ing process
□ Finland – Concern over security of outsourced account-
□ The EU (other than Finland) ing process
□ The rest of Europe 10. Please indicate the most important factor for out-
□ Asia sourcing each of the outsourced processes
□ Africa [The list of the outsourced processes, based on
□ N. America Question 8]
□ S. America – Cost reduction
□ Australia – Ability to focus on core business
7. Do you use cloud-based accounting information – Expertise of an external accountant
systems? – Opportunity to improve accounting process
○ Yes – Ability to quickly scale up/down accounting
○ No services based on the needs of our company
○ I do not know – Faster delivery times by an external accountant
8. If you are outsourcing your accounting partially, than internal delivery
please mark the functions that are outsourced/kept – Ease of use
in-house accordingly (in-house/outsourced): – Fear of losing control over outsourced account-
1. Client register maintenance ing process
2. Product register maintenance – Concern over security of outsourced account-
3. Sending sales invoices ing process
 Asatiani et al.

Appendix D
Correlations table
Table 9. Correlations table.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Cost reduction 1 –
Focus on core 2 0.2927 –
competence
Access to expertise 3 0.2053 0.4208 –
Process improvements 4 0.3961 0.4292 0.4508 –
Scalability 5 0.3743 0.4131 0.3313 0.4538 –
Rapid delivery 6 0.3895 0.4825 0.3761 0.6336 0.5144 –
Ease of use 7 0.3075 0.5526 0.4980 0.5292 0.4153 0.5500 –
Concerns for security 8 0.2109 0.0647 0.1395 0.0981 0.1607 0.1015 0.0560 –
Fear of losing control 9 0.1004 0.1339 0.1366 0.1017 0.1219 0.1073 0.0497 0.5994 –
Use of cloud-based IS 10 −0.0128 0.0444 −0.0266 0.0930 0.0404 0.1676 0.0232 −0.0082 −0.0333 –
Number of employees 11 0.0667 0.0250 0.0518 0.0589 0.0700 0.0243 0.0020 0.0622 0.0183 0.0106 –
Age of the company 12 −0.0820 −0.1158 0.0128 −0.0942 −0.0949 −0.1112 −0.0671 0.0283 −0.1024 −0.1053 0.0941 –
Multinational 13 0.0593 −0.0101 −0.0526 −0.0576 0.0549 −0.0673 −0.0157 −0.0600 −0.1361 0.0817 0.0226 0.0225 –
operations
Turnover 14 0.0048 0.0137 0.0563 0.1061 0.0849 0.0461 0.0270 −0.0132 −0.0208 0.0953 0.1172 0.0599 −0.0041 –
Secondary sector 15 −0.0225 −0.1096 −0.0428 −0.0493 −0.0891 −0.0784 −0.1325 0.0111 −0.0345 −0.0053 0.1000 0.1167 0.1390 −0.0329 –
Tertiary sector 16 0.0205 0.1006 0.0499 0.0447 0.0687 0.0746 0.1206 −0.0059 0.0541 −0.0186 −0.1000 −0.1095 −0.1316 0.0305 −0.9703
55
56 Journal of Information Technology 34(1)

Appendix E vivi
g (δ i ) = δ ivi−1 e( −δi vi ) for vi > 0
Negative binomial distribution model Γ ( vi )
(13)
∞
We derive the expression for negative binomial distribu-
and Γ(v) = ∫ t v −1e − t dt
tion (NBD) model starting from a basic Poisson model. 0
The Poisson model provides estimates of the probability of
the observed count, y=CntOutProc, as follows
Integrating Equation (12) with Equation (13) gives the
e − µi
µ yi probability density of the negative binomial as follows
Pr ( yi xi ) = i
(9)
yi !
Γ ( yi + vi )  vi   µ 
vi yi

which is conditional on an expected mean ∝i , expressed in Pr ( yi xi ) =     (14)

yi !Γ ( vi )  vi + µi   vi + µ 
exponential form to avoid negative values

E ( yi xi ) = µi = e( xi β ) Appendix F (10)
Process-level and process group-level
The negative binomial model replaces ∝i with random
motivations
variable λi
In Table 10, we report the responses to the individual pro-
λi = e ( xi β + ε i )
=e ( xi β ) εi
e = µi δ i (11) cess-level motivations asked in Question Number 10 in
Appendix C. In the first block, we indicate how many
Thus, the probability density function of the negative respondents had outsourced the process and the corre-
binomial function becomes sponding percentage of the total number of respondents
(n = 337). In the second block (and third block), we present
e − µi δ i ( µ i δ i )
yi
how many times each motivation item was cited as the
Pr ( yi xi ) = (12)
yi ! main motivation to that specific process (and the corre-
sponding percentage). In the fourth and fifth blocks, we
In NBD model, the probability of the observed count present the process group-level findings in terms of how
becomes conditional on the error distribution. We get the many times each motivation item was cited as the main
unconditional distribution by specifying the gamma distri- motivation to outsource any of the processes associated to
bution for δ i with parameter vi and probability density the process group (sales, purchases, payroll, reporting and
function g (δ i ) as follows payments).
Asatiani et al. 57

Table 10. Process-level and process group-level motivations.

Process level Sales Purchases

P1 P2 P3 P4 P5 P6 TOTAL P7 P8 P9 P10 P11 TOTAL

# outsourced 10 14 65 32 84 125 330 51 127 76 77 152 483
% outsourced 3% 4% 19% 9% 25% 37% 16% 15% 38% 23% 23% 45% 29%
Process-level motivations (#) P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11
Cost reduction 2 2 5 1 4 6 3 15 6 10 10
Focus on core competence 2 3 18 7 25 31 13 25 25 21 35
Access to expertise 1 2 2 4 9 23 4 7 5 11 29
Process improvements 1 2 12 5 15 15 8 22 10 9 22
Scalability 1 2 7 4 2 8 2 4 2 3 8
Rapid delivery 0 1 12 3 14 20 3 26 12 8 20
Ease of use 3 2 9 8 15 22 18 28 16 15 28
Process-level motivations (%) P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11
Cost reduction 20% 14% 8% 3% 5% 5% 6% 12% 8% 13% 7%
Focus on core competence 20% 21% 28% 22% 30% 25% 25% 20% 33% 27% 23%
Access to expertise 10% 14% 3% 13% 11% 18% 8% 6% 7% 14% 19%
Process improvements 10% 14% 18% 16% 18% 12% 16% 17% 13% 12% 14%
Scalability 10% 14% 11% 13% 2% 6% 4% 3% 3% 4% 5%
Rapid delivery 0% 7% 18% 9% 17% 16% 6% 20% 16% 10% 13%
Ease of use 30% 14% 14% 25% 18% 18% 35% 22% 21% 19% 18%
Process group-level motivations (#) Sales Purchases
Cost reduction 20 44
Focus on core competence 86 119
Access to expertise 41 56
Process improvements 50 71
Scalability 24 19
Rapid delivery 50 69
Ease of use 59 105
Process group-level motivations (%) Sales Purchases
Cost reduction 6% 9%
Focus on core competence 26% 25%
Access to expertise 12% 12%
Process improvements 15% 15%
Scalability 7% 4%
Rapid delivery 15% 14%
Ease of use 18% 22%

(Continued)
58 Journal of Information Technology 34(1)

Table 10. (Continued)

Process level Payroll Reporting Payments

P12 P13 P14 TOTAL P15 P16 P17 P18 TOTAL P19 P20 P21 P22 TOTAL
# outsourced 110 157 231 498 332 310 254 244 1140 163 188 113 198 662
% outsourced 33% 47% 69% 49% 99% 92% 75% 72% 85% 48% 56% 34% 59% 49%
Process-level motivations (#) P12 P13 P14 P15 P16 P17 P18 P19 P20 P21 P22
Cost reduction 11 10 11 6 7 10 10 7 10 8 6
Focus on core competence 33 40 58 45 59 57 60 36 43 25 39
Access to expertise 23 59 96 242 170 120 113 62 51 21 89
Process improvements 12 9 16 13 13 20 16 10 15 14 13
Scalability 1 2 5 1 3 3 2 3 2 1 3
Rapid delivery 4 6 8 6 13 11 6 11 16 15 13
Ease of use 26 31 37 19 45 33 37 34 51 29 35
Process-level motivations (%) P12 P13 P14 P15 P16 P17 P18 P19 P20 P21 P22
Cost reduction 10% 6% 5% 2% 2% 4% 4% 4% 5% 7% 3%
Focus on core competence 30% 25% 25% 14% 19% 22% 25% 22% 23% 22% 20%
Access to expertise 21% 38% 42% 73% 55% 47% 46% 38% 27% 19% 45%
Process improvements 11% 6% 7% 4% 4% 8% 7% 6% 8% 12% 7%
Scalability 1% 1% 2% 0% 1% 1% 1% 2% 1% 1% 2%
Rapid delivery 4% 4% 3% 2% 4% 4% 2% 7% 9% 13% 7%
Ease of use 24% 20% 16% 6% 15% 13% 15% 21% 27% 26% 18%
Process group-level motivations (#) Payroll Reporting Payments
Cost reduction 32 33 31
Focus on core competence 131 221 143
Access to expertise 178 645 223
Process improvements 37 62 52
Scalability 8 9 9
Rapid delivery 18 36 55
Ease of use 94 134 149
Process group-level motivations (%) Payroll Reporting Payments
Cost reduction 6% 3% 5%
Focus on core competence 26% 19% 22%
Access to expertise 36% 57% 34%
Process improvements 7% 5% 8%
Scalability 2% 1% 1%
Rapid delivery 4% 3% 8%
Ease of use 19% 12% 23%