|
Offshoring the Financial Services Industry:
Implications for the Evolution of Indian IT Clusters
Forthcoming in Environment and Planning A
Accepted 20 November 2004
Michael H. Grote and Florian A. Täube
Johann Wolfgang Goethe-University, Frankfurt
Version November 2004
J. W. Goethe-University
Faculty of Economics and Business Administration
Schumannstraße 60
60325 Frankfurt am Main
Germany
Michael H. Grote:
Phone: +49 (0) 69 798-23566
Fax: +49 (0) 69 798-28121
grote@stud.uni-frankfurt.de
Florian A. Täube:
Phone: +49 (0) 69 798-23411
Fax: +49 (0) 69 798-28121
taeube@wiwi.uni-frankfurt.de
- 2 -
Offshoring the Financial Services Industry:
Implications for the Evolution of Indian IT Clusters
Abstract
This paper explores the opportunities for existing Indian IT clusters to upgrade and
undertake financial research activities. Wholesale financial activity and the accompanying
financial research in banks are still highly concentrated in Western financial centers.
Increasing competition in the financial services industry as well as regulatory pressure
place the options of outsourcing and offshoring activities, especially research, to low-cost
locations high on the agenda of financial institutions. For the first time complex tasks at
the core of financial activity are offshored which makes it an interesting case for a lot of
other industries and their spatial economic organization in an ever globalizing world. Will
there be a World Financial Research Centre in Mumbai? Using qualitative interview data
as well as a quantitative analysis, this paper argues that research activities are locally
embedded in Western financial centers to an extent that such a development is not likely.
Two different research activities, viz. country analysis and institutional equity analysis
are examined. This analysis shows, however, that there is a certain potential for some
research activities to be relocated to India. So far investments take place in very few
existing IT clusters which have already gained reputation in the financial community.
:
- 3 -
“National economic prosperity depends not only on the pattern [of success and
failure in its economy] at any point in time, however, but even more so on the
capacity of a nation’s industry to upgrade itself over time.”
(Porter, 1990: 277-8, italics in original)
Introduction
This paper explores the possibilities of upgrading existing Indian IT clusters in order to
take over financial research activities. India is so far the prime location for offshoring and
outsourcing IT services and financial services firms are among the largest customers with
activities like billing and data entry. Although such tasks normally do not yield much of
the value-added for the developing country, such a division of labor can be the
(necessary) initial step in a process of upgrading the capabilities of firms in these
countries through international cooperation (Lateef, 1997). For instance, the rise of the
Silicon Valley model with a network of highly specialized companies over the vertically
integrated firm of previous decades gave these entrepreneurs the opportunity to venture
into niche markets and outsource a large part of the global value chain to their home
country (Saxenian et al. 2002). Previous studies on upgrading in global value chains have
usually looked at upgrading trajectories of specific, and rather traditional, industries or
sectors, e.g. Asian apparel (Gereffi, 1999). Rabellotti and Pietrobelli (2004) conducted a
comparative study on 12 different Latin American clusters. Whereas process and product
upgrading are found to be common, intersectoral upgrading is found only in a Chilean
salmon cluster that ventured into biotechnology and genetics. Regarding Indian IT
clusters, the literature so far also focuses on product and process upgrading (Krishnan
and Prabhu, 2002; D’Costa, 2002). Using an enhanced value chain approach, this paper
analyzes the possibilities of shifting research activities within the financial sector to
existing Indian software clusters, i.e. the possibility of intersectoral upgrading.
- 4 -
At present, wholesale financial activity and the accompanying financial research in banks
are highly concentrated in Western financial centers (Clark, 2002). Increasing
competition in the financial services industry as well as regulatory pressure place the
options of outsourcing and offshoring activities, especially research, to low-cost locations
high on the agenda of financial institutions (Deloitte Consulting, 2003). For the first time
complex tasks at the core of financial activity are offshored, which makes it an
interesting case for a lot of other industries and their spatial economic organization in an
ever globalizing world. Will there be a World Financial Research Centre in Mumbai
(formerly called Bombay), the financial center of India? This paper argues that many
research activities are locally embedded in Western financial centers to an extent that
such a development is not likely. Two different research activities, i.e. country analysis
and institutional equity analysis are examined. The analysis shows, however, that there
is a certain potential for some research activities to be relocated to India.
The paper is structured as follows: In the two theoretical sections the value chain
framework for the analysis of possible business process reorganizations in the financial
sector is developed and the potential upgrading channels of Indian software clusters are
examined. The subsequent chapter describes the methodology mix that is used in the
empirical analysis. The empirical section then presents micro-level results from
interviews with financial analysts about their organizational and local ‘embeddedness’,
i.e. an assessment of the relocation possibilities of financial research to India. A
quantitative analysis of host region factors determining the location of Brownfield and
Greenfield investments by financial institutions in India figures as a complementary
meso-level approach. The last section concludes.
- 5 -
Reorganization in the Financial Industry
Financial activities above the retail level are concentrated to a large extent in financial
centers, which form an intricate global production network of interconnected locations
between high-end members of the global city hierarchy (Friedmann, 1995; Poon, 2003;
Sassen, 1991). Strong agglomerative forces “shape the evolving geographies of domestic
and global finance” (Martin, 1999, 15). In the case of financial production, centripetal
forces are linked to socio-institutional and cultural factors, labor market externalities,
access to intermediate services, and above all informational spillovers (Porteous, 1999;
Thrift, 1994; Kindleberger, 1974) – and thus to proximity to other actors. Centrifugal
forces, on the other hand, so far played only a minor role. Intensified competition and reregulation
lead to increasing pressure on the costs of financial services firms (Walter,
2004). Outsourcing and offshoring are regarded as important means of reducing costs in
general as well as in financial services. Cost reduction can be achieved via reduction in
wages (offshoring), via economies of scale within the service provider delivering the
same services to more customers (outsourcing), or both. Information and communication
technologies (ICT) in the financial industry now might enable the splitting up of existing
production processes: hence, we propose a possibility of integrating new entrants, e.g.
from India, into the global financial production network.
Outsourcing is understood here as a process in which certain service providers external
to the firm take over business processes formerly conducted within the firm. A relocation
of these processes is thus not required for outsourcing. Contrary to that, offshoring
means the relocation of activities from one site to another, which is often in low-cost
regions, within the same firm or at least the same corporate group. Offshoring can occur
in combination with outsourcing, but not necessarily so. The reasons for not outsourcing
and offshoring everything to the cheapest location or specialized provider are transaction
costs and competencies of the firm that are not easily transferable (see Mahnke, 2001,
for a comprehensive overview). It is argued here that the amount and sort of information
- 6 -
and communication exchanged in business processes is crucial for the possibilities of
outsourcing and offshoring research activities.1
An analysis of the spatial consequences of these decisions has to take into account firm
and even department levels. The concept of value chains provides a simultaneous and
disaggregated analysis of the process and actors, allowing for the appraisal of
competitive advantages of firms (looking for the processes that could be outsourced) and
comparative advantages of countries or regions (looking for the places to which they
could be offshored – see Kogut, 1985). The concept of value chains is mostly used for the
analysis of production systems while services are regarded merely as facilitating links
between production stages (Dicken, 1998; Rabach and Kim, 1994). We build upon an
approach suggested by Grote et al. (2002) that allows for analyzing those junctures
(‘breaking points’) where production stages can be separated, either spatially
(offshoring) or organizationally (outsourcing). While traditional value chain analyses
concentrate on vertical relationships (Leslie and Reimer, 1999), this enhanced approach
integrates also horizontal links to co-operating firms, customers, or other sources of
information and knowledge. Grote et al. (2002) utilize a combination of vertical analysis
with horizontal connections in order to locate the different stages of financial production;
these horizontal linkages occur predominantly in geographical agglomerations or clusters.
Thus, each stage is regarded as a node which “is in itself a network connected to other
nodes concerned with related activities” (Appelbaum et al., 1994, page 188).
Here we are interested in the organizational as well as spatial ‘embeddedness’ of financial
research activities.2 In this context, embeddedness is understood neither in a pure
geographical nor in an exclusively organizational sense. Basically we look at the
“structure of the overall network of relations” (Granovetter, 1990, page 98) of analysts
and especially the means of information and communication they need to perform their
1 The financial sector is heavily regulated and the decision to outsource and offshore activities is often subject
to approval of regulatory bodies. However, the regulatory point of view does not consider research activities to
be critical for the survival of financial institutions and will not be considered here.
2 A large body of literature on (regional and other kinds on) embeddedness exists that will not be covered here.
- 7 -
tasks. The ability to slice up the value chain and to outsource certain stages or phases of
business processes and to offshore parts of the value chain within their firms to low-wage
locations depends crucially on how the processes are ‘embedded’ in relation to a) the
other departments of the bank and b) to the corresponding actors outside the bank. The
need for the interchange of information and knowledge between research departments
and actors outside as well as inside their banks in other departments, i.e. along the value
chain, determines the possibility of relocating activities. We argue that different kinds of
proximity are necessary for the communication of complex information or knowledge.
Information (and codified knowledge) is easily transferable between actors via ICT, while
tacit knowledge is not. Tacit knowledge is strongly linked to subject and context, and
thus difficult to pass on (Nonaka and Takeuchi, 1995). There is, however, no clear-cut
distinction between tacit knowledge and information: the “degree of tacitness” (Nelson
and Winter, 1982, page 80) is determined by time, complexity, and depth of explanation
needed. Whether a specific type of knowledge could in principle be codified and
transferred via ICT is not considered here; when the costs of codifying become too high,
the use of ICT becomes de facto impossible. ICT allows for more and more knowledge to
become codified and thus transferable in space. However, to interpret this codified
knowledge, most often a specific code is needed which has to be learned by actors
beforehand – with that learning process itself being based on tacit knowledge (Nelson
and Winter, 1982; Cowan and Foray, 1997).
Knowledge with a high degree of tacitness is still passed on only between co-present
actors; face-to-face contacts are described as an ‘efficient communication technology’,
where “verbal, physical, contextual, intentional and non-intentional” levels of
communication are addressed at the same time (Storper and Venables, 2004, page 355).
The higher the need for spontaneous meetings and/or regular informal contacts, the
more important spatial proximity becomes. In addition, face-to-face contacts display
several other advantages such as the possibility of judging other persons and being
- 8 -
judged by others facilitating the detection of lies and thus the building up of trust, and
better motivation in co-presence (Storper and Venables, 2004). Trust is an essential
aspect in the financial business and especially in communication via ICT (Banerjee and
Duflo, 2000); the development of trust requires at least initial spatial proximity (Giddens,
1990).
Other proximities play a role in the spatial organization of the financial research sector as
well (Grote et al., 2002; Stein, 2002; cf. Zeller, 2004)3: since “nation-states create those
institutions which actively define and maintain distinct industrial practices” (Gertler,
1993, page 676) it is a kind of ‘cultural’ or ‘national’ proximity that is needed to
understand certain business practices, local regulations (or local interpretations of
international regulations) and not the least language of the actors. The less transparent
an object or transaction, the higher is the relevance of cultural proximity (see Bathelt,
2000). Face-to-face contacts and cultural proximity can be partially compensated
through vertical integration or “organizational proximity” (Lundvall, 1988). Organizational
proximity exists between actors working in the same company regardless of their
geographical location and thus gains a new relevance with the implementation of ICT
(Bathelt, 2000). It refers to corporate identity, corporate philosophy, organizational
rules, and codes (Blanc and Sierra, 1999). In a similar manner, proximity between actors
in the same type of job can bridge spatial and cultural distance to a certain extent. Actors
in close ‘professional proximity’ possess an understanding of each others’ methods,
practices and aims, share similar interests, and professional language. Both
organizational and professional proximity facilitate the building up of trust and furnish a
common background for the actors and hence a context for interaction, thereby
simplifying knowledge exchange. They are based on shared conventions, thereby
providing a common “framework of action [...] with other actors engaged in that activity”
(Storper 1997, page 45). Where organizational and professional proximity is strong, trust
that enables the disembedding mechanism to operate is supplied; hence ICT can be used
3 The following proximities are similar to the different concepts of embeddedness proposed by Hess (2004).
- 9 -
to bridge great spatial distances. In contrast, the value provided by spatial and cultural
proximity acts predominantly to keep activities within a certain territory. However,
parallel to organizational proximity ethnic networks also create a common background
and thus help in extending communication from a local, regional or national to a
transnational dimension, as exemplified in the case of the Indian software industry and
its connections with Silicon Valley (Saxenian et al., 2002, Taeube, 2004b).
In general, outsourcing becomes an option when organizational proximity is not
necessary. Moreover, the content of the processes to be outsourced ought not to be of
strategic or otherwise critical value to the outsourcing company (Cronin et al., 2004). In
this case the increased dependence on specialized suppliers offering specific assets might
lead to the classic hold-up problem. Offshoring of complex tasks is possible for parts of
the value chain that do not require cultural proximity and face-to-face contacts and
where professional and organizational proximity ensures sufficient common background
for communicating.
When it is possible to offshore certain tasks, the offshoring location has to be
determined. Investment banks’ research is a complex task that requires capabilities and
training which is not available ubiquitously, contrary to, e.g. call center activities. So, just
moving to cheaper locations in the countryside is not an alternative for offshoring: banks
have to shift their research activities to places where there is already a pool of
adequately trained staff – i.e. to clusters. From a strategic management perspective,
Multinational Corporations (MNCs) would locate in clusters that have the highest
probabilities of delivering value. Apparently, the local resources need not be idiosyncratic
given the nature of ever more interconnected or networked organizations and business
relations (Birkinshaw and Hagström, 2002). Hence, one would expect higher value-added
activities to be localized in those existing clusters exhibiting features such as labor
markets with experience specific to the requirements of MNCs (Fromhold-Eisebith, 2002;
see also Cantwell and Mudambi, 2004). The next section examines the requirements of
- 10 -
Indian IT clusters to become hosts for research activities of investment banks and thus
lock into the global financial production network.
Upgrading Channels of Indian IT Clusters
Clusters – and cities with clusters - are long established in the literature as important
places for learning, innovation and economic development (Glaeser, 1999; Porter, 1998;
Maskell and Malmberg, 1999). Besides the traditional Marshallian externalities, external
economies like knowledge spillovers (Almeida and Kogut, 1999) derive from collective
efficiency, social capital or some other form of social cohesiveness (Uzzi, 1997; Nahapiet
and Ghoshal, 1998). There is evidence from very heterogeneous developing country
clusters having faced and mastered economic challenges that the higher the level of local
cooperation the higher the success and future performance. Social networks and ties at
the local level allow for collective social action and knowledge spillovers that lead to the
collective efficiency emphasized by Schmitz (1995) as an important ingredient of cluster
upgrading.
A significant role is played by geographical and social proximities: firms and people
located in the same region possess some form of a shared culture or collective identity.
Geographical proximity enables the creation of common cultural contexts which, in turn,
facilitate interactive learning processes crucial for innovation, because regional cultures
tend to become institutionalized as rules of conduct which govern the relations and
interactions of economic agents within the geographical area (Dosi, 1988; Storper, 1995;
Lorenzen & Mahnke, 2004, forthcoming). Other authors stress the importance of learning
from competitors in clusters (Malmberg and Maskell, 2002) and the predominance of
vertical links (Humphrey and Schmitz, 2000).
According to Humphrey and Schmitz (2000) there are four types of upgrading: 1)
process upgrading, which is a more efficient way of production, 2) product upgrading,
which means selling similar products in higher market segments, 3) functional upgrading,
- 11 -
that is assuming roles with a higher value-added within the production process, and 4)
intersectoral upgrading, using competences acquired in one sector for production in other
sectors. Typical upgrading channels are all kinds of (knowledge) spillovers, for instance,
the mobility of human capital (Franco and Filson, 2000), spin-offs (Klepper, 2001), or
knowledge transfer from MNCs (Fromhold-Eisebith, 2002). Schmitz (1995) introduced the
concept of collective efficiency, defined as the competitive advantage derived from local
external economies and joint action in order to assess the impact on the competitiveness
of firms located in clusters. This implies that the capabilities that are required to upgrade
are supposedly built up through local processes. It is the concept of absorptive capacity
that analyzes the “ability to exploit external knowledge” (Cohen and Levinthal, 1990,
128). Originally developed for analysis at firm level, this concept has been extended to
the analysis of countries – developing countries in particular – where education and
infrastructure are among the most critical factors (Dahlman and Nelson, 1995) and
clusters (Giuliani, 2004, forthcoming).
The success of the Indian software industry is well-researched, with factors like first-class
higher education and research institutions, both public and private, low labor costs and
stimulating policies commonly accepted as systemic components. Since the late 1990s,
an increasing number of studies on its upgrading potential has been published (e.g.
Krishnan and Prabhu, 2002; D’Costa, 2002), which usually focus on functional or product
upgrading. Nowadays, MNCs increasingly locate not only low-level programming but also
research and development (R&D) centers or laboratories in India; many already have
more than one research lab (Fromhold-Eisebith, 2002). Generally, the quality of
software-exporting firms is assessed at high levels. Nevertheless, the innovative
capabilities of the industry are viewed rather skeptically as being still rather low in the
value chain (Arora et al., 2001, Tschang, 2001).
India has reached stage 4 of Yourdon’s (1992) “stages of development”-model (see
figure 1). However, this kind of upgrading is still rather of the product or functional type.
Bhatnagar and Madon (1997) argue that reaching a higher stage necessarily requires
- 12 -
certain technological competencies as well as an understanding of international markets.
Furthermore, they attribute the maturity the Indian software industry has already
achieved to acceptance and reputation in international markets (cf. Banerjee and Duflo,
2000), being endowed with technical competence and capability building among other
factors (Fromhold-Eisebith, 2002; Taeube, 2004a).
Stage Objective Description
1 Build reputation Low value-added body shopping
2 Onshore to offshore Offshore customized software development
3 Improve value addition Starting up offshore package development
4 Product development Total offshore product development
5 Innovation Identify new software-intensive products
Figure 1: Stages of Development (Source: Yourdon, 1992)
A relatively recent tendency in the software industry is to venture into Business Process
Outsourcing (BPO) which sometimes even refers to processes at the core of a firm’s
activities. It started with captive centers founded by MNCs which basically converted data
from one kind of medium (paper) to another (digitized) (Aron and Singh, 2003). This
involves a high degree of human intervention, since in many cases the documents cannot
be reasonably transformed without interpretation. Thus, it also embodies a certain extent
of learning (by doing) and capability building in terms of client-specific as well as more
generic project-management knowledge (Ethiraj et al., 2004). Financial services already
compose the largest part of the business of Indian software companies and are still
gaining in its share; in 2003 the financial sector made up for about 39% of the software
industry revenues, followed by manufacturing (12%) and telecommunication (9%)
(Nasscom, 2003). Having accumulated knowledge and capabilities through supplying
intermediate, rather technical, inputs to the financial services industry some of the
companies venture into new domains by providing financial services themselves
(Economist, 2004, page 9). We assess this potential for upgrading existing IT clusters in
- 13 -
India intersectorally in order to undertake financial research, which would provide
another opportunity for some parts of the developing world.
Methods
The empirical evidence for the analysis presented here draws on qualitative interview
data as well as on quantitative data on Foreign Direct Investment (FDI) transactions of
financial firms in India. Information pertaining to analysts stems from the findings of
interviews carried out during certain former research projects of the authors on the
spatial organization of the financial industry (see Grote, 2004; Grote et al., 2002; Lo and
Grote, 2002; Taeube, 2004c). Six in-depth interviews with analysts that lasted from one
to two and a half hours were conducted between Winter 2003 and Fall 2004 in Frankfurt.
The interviews were open ended; notes were taken during the process. The interviewees
are research analysts and senior analysts in investment banks based in Frankfurt and
London. The interviews focused on the frequency of contacts to any other actors and
information sources and the communication method used. We do not claim to provide a
representative overview in the statistical sense. However, we asked the interview
partners to reflect on other industry practices and former work experiences, which
yielded no different results. This part of the survey is not intended to furnish a
quantitative analysis of the spatial restructuring of financial research activities, but to
reveal the underlying rationales, possibilities and targets for outsourcing and offshoring
different research activities in banks.
The interviews with financial analysts are supplemented by background information from
33 semi-structured interviews with senior executives of IT firms, universities and public
sector entities conducted in Bangalore and Mumbai in Fall 2003. The questions in these
interviews were centered on general business and company information, social, regional
and international networks, and policy. The qualitative evidence from 16 semi-structured
interviews with senior executives of small, medium, and large Indian IT companies in
- 14 -
Frankfurt conducted in October and November 2002 complements the findings from
Bangalore (see Täube, 2004c).
In order to test the hypothesis whether Indian software clusters are attractive
destinations for offshoring such activities by means of FDI, a multivariate regression
analysis is used. Foreign direct investment can either take place as Brownfield (M&A) or
as Greenfield investment; we have compiled data on both. Data on Brownfield
investment of financial firms in India are regressed on several characteristics of target
locations. The results show that M&A activities of financial institutions are highly
concentrated in very few cities with IT clusters in India. Since the data set does not allow
for distinguishing between offshoring and other investments, self-collected newspaper
articles on Greenfield investments in India by banks corroborate the evidence from the
regression analysis.
Qualitative Evidence: Embeddedness of Research Analysts
‘Research’ in banks is a heterogeneous field. In our analysis we concentrate for
illustrative purposes on two examples of research analysts that are the most and least
embedded respectively: institutional equity analysts and country analysts. By asking
analysts in open interviews about their daily work, their data gathering, their contacts
and frequency thereof, a picture of their embeddedness in local and organizational
structures emerges.
Institutional equity analysts work within the wholesale brokerage departments. Clients
are big, mostly institutional investors like pension funds or insurance companies investing
large sums of money. Analysts get into touch with clients often together with sales
persons who maintain relationships with the clients. An analyst’s main task is to
‘generate trading ideas’ for their clients, on which they could (should) trade. These
trading ideas have a time horizon that could last from one day to a few months. In
- 15 -
general, analysts are paid to know more than ‘the market’ does – or at least, before the
market does. Clients do not pay the banks a set fee for the analysis but route trading
volume to that bank in exchange (these are the so-called ‘soft dollars’, a practice now
under interrogation by several regulatory authorities; see, e.g. Financial Services
Authority, 2003). Analysts are paid according to the amount of trading volume they
generate and their position in the ranking lists, which are compiled by industry
observers.
Institutional equity analysts typically cover not more than five big companies in depth
and up to about the same number of smaller ones, mostly in related industries. The
value of a company depends on its future earnings possibilities; therefore, a large part of
what analysts do is project the future. All available data about the company are taken
into account: company sources, analyst meetings, road shows, specialized data
providers, newspaper articles, etc., often collected by research assistants (see Wrigley et
al., 2003, for similar findings). Almost as important are the consequences for the
company ensuing from any other, also seemingly unrelated, events. Analysts interpret a
wide range of signals, including price movements of the company’s or other stocks,
market rumors, and the ‘mood’, etc. There is a lot of number-crunching involved, but this
is used as a tool only. It was stressed that ‘understanding’ what is going on – and what
will be going on – in the company and in the industry is most important (see Agnes,
2000, for analogous results for swap traders).
To get as much information as possible from the current market situation institutional
equity analysts are commonly located right within the trading room of banks and
communicate with all sorts of traders intensively. Since compliance regulations do not
allow even the traders of the own bank to see the recommendations before they are
officially published, the analyst’s desk is situated on a dais about one and a half meters
high above the trader’s level with access restricted only to analysts. The interchange and
testing of ideas with other equity analysts of the own bank is considered as absolutely
- 16 -
necessary; the analysts sit cheek by jowl on the dais in the trading room. Thus, analysts
get to know what is going on in the trading room without any trader being able to see
what they are doing. Interestingly, contact to the own economists – who make longterm,
macroeconomic projections about business cycles and growth, both country- and
industry-wise that analysts use as inputs in their studies – is mostly via email and
reading and writing reports.
Equity analysts do frequently have contacts with the firms they are covering (cf. Palmer
and Sparks, 2004). They use their homepages, telephone contacts with the investor
relations department, attend analyst meetings and sometimes visit the sites. A close
understanding of local regulations and the interpretation of accounting regulations is
viewed as essential. Personal contacts with the heads of investor relations and the chief
financial officers (CFOs) are common and regularly renewed at investor conferences. That
allows analysts to solve problems with the data and to confirm rumors via telephone
more easily. Two analysts stressed that when talking to CFOs, it is crucial how the latter
say something and what they do not say. All interview partners emphasized the high
relevance of close and personal contacts to board members of the firms they cover.
Also of interest is the “demand side”, viz. investors or potential investors. Analysts
frequently speak to the sales persons of the bank. When new studies come out, instead
of just mailing them to investors and informing them of trends, most often they are
presented face-to-face to investors (sometimes in one-to-one meetings) to brief them
about the companies and potential developments relative to the market. This involves
traveling to meet clients – who might be located in Europe, the US or Asia – as often as
three to four times a year. Asked for the reason, one interviewee answered “They won’t
take you seriously otherwise. They’re getting loads of studies everyday via email. You
have to show a commitment.” This confirms the findings by Storper and Venables
(2004), who conclude: “[…] for complex context-dependent information, the medium is
the message” (Storper and Venables, 2004, page 356; cf. Sidaway and Bryson, 2002).
- 17 -
The embeddedness of institutional research analysts is summarized on the left side of
figure 2 below.
Country analysts are to be found on the opposite side of the scale of embeddedness.
They assess country risks and prospects for further regional development. Findings are
published sometimes in public country reports distributed by banks but mostly in internal
information systems, where they are used as inputs in a variety of decisions: in assessing
the credit risk of countries and thus the loan terms for lenders in that country, in
generating a benchmark world-portfolio of equity investments, or in advising clients
before large real or financial investments. These evaluations are updated frequently on
the margin but have an average time horizon of half a year to even a few years.
Typically, a certain number of countries in one region are covered together, depending
on their importance for the bank; e.g. there might be two or three country analysts for
South America in a bank and five for the US and Canada alone.
“Clients” of country analysts are almost exclusively other departments of the same bank.
Face-to-face contacts almost never occur. Sometimes, a department has a specific
question regarding a particular development, but this is solved via telephone in general.
Information is gathered from a variety of sources, mostly internet, newspapers and
research reports from specialized vendors. New developments are discussed with
colleagues via telephone, email and face-to-face when they are close by. In large banks,
the analysts covering a specific area like, e.g. South America are scattered in different
financial centers. Analysts visit the countries they cover to confer with government
officials about once a year. The findings are summarized in the right part of figure 2
below.
--- insert figure 2 about here
Figure 2: Proximities and embeddedness of analysts
- 18 -
The offshoring and/or outsourcing potential of analysts depends on their embeddedness
in local or organizational structures. The two types of analysts described above are two
extreme cases of embeddedness, with many other kinds of economic and financial
research in between possessing characteristics of both sorts. As summarized in figure 2
above, institutional equity analysts need cultural proximity to the firms they cover and
maintain sporadic face-to-face contact with the firms as well as investors. This
requirement confines them to the same nation or at least region as the firms they cover,
such as the EU (the smaller the firms get, the more important is cultural proximity, see
Hau, 2001). Since a lot of tacit knowledge is frequently exchanged with traders and sales
persons, outsourcing as well as offshoring away from these departments seem highly
unlikely. Banks trying to reduce costs in their investment banking departments therefore
try not to outsource or offshore the – expensive – equity analysts but to relieve them
from a lot of non-essential work. Organizational and professional proximity allows for
offshoring tasks that is more than just data processing: internet searches, processing of
data from publications of firms into proprietary databases, standardized calculations and
preliminary firm valuations can be offshored, since very little face-to-face contact is
necessary.
Country economists do not have to be close to the countries they cover and face-to-face
contacts hardly play a role in their daily business. Incoming and outgoing information is
mostly processed electronically. Since they feed mostly proprietary information systems,
offshoring seems to be the best option to cut costs. However, there is not much evidence
on such a move besides sporadic newspaper reports of single banks and consulting firms
that open economic research departments in India. One reason for this missing
offshoring activity might be the small size of these departments (even the largest
international banks employ not more than about 30 country analysts) and the fact that
they are paid considerably less than their investment bank peers.
- 19 -
Summing up the findings so far, offshoring of wholesale equity research is not a
straightforward undertaking due to the strong embeddedness, both local (within the
bank) and regional (with the analyzed firms). Country analysts could be moved more
easily but are not a big cost factor anyway, so the pressure to offshore is not high.
Hence, the main strategy of banks could be to offshore assistant work like gathering and
preparing data and certain calculations – tasks for which Indian IT clusters seem to be
prepared quite well. Given the rising importance of the Indian domestic capital market
there will be a growing presence of financial firms in the Indian financial center of
Mumbai (Shah, 2004). Since these offices will prepare wholesale analysis for the
domestic market anyway there might be attempts to use the existing manpower and
facilities to conduct research on foreign firms, too.
Quantitative Evidence: Brownfield and Greenfield FDI in India by Financial
Institutions
According to our interview partners, pure outsourcing activities of banks’ research are
rare (in fact, none was able to give an example for that). Most banks still regard research
as their core competence and an important source of competitive advantage which
cannot be outsourced (cf. Eisebith, 2002, page 2157). Banks do, however, offshore parts
of their research activities, but only within their own corporate group. They either pursue
Brownfield investments – buying existing Indian companies or parts thereof – or
Greenfield investments, i.e. setting up new branches and subsidiaries. There is only
anecdotal evidence for Greenfield investments of banks, but there are some data
available for M&A transactions of foreign financial institutions in India. Both are analyzed
in order to verify the findings of our interviews with analysts and IT firms in India.
We use an OLS regression analysis to explore the factors that might attract Brownfield
FDI (i.e. inward M&A transactions) in India by financial institutions. The main hypothesis
- 20 -
inferred from the interviews and reasoning above states that the local concentration of IT
firms should strongly influence the localization of banks’ investments in India. Due to
strong agglomeration forces in the financial sector, however, most investments are
expected to take place in locations with an already strong presence of other banks. We
control for the number of commercial banks in these areas and also include a dummy for
Mumbai as the predominant financial centre in India to take care of the other financial
activities (e.g. stock exchange trading and investment banking) that are predominantly
located in Mumbai (Derudder et al., 2003; Poon, 2003). The size of the local economy is
controlled for (approximated by the population of the larger metropolitan areas) as well
as the regional educational level (measured by enrollment in higher education per capita
at the state level). These independent variables were selected to represent forces
commonly accepted as attracting FDI by multinational banks as identified for US (Nigh et
al., 1986), Japanese (Yamori, 1998), and Italian banks (Mutinelli and Piscitello, 2001) as
well as inward FDI in the US (Nachum, 2000).
Data taken from the Thomson Financial SDC Platinum database cover all inward M&A
transactions in India by financial institutions from 1985 to the first quarter of 2004. M&A
data are notoriously scarce and incomplete outside the US and only precious little
information on deal value, locations, and other characteristics is available. With regard to
the hypothesis, the dataset has two shortcomings: it does not allow distinguishing
between pure portfolio investments and real offshoring investments, and it does not
record the specific type of activities pursued in the acquired firms after the transaction.
Therefore, we use the analysis of Brownfield investments only to test whether there is a
general locational trend for banks’ investments in India. We complement the findings
with newspaper reports on Greenfield investments focusing exclusively on the offshoring
of banks’ research activities. Between 1985 and end of the first quarter 2004 Thomson
Financial has recorded 1,471 inward M&A deals in India of which 399 were done by
acquirers with SIC-codes from 6000-6999, i.e. financial institutions and insurances.
Excluding target firms that are neither in the IT nor in the financial industry leaves 219
- 21 -
deals, of which 133 have a known target location in India. None of these deals occurred
before 1992 and only 11 before 1997. Although the number of transactions has not yet
reached the level of the year 2000 again, there has not been a sharp drop either after
the burst of the dotcom-bubble.
Figure 3: Number of inward M&A transactions by year
Source: Thomson Financial, own calculations
The dependent variable in this analysis is the number of transactions (TRANS) that take
place in larger metropolitan areas in India with more than 1 million inhabitants between
1992 and 2004. We use data provided by “The World Gazetteer” for the number of
inhabitants (POP) in each larger metropolitan area in 2004. There are 35 of these areas
in which 131 of the 133 deals with a known target location occur. The remaining two
(target companies in Solan and Trivandrum respectively) are excluded from the analysis.
The number of IT companies per larger metropolitan area is taken from the National
Association of Software and Service Companies (NASSCOM), the apex body and umbrella
organization of IT and IT Enabled Services organizations in India, as on September 2003.
In order to avoid problems of multicollinearity, the number of IT firms per million
0
5
10
15
20
25
30
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 1Q2004*4
(est.)
- 22 -
inhabitants in the metropolitan areas have been calculated (ITCAP). The Reserve Bank of
India, the Indian central bank, provided data on the number of different commercial
banks in the metropolitan areas, a figure commonly used as an approximation for the
size of a financial center (BANK) as on 30th September 2003. To test for the influence of
education on the location of Brownfield investments, the enrolment in higher education
per 1000 capita (EDUCAP) at the state level in 2001 serves as a proxy. Data on
education are provided by the Indian Ministry of Education but are only available for the
states of 34 of the 35 larger metropolitan areas, population data is taken from the
Census of India 2001. The dummy for the Financial Centre (FIN) is 1 for Mumbai and 0
for all other metropolitan areas. Descriptive statistics for the data are shown below in
table 1.
TRANS BANK ITCAP POP EDUCAP
Mean 3.74 44.14 3.25 3,409,003 7.04
Median 0.00 42.00 0.71 1,532,000 7.08
Std. Dev. 9.67 16.09 6.30 4,320,171 2.58
Observations 35 35 35 35 34
Table 1: Descriptive statistics
The model adopts the following form:
TRANS =α + β log(POP) + β FIN + β BANK + β ITCAP + β EDUCAP +ε 1 2 3 4 5
In order to understand the influence of the independent variables on the number of
inward FDI per larger metropolitan area better and to take care of multicollinearity (a
correlation of .76 occurs between the number of banks and the size of the metropolitan
- 23 -
area), a stepwise multiple regression analysis is performed. The results are displayed in
table 2.
dependent variable: TRANS
independent variables (1) (2) (3) (4) (5) (6) (7)
log(POP) 8.85*** 5.74*** 4.16** 2.67*** 3.48*** 2.98*** 1.47*
(6.49) (6.43) (2.63) (3.58) (4.41) (2.78) (1.88)
FIN 34.43*** 33.18*** 37.63*** 37.17*** 38.00*** 41.27***
(7.98) (7.52) (13.15) (11.45) (12.50) (19.06)
BANK 0.10 -0.02 -0.03
(1.20) (-0.41) (-0.71)
ITCAP 0.58*** 0.60*** 0.58***
(6.63) (6.30) (7.01)
EDUCAP 0.98*** 0.23
(4.87) (1.43)
constant -125.68*** -81.07*** -62.37*** -38.17*** -55.38*** -41.71*** -21.61**
(-6,29) (-6.23) (-3.09) (-3.57) (-5.01) (-3.01) (-2.06)
no. of observations 35 35 35 35 34 35 34
R² 0.56 0.85 0.86 0.94 0.92 0.94 0.97
adjusted R² 0.55 0.84 0.85 0.93 0.91 0.93 0.97
F-value 42.05 92.71 63.16 159.39 115.23 116.37 190.64
t-statistics in parentheses; ***, **, * indicate significance at the 1%, 5%, 10% level respectively
Table 2: Regression findings
Foreign direct investments by banks are concentrated in very few cities only (the top
three are Mumbai with 49 transactions; Bangalore with 24, and Delhi with 21), as could
be expected from the literature. The results are highly significant and deliver a high and
robust R-squared and adjusted R-squared. In general, Brownfield investments by
financial institutions in India can be estimated on the basis of a few variables: not
surprisingly, the size of the economy of the larger metropolitan area – the log of the
number of inhabitants – explains to a great extent their attractiveness for foreign
investments (model 1 in table 2). This is in line with the results of Bajpai and Sachs
(2002) and Nachum (2000) that FDI in India is largely determined by the urbanization
rate of a state. Moreover, almost all FDI flows into six states out of which Maharashtra,
where Mumbai is located, has the most favorable investment climate (Dollar et al.,
- 24 -
2002). As expected, the dummy (FIN) for Mumbai, the financial centre of India is highly
significant and adds a large share to the explanatory power of the estimations (model 2).
The number of banks in the metropolitan areas turns out to be never a statistically
significant variable in explaining the attractiveness of a metropolitan region for foreign
banks’ direct investments (models 3, 6, 7). Since the number of different commercial
banks is strongly correlated with the size of the local economy, we conducted a variety of
robustness checks - without any qualitative change in the results. Among others, data
from the Reserve Bank of India (available only at state level) were taken into account:
neither the number or the share of employees in the financial sector, the size of the
financial sector relative to the state GDP, nor the state relative to the national GDP in
banking explain the attractiveness of metropolitan areas. Whereas banks tend to invest
in the predominant financial center, the size of other financial centers does not influence
the number of investments in a statistically significant way. This is consistent with the
findings by Yamori (1998) and Nachum (2000).
We are mainly interested in the relationship between the number of IT companies and
the attractiveness of a metropolitan area for financial institutions: ITCAP, the number of
IT firms per capita, is always highly significant – and, as all the variables – has the
expected sign. The locations of banks’ investments in India are strongly influenced by the
local density of IT firms in the metropolitan area (models 4, 6 and 7). This is in line with
Gholami et al. (2003) who find that the general level of inward FDI increases with the
level of investment in information and communication technologies. Here it is the IT firms
themselves that attract investments.
The educational level of the population is significant (model 5) when analyzed in
isolation. However, when included in the analysis in combination with IT and banking
(model 7) or only with IT (not shown in the table), the significance is lost. Similarly, Self
- 25 -
and Grabowski (2004) were able to establish a causal relationship from education to
growth for both primary and secondary, but not for the tertiary education.
In general, banks tend to invest in the main financial center Mumbai; further investments
are driven by the size of the local economy and the local density of IT firms. The size of
the local financial centers, measured by the presence of commercial banks in the areas,
does not yield significant results. Foreign investments by financial institutions in India are
concentrated in a few – broadly defined – clusters. This is in line with the argument
developed in the paper that Indian IT clusters can profit from intersectoral upgrading into
the financial sector. However, the data underlying these findings do not allow for
distinguishing between banks’ investments to shift research activities and other FDI.
Newspaper reports on banks that set up research departments in India (“Greenfield
investments”) are taken into account to substantiate the observed trend.
On top of the Brownfield investments made by foreign banks, more recently there is also
some (anecdotal) evidence of investment banks relocating their research activities (e.g.
Atal and Niranjan, 2004; Kulkarni, 2004). We have collected data from newspaper
articles and the internet on Greenfield investments by foreign financial service firms, in
particular investment bank research activities over the last 18 months. The following map
1 provides an overview of FDI of financial services firms in India, both Brownfield and
Greenfield.
---- insert map 1 about here
Map 1: FDI by financial institutions in India
Legend: The bars indicate the number of Brownfield investments; the stars denote
the number of Greenfield investments.
- 26 -
Most of the Greenfield investments take place in the existing IT centers Bangalore,
Hyderabad, and Chennai (Madras) as well as the financial capital Mumbai and the
national capital region Delhi. The dataset focuses exclusively on the setting up of
research activities of financial firms without the acquisition of an Indian company. These
data show a pattern quite similar to the distribution of Brownfield investments and thus
substantiate the findings from the regression analysis above.
Conclusion
The long-run performance of clusters depends on their ability to capture increasing
shares of value chains and to adapt to changing environments. In the paper, an
enhanced value chain concept is developed that allows for an assessment of the local
embeddedness of actors within the value chain. This general concept could be applied to
a variety of industries in order to analyze opportunities for spatial reorganization. We
have looked particularly at the potential to do financial research in Indian IT clusters.
Indian IT firms have built up a reputation and capabilities that would enable an
intersectoral upgrading of the IT clusters, especially with regard to banking and financial
services. The analysis of research departments in Western financial centers revealed the
embeddedness of the highly debated wholesale equity research analysts that makes
offshoring unlikely. Other analysts – covered here are country analysts – are embedded
only to a very low extent and could therefore be transferred to a low-wage country like
India without greater loss of information and knowledge. Most likely is the migration of
technical analyses and data gathering assistant work to India. Generally, in terms of the
governance structure most financial firms will not outsource their critical and high valueadded
research activities, but rather keep them offshore in-house or in captive centers.
Hence, the potential for upgrading lies in the first place in offshoring of foreign banks’
activities to Indian IT clusters, allowing them to tap into the global financial production
network.
- 27 -
As the quantitative analysis shows, investments by financial institutions in India are quite
concentrated in a few – largely defined – clusters. More than 98% of Brownfield
investments take place in these clusters, with Mumbai leading by a large margin,
whereas Greenfield investments are divided between Delhi, Bangalore, Mumbai, and
Hyderabad. The density of IT companies (IT companies per capita) contributes
significantly to the explanation of the location of investments by financial institutions, as
indeed financial firms look for IT clusters to invest in. This confirms the hypothesis that
upgrading to financial research might be a viable option for the Indian IT clusters. In
order to address the specificities of each cluster and to identify the features that attract
financial institutions, research at the micro-level within these clusters is needed to
complement the findings of the macro approach presented here.
Acknowledgements. We would like to acknowledge the comments of participants in the
2003 EADI Workshop, Novara, the 2004 Centennial Meeting of the American Association
of Geographers, Philadelphia, the 10th International Schumpeter Society Conference,
Milan, and the 2004 DRUID Summer conference, Helsingør. We wish to thank Martin
Hess, Henry Yeung and Jamie Peck and three anonymous referees for critiques and
suggestions which helped improving the flow of our argument to a great extent.
- 28 -
References
Agnes P, 2000, “The "End of Geography" in Financial Services? Local Embeddedness and
Territorialization in the Interest Rate Swaps Industry.” Economic Geography 76
347-366
Almeida P, Kogut B, 1999, “The localization of knowledge and the mobility of engineers in
regional networks” Management Science 45 905–917
Appelbaum R P, Smith D, Christerson B, 1994, “Commodity chains and industrial
restructuring in the Pacific Rim”, in Commodity chains and global capitalism Eds G
Gereffi, M Korzeniewicz (Westport: Praeger) pp 187-198
Aron R, Singh J, 2003 IT Enabled Strategic Outsourcing: Knowledge Intensive Firms,
Information Work and the Extended Organizational Form The Wharton School
(University of Pennsylvania)
Arora A, Arunachalam V S, Asundi J & Fernandes R, 2001 The Globalization of software:
The Case of the Indian Software Industry, a report submitted to the Sloan
foundation (Carnegie Mellon University)
Atal S, Niranjan I, 2004, “Higher-end BPO jobs for MBAs, CAs” The Economic Times 20
September, http://economictimes.indiatimes.com/articleshow/856464.cms
Bajpai N, Sachs J, 2002, “Understanding regional economic growth in India” Asian
Economic Papers 1 (3) 32-62
Banerjee A V & Duflo E, 2000, “Reputation Effects and the Limits of Contracting: A Study
of the Indian Software Industry” Quarterly Journal of Economics 115 (3) 989-
1017
Bathelt H, 2000, „Räumliche Produktions- und Marktbeziehungen zwischen Globalisierung
und Regionalisierung – Konzeptioneller Überblick und ausgewählte Beispiele“
Berichte zur deutschen Landeskunde 74 (2) 97-124
Bhatnagar, S C, Madon, S, 1997, “The Indian software industry: moving towards
maturity” Journal of Information Technology 12 277-288
Birkinshaw J, Hagström P, 2003 The flexible Firm: Capability Management in Network
Organizations (Oxford University Press, Oxford and New York)
Blanc H, Sierra C, 1999, “The internationalisation of R&D by multinationals: a trade-off
between external and internal proximity” Cambridge Journal of Economics 23
187-206
- 29 -
Cantwell J, Mudambi R, 2004, “On the Geography of Knowledge Sourcing: A Typology of
MNE Subsidiaries”, paper presented at the DRUID Summer Conference, Helsingør
Clark G L, 2002, “London in the European financial services industry: locational
advantage and product complementarities” Journal of Economic Geography 2 433-
453
Cohen W M, Levinthal D A, 1990, “Absorptive capacity: a new perspective on learning
and innovation” Administrative Science Quarterly 35 (1) 128-152
Cowan R, Foray D, 1997, “The Economics of Codification and the Diffusion of Knowledge”
Industrial and Corporate Change, 6 (3) 595-622
Cronin B, Catchpowle L, Hall D, 2004, “Outsourcing and Offshoring” CESifo Forum, 5 (2)
17-21
Dahlman C, Nelson R R, 1995, “Social Absorption Capability, National Innovation
Systems and Economic Development”, in Social Capability and Long-term Growth
Eds D Perkins & B Koo (Macmillan Press, Basingstoke)
D’Costa A, 2002, “Export Growth and Path Dependence: The Locking-in of Innovations in
the Software Industry” Science, Technology & Society 7 (1) 13-49
Deloitte Consulting, 2003 Right-Sizing Research. New business economics: the future of
investment banking research (New York)
Derudder B, Taylor P J, Witlox F, Catalano G, 2003, “Hierarchical Tendencies and
Regional Patterns in the World City Network: A Global Urban Analysis of 234
Cities” Regional Studies 37 (9) 875-886
Dicken P, 1998 Global shift (Paul Chapman, 3rd edition, London)
Dollar D, Iarossi G, Mengistae T, 2002, “Investment Climate and Economic Performance:
Some Firm Level Evidence from India”, WP 143, Center for Research on Economic
Development, Stanford
Dosi G, 1988, “Sources, procedures and microeconomic effects of innovation” Journal of
Economic Literature 26 1120-1171
The Economist 2004, “A world of work - A survey of outsourcing”, November 13th
Ethiraj S, Kale P, Krishnan M S, Singh J, 2004, “Where do capabilities come from and
how do they matter? A study in the software services industry”, Strategic
Management Journal (in press)
Financial Services Authority, 2003 Bundled Brokerage and Soft Commission
Arrangements, Consultation Paper 176 (London)
- 30 -
Franco A, Filson D, 2000 Knowledge Diffusion through Employee Mobility Research
Department Staff Report 272 (Federal Reserve Bank of Minneapolis)
Friedmann J, 1995, “Where we stand: a decade of world city research”, in World cities in
a world-system Eds P Knox & P Taylor, (University Press, Cambridge) pp 21-47
Fromhold-Eisebith M, 2002, “Regional cycles of learning: foreign multinationals as agents
of technological upgrading in less developed countries” Environment and Planning
A 34 (12) 2155-2173
Gereffi G, 1999, “International Trade and Industrial Upgrading in the Apparel Commodity
Chain” Journal of International Economics 48 37-70
Gertler M, 1993, “Implementing Advanced Manufacturing Technologies in Mature
Industrial Regions: Towards a Social Model of Technology Production” Regional
Studies 27 (7) 665-680
Gholami R, Lee S T, Heshmati A, 2003, “The Causal Relationship between Information
and Communication Technology and Foreign Direct Investment”, WIDER
Discussion Paper No. 2003/30
Giddens A, 1990 The consequences of modernity (Stanford University Press, Stanford)
Giuliani E, 2004 forthcoming, “Cluster Absorptive Capacity: Localised Learning in
Developing Countries” European Urban and Regional Studies
Glaeser E, 1999, “Learning in Cities” Journal of Urban Economics 46 254-277
Granovetter M, 1990, “The Old and New Economic Sociology: A History and an Agenda”,
in Beyond the Market Place: Rethinking Economy and Society Eds R Friedlang, A F
Robertson (de Gruyter, New York) pp 89-112
Grote M H, 2004 Die Entwicklung des Finanzplatzes Frankfurt seit dem Zweiten Weltkrieg
- Eine evolutionsökonomische Untersuchung (Duncker & Humblot, Berlin)
Grote M H, Lo V, Harrschar-Ehrnborg S, 2002, “A value chain approach to financial
centers – The case of Frankfurt” TESG Journal of Economic and Social Geography
93 (4) 412-423
Hess M, 2004, “‘Spatial’ relationships? Towards a reconceptualization of embeddedness”
Progress in Human Geography 28 (2) 165-186
Hau H, 2001, “Location Matters: An Examination of Trading Profits” Journal of Finance 56
(5) 1959-1983
Humphrey J, Schmitz H, 2000, “Governance and Upgrading: Linking Industrial Cluster
and Global Value Chain Research”, IDS WP 120
- 31 -
Kindleberger C, 1974, “The formation of financial centers: A study in comparative
economic history” Princeton Studies in International Finance 36
Klepper S, 2001, “Employee Startups in High-Tech Industries” Industrial and Corporate
Change 10 (3) 639-674
Kogut B, 1985, “Designing global strategies: comparative and competitive value-added
chains” Sloan School of Management Review 26 (4) 15-28
Krishnan R T, Prabhu G N, 2002, “Innovation in the Indian Information Technology
Industry: A Study of the Software Product Development Process” Science,
Technology & Society 7 (1) 91-115
Kulkarni S, 2004, “JP Morgan to hire up to 4000” The Economic Times September 23,
http://economictimes.indiatimes.com/articleshow/860250.cms
Lateef A, 1997, “Linking up with the Global Economy: A Case Study of the Bangalore
Software Industry”, Discussion Paper 96/1997, International Institute for Labour
Studies, Geneva
Leslie D, Reimer S, 1999, “Spatializing commodity chains” Progress in Human Geography
23 401-420
Lo V, Grote M H, 2002, “Where traders go when stock exchanges go virtual –
concentration, dissemination or persistence?”, in Technology and Finance:
Challenges for financial markets, business strategies and policy makers Eds M
Balling, F Lierman, A Mullineux (Routledge, London) pp 190-203
Lorenzen M, Mahnke V, 2004 forthcoming, “Global strategy and the acquisition of local
knowledge: How MNCs enter regional knowledge clusters” Scandinavian Journal of
Management
Lundvall B A, 1988, “Innovation as an interactive process: from user-producer interaction
to the national system of innovation”, in Technical Change and Economic Theory
Eds G Dosi et al. (Pinter, London) pp 349-369
Mahnke V, 2001, “The Process of Vertical Dis-Integration: An Evolutionary Perspective on
Outsourcing” Journal of Management and Governance 5 353–379
Malmberg A & Maskell P, 2002, “The elusive concept of localisation economies: towards a
knowledge-based theory of spatial clustering” Environment and Planning A 34
429-449
Martin R, 1999, “The new economic geography of money”, in Money and the space
economy Ed R Martin (John Wiley & Sons, Chichester) pp 3-27
Maskell P, Malmberg A, 1999, “Localized Learning and Industrial Competitiveness”
Cambridge Journal of Economics 23 (2) 167-185
- 32 -
Mutinelli M, Piscitello L, 2001, “Foreign direct investment in the banking sector: the case
of Italian banks in the ’90s” International Business Review 10 (2001) 661–685
Nachum L, 2000, “Economic Geography and the Location of TNCs: Financial and
Professional Service FDI to the USA” Journal of International Business Studies 31
(3) 367-385
Nahapiet J, Ghoshal S, 1998, “Social Capital, Intellectual Capital, and the Organizational
Advantage” Academy of Management Review 23 (2) 242-266
Nasscom, 2003, www.nasscom.org
Nelson R R, Winter S G, 1982 An Evolutionary Theory of Economic Change (Belknap,
Cambridge)
Nigh D, Cho K R, & Krishan S, 1986, “The role of location-related factors in US banking
involvement abroad: an empirical examination” Journal of International Business
Studies 17 (3) 59–72
Nonaka I, Takeuchi I, 1995 The knowledge-creating company: How Japanese companies
create the dynamics (Oxford University Press, Oxford)
Palmer M, Sparks L, 2004, “Investment bank analysts and retail research” Environment
and Planning A 36 1521-1528
Poon J P H, 2003, “Hierarchical Tendencies of Capital Markets Among International
Financial Centers” Growth and Change 34 (2) 135-156
Porteous D, 1999, “The development of financial centres: Location, information
externalities and path dependence”, in Money and the space economy Ed R Martin
(John Wiley & Sons, Chichester) pp 95-114
Porter M E, 1990 The Competitive Advantage of Nations (MacMillan, Basingstoke and
London)
Porter M E, 1998, “Clusters and the New Economics of Competition” Harvard Business
Review 76 (6) 77-91
Rabach E, Kim E M, 1994, “Where is the chain in commodity chains? The service sector
nexus”, in Commodity chains and global capitalism Eds G Gereffi, M Korzeniewicz
(Praeger, Westport) pp 123-141
Rabellotti R, Pietrobelli C, 2004, Upgrading in Clusters and Value Chains in Latin America:
the Role of Policies Inter-American Development Bank (Washington D.C.)
Sassen S, 1991 The global city: New York, London, Tokyo (University Press, Princeton,
NJ)
- 33 -
Saxenian A, Motoyama Y, Quan X, 2002 Local and Global Networks of Immigrant
Professionals in Silicon Valley Public Policy Institute of California
Schmitz H, 1995, “Collective Efficiency: growth path for small-scale Industry” Journal of
Development Studies 31 (4) 529-566
Self S, Grabowski R, 2004, “Does education at all levels cause growth? India, a case
study” Economics of Education Review 23 47–55
Shah N, 2004, “Destination India for big foreign brokers” Rediff.com 16 Nov,
http://in.rediff.com/money/2004/nov/16broker.htm
Sidaway J D, Bryson J R, 2002, “Constructing knowledges of ‘emerging markets': UKbased
investment managers and their overseas connections'' Environment and
Planning A 34 401-416
Stein R, 2002, “Producer Services, Transaction Activities, and Cities: Rethinking
Occupational Categories in Economic Geography” European Planning Studies 10
(6) 723-743
Storper M, 1995, “The resurgence of regional economies, ten years after; the region as a
nexus of untraded interdependencies” European Urban & Regional Studies 2 191-
221
Storper M, 1997 The Regional World: Territorial Development in a Global Economy
(Guilford Press, New York)
Storper M, Venables A J, 2004, “Buzz: Face-to-Face contact and the Urban Economy”
Journal of Economic Geography 4 (4) 351-370
Taeube F, 2004a, “Culture, Innovation and Economic Development: The Case of the
South Indian ICT Clusters”, in Innovation, Learning and Technological Dynamism
of Developing Countries Eds S Mani, H Romijn (UNU Press, New York) pp 202-228
Taeube F, 2004b, “Transnational Networks and the Evolution of the Indian Software
Industry: The Role of Culture and Ethnicity”, in The Role of Labour Mobility and
Informal Networks for Knowledge Transfer Eds D Fornahl, C Zellner, D Audretsch
(Kluwer, Dordrecht) pp 97-121
Taeube F, 2004c, “Proximities and Innovation: Evidence from the Indian IT industry in
Bangalore”, DRUID WP 04-10
Thrift N, 1994, “On the social and cultural determinants of international financial centres:
The case of the City of London”, in Money, power and space Eds S Corbridge, R
Martin & N Thrift (Blackwell, Oxford) pp 327-355
Tschang T, 2001, “The Basic Characteristics of Skills and Organizational Capabilities in
the Indian Software Industry”, ADB WP 13, Asian Development Bank, Manila
- 34 -
Uzzi B, 1997, “Social Structure and Competition in Interfirm Networks: The Paradox of
Embeddedness” Administrative Science Quarterly 42 (1) 35-67
Walter I, 2004, “Conflicts of Interest and Market Discipline Among financial Service
Firms” European Management Journal 22 (4) 361-376
Wrigley N, Currah A, Wood S, 2003, “Investment bank analysts and knowledge in
economic geography” Environment and Planning A 35 381-387
Yamori N, 1998, “A note on location choice of multinational banks: the case of Japanese
financial institutions” Journal of Banking and Finance 22 109–120
Yourdon E, 1992 Decline and Fall of the American Programmer (Prentice Hall, Upper
Saddle River, NJ)
Zeller C, 2004, “North Atlantic Innovative Relations of Swiss Pharmaceuticals and the
Proximities with Regional Biotech Arenas” Economic Geography 80 (1) 83-111Management of Information Systems:
Insights from Management Accounting Research
Neale G. O’Connor*
The University of Hong Kong
Pokfulam Road
HONG KONG
E-mail: acno@cityu.edu.hk
and
Maris G. Martinsons
City University of Hong Kong
83 Tat Chee Avenue
Kowloon Tong, Kowloon
HONG KONG
E-mail: mgmaris@cityu.edu.hk
* Corresponding author
May 13, 2006
Management of Information Systems:
Insights from Management Accounting Research
Abstract
This paper seeks to advance our knowledge in key areas of information systems
(IS) research by applying ideas and insights from accounting to improve our
understanding of IS management. An integrative cost-benefit framework is
proposed and used to examine four areas of research that are related to the
management of an IS: chargeback, outsourcing, decision support, and business
process re-engineering and improvement. The paper specifies how the
accounting literature contributes significantly to each of the four areas of IS
management research, and highlights key questions for further study.
KEYWORDS: information systems value, IT management, outsourcing,
chargeback, decision support, business process re-engineering, business
performance.
1
I. INTRODUCTION
Information systems must be managed at various levels. Enterprise level
decisions primarily revolve around the allocation of resources to IS-related activities.
Functional level IS decisions include where and when to apply IT, and whether new
applications should be made or bought, either purchased off the shelf or developed
through an outsourcing arrangement. A decision to “make” a new IT application
internally will inevitably lead to the initiation of a project that must be managed.
Important issues that are related to each of the aforementioned decisions
have been considered by both IS and accounting researchers. For example, a
steady stream of IS research on outsourcing began to emerge after the landmark
decision by Kodak to outsource its IS [59, 5]. Similarly, accounting research has
examined the initial outsourcing decision [89, 99, 48]. However, the IS literature
has inadequately addressed other research questions that pertain to the
management of IT, many of them involving post-implementation decisions. This
paper addresses these inadequacies directly by applying an accounting
perspective to four specific research areas within the post-implementation realm:
chargeback for IS services, post-contractual management of IS outsourcing,
management decision support, and IT-enabled operational process improvement.1
Research on each of these four areas has been published widely in major
IS journals [92]. Decision support and process re-engineering affect the
managerial and operational processes of an organization, respectively, while
outsourcing has been identified as a major technological change that is affecting
the IT organization. Meanwhile, the management of chargeback is critical to two
1 These four areas are by no means an exhaustive list of areas in which accounting research
has contributed. For example, researchers in the area of business value of IS investments
can draw insights from accounting researchers application of event study methodology, which
they commonly employ in capital markets research. Further, IS project management is
another well researched area that shares theory and methods with the accounting research
into the implementation of various technologies such as activity based costing, balanced
scorecard and enterprise resource planning.
2
of the eight imperatives identified by [87] for the new IT organization—achieving
strategic alignment and developing a high performance culture.
The four areas have also been examined frequently in the management
accounting literature [62, 14]. Although these areas address different issues and
have tended to be examined in isolation, they share a dependence on accounting
measures. First, the measurement of costs and cost drivers underpins the use of
chargeback, facilitates the decision to outsource and often helps in the control of
the outsourcing relationship. Second, measurement systems that span the value
chain (e.g. the balanced scorecard) are commonly used as part of a managerial
decision support system. Such measures are also used for the purpose of
controlling operational process change, because you can only manage and
change what you can measure.
While the accounting perspective is relevant to many IS issues, the most
recent advances in accounting research has focused on decision facilitating and
management control issues after an IS has been implemented [43]. As a result,
this paper aims to help researchers consider the role of accounting in managing:
(i) the intangible aspect of IT projects (both inhouse and outsourced), which
raises the problems of risk assessment, control, and coordination; (ii) the
judgment and decision-making biases that are associated with using a DSS, and
(iii) the authority and incentive structure of the firm, such that they complement
operational process change.
The next section presents a conceptual framework, which is first used to
review the overlap between IS and accounting research. We then consider each
area in turn, to 1) summarize the IS literature, 2) specify how the accounting
perspective can augment mainstream knowledge, and 3) identify key issues for
further research.
3
II. CONCEPTUAL FRAMEWORK: ACCOUNTING COSTS/BENEFITS
Accounting scholars commonly believe that by more accurately attributing
costs to products, services and customers, accounting can greatly improve the
quality of information for operating decisions. Similarly, by more accurately
attributing employee effort to organizational outcomes, accounting can improve
the motivation and goal alignment of employees. Both of these roles are aimed at
enhancing the management of IS. The success of these roles ultimately depends
on what objects (e.g., costs, activities, or processes) are measured and how well
they are measured. Typically, these measurements are translated into monetary
terms to evaluate the costs and benefits of various decisions or activities.
To introduce and relate the four areas of IS, it is helpful to consider the
process of IT application depicted in Figure 1. An organization must first decide
to the extent to which it will develop and operate its own IS. It may decide to
outsource some or all of these activities. The value of each choice is based
primarily on (i) the net benefits of outsourcing—the economic contribution of IT
less the costs of contracting and control—and (ii) the benefits of in-house IS less
the initial implementation and ongoing management costs.
Figure 1 – The Process of IT Application
The economic benefits of IS generally take one of two forms: i) the
reduction of operational costs through process improvement (by automating,
streamlining or re-engineering capital and labor intensive activities) [35, 65], and
ii) the improvement of resource allocation by providing more accurate and timely
information to decision makers. The potential scope of IT application in an
Make
Buy
Commitment
to a new IT
application
Outsource development
Internal development
Outsource operation
Internal operation
4
organization ranges from the strategic level, where decisions concern the
allocation of resources and the development of new ventures, to the operational
level, where frontline workers can access information systems to make more
accurate and timely decisions concerning suppliers, production, marketing and
sales [66]. The cost/benefit framework in Figure 2 shows these two general types
of IT benefits. It also depicts IT costs, which include: (i) initial development, which
entails decisions about a combination of in-house programming and off the shelf
software packages plus hardware acquisition versus outsourcing, and (ii) ongoing
maintenance costs, based upon a decision of whether or not to outsource
IS operations. The core issues (questions) from an accounting perspective in
each area of IS research are highlighted, because these issues drive the relevant
part of IS design and thus ultimately affect the effectiveness of such design.
Initial Development Ongoing maintenance costs/management
(Accounting research questions)
Outsource operation
Contracting costs
Post contracting controls
(What is the extent to which formal
measures can be used to strengthen
cooperative ties?)
Management/Decision Aid
Value of IT
(How can the use of de-biasing
mechanisms help to improve the
effectiveness of DSS?
How can the benefits of decision aid be
measured?)
Business Process Improvement
(How can structure and control in the
organizational change process be
balanced?)
Internal operation
IT Planning (in-house
programming, hardware
acquisition)
IT control – Chargeback
(How can accountability and flexibility be
balanced?)
Figure 2: Cost/Benefit framework – with questions from an
accounting perspective
Benefits
Costs
Costs
5
III. FOUR AREAS OF INFORMATION SYSTEMS RESEARCH
MANAGING THE COSTS OF IS SYSTEMS
A significant concern for both IS researchers and managers is whether and
how to recover the costs of operating and maintaining computer-based
information systems. Within the IS literature, this has been defined as the
“chargeback” issue [10, 71, 90]. A related issue is the outsourcing of an IS.
Outsourcing decisions are important to IS researchers because, compared to
other servicing processes (such as human resources and accounting), IT usually
represents a larger share of an organization’s budget, and outsourcing provides
an opportunity to leverage the core competence of another organization that
specializes in IT services.
CHARGEBACK (INTERNAL TRANSFER PRICING)
A chargeback system traditionally bills cost centers or user departments for
in-house IS services. There are several chargeback alternatives, such as cost
minimization [23], flexible pricing—which was advocated to resolve peak load
problems [30,101]—and the setting of standard rates based on the elapsed time,
the estimated fraction of the system used by the job, and a time-adjusting factor
that accounts for the job’s priority and mix in the system [94].
The rationale in most cases is to encourage the responsible and efficient
use of IS resources, which has become more important with increasing
technology investment and a growing diversity of users and uses. However, such
a system inevitably raises the question of what terms and conditions should be
applied. What chargeback system would be fair to both providers and users?
Several tradeoffs are depicted in Figure 3. According to Prendergast [86],
chargeback helps to create a culture of accountability that communicates the cost
of IT to users, who in turn act as independent monitoring devices. However, the
demand for accountability may have to be tempered by a flexible approach that
encourages volitional use that is critical to a firm’s strategy (see Figure 3). Finally,
6
can chargeback be used to match the interdependency between service
department capacity and external (non-service) department demand?2 For
example, service department capacity will influence IS service cost, but it depends
on external department demand. In an uncertain environment, the notion of IT
capacity will constantly mirror changes in the interdependency between service
and external departments. This in turn will influence the optimal cost levels for
external departments and services.
Accountability Flexibility
Objective: Resource
control
Objective: Strategic
resource use
Interdependency
Objective: Efficient resource planning
and allocation
Externalities: Congestion, information,
incentives
Trade-off between User demand and IS
support capacity
Figure 3. Drivers of chargeback design
IS researchers have actively examined various chargeback issues from the
perspective of what determines chargeback practices. The drivers of chargeback
have their roots in: (i) the management control philosophy, (ii) the strategic role of
IT, and (iii) the organization’s ability to measure costs and performance using
other techniques. Although chargeback practices have not changed significantly,
the difficulties of both allocating costs and effectively educating users about using
chargeback information have increased steadily over time. Moreover, research
into the role of chargeback between the two extremes of volitional use and control
has yet to address issues of fairness, equity, responsibility, and controllability.
These issues require new frameworks and instruments that go beyond those used
in traditional IS effectiveness studies.
2 Depending on the organization’s information needs, service department capacity can relate
to either service quality or information quality.
7
Chargeback studies from the 1980s may have limited relevance in today’s
IT-intensive business environment [108]. For example, the growing
portability/mobility and declining cost-to-functionality ratio of IS hardware allows
user departments to bypass the centralized IS department and purchase both
hardware and software with their own budgets. The traditional IS service center is
also imperiled as departments acquire their own IS specialists, while networking
and maintenance services remain centralized. The organizational dispersion of IT
resources reduces the central transparency of costs and IT resource
consumption. The consequences of this are (i) understated IT costs and (ii)
duplicated effort (multiple applications being run on multiple platforms). To
address these developments, the charge out framework must be expanded to
consider the IS budget setting process, IS performance management [67], and
communication between IT providers and users to clarify their respective roles
and responsibilities [109].
Similarly, the adoption of new technology involves user apprehension that
is associated with the perceived benefits of use, steep learning curves, and
uncertainty about the real costs of use. As IT adoption and upgrades take place at
increasing rates, many organizations simultaneously have at least two
generations of IT. As this increases the complexity of the chargeback system,
Keller and Allen [49] suggest that a sound cost-benefit framework should be set
up to identify the costs in question before the chargeout system is designed.
In short, researchers have yet to fully consider the dynamic and dispersed
nature of IT applications, and have failed to view IS services as different from
other services. The nature of IS services, being an inherent part of organizational
structure and management, is such that complementarities and externalities are
not fully considered.
Accounting perspective
Two accounting concepts provide guidance for designing and employing
an IT chargeback system: internal cost-resource utilization within functional areas,
8
and the effects of externalities in the cost allocation process. Fairness in charging
users for IT services is a commonly accepted principle [40]. However, putting this
principle into practice can be difficult when organizational units are crossconnected.
In such a case, IT use by one department will influence that of
another department, and ultimately affect service department capacity. A poorly
designed chargeback system will result in negative externalities, whereby costs
will be imposed on other individuals without their participation in the decision. For
example, Zimmerman [113] cites incentive effects, information effects, and
congestion effects. Incentive effects arise because managers over-consume
resources to the detriment of the firm's overall profitability. Information effects
relate to the effect that charging systems have in establishing a mutual monitoring
relationship which overcomes the problem of asymmetric information between
users and suppliers. Congestion effects refer to the imposition of delays and
rationing costs on other users within the organization. These result from the
absence of chargeback, whereby no cost is placed on the use of limited resources.
The tradeoff between flexibility, which results in no self-monitoring and little
concern for the resources that are consumed by the IT department, and
accountability, can also stifle volitional use of IT services and resources. We
identify three specific aspects within the chargeback area where accounting
provides guidance for future research. They relate to (i) balance of control and
volitional use, (ii) where strategic value (and thus a greater need for volitional
use) is located, and (iii) the control mechanisms that will overcome the absence of
chargeback-type accountability in these areas.
First, although it is logical to recoup the cost of IT investment according to
a user-pays formula, the practical situation in which a balance is struck between
accountability and flexibility is more complex for several reasons. It would thus be
useful to have a better understanding of the extent to which minimum levels of
chargeback act to increase user awareness of IT services. There appears to be a
spectrum in which too little chargeback results in no self-monitoring by IT users
and little concern for the resources consumed by the IT department, whereas too
9
much chargeback stifles the volitional use of the IT services and resources. How
this balancing act between control and use is related to the economic value of IT
is a question for future research.
Second, the accounting perspective acknowledges the need for volitional
activity to be encouraged in strategic areas, such as research and development.
However, this acknowledgement does not eliminate the need to evaluate and
control the progress of a strategic project. The ability to determine a set of critical
success factors for a strategic information system makes it appropriate to identify
and monitor a few key financial measures that can indicate how well these critical
success factors are being achieved while comparing the actual and expected
costs of achieving important milestones [70]. This is consistent with the position
of Rockness and Shields [88], who concluded that input and behavioral controls
are appropriate when an organizational task involves a high degree of
technological uncertainty or has outputs that are difficult to measure.
Third, it is important to recognize that chargeback systems are not the only
means of guiding and controlling the actions of IS users. Research into nonchargeback
systems can shed light on the effectiveness of chargeback. Research
into how incentive systems that are tailored for the IS department and user
departments can direct and motivate innovative behavior would be particularly
useful. For example, Drake, Haka and Ravencroft [20] show that the interaction of
the type of incentive system with cost allocation can affect profitability,
productivity, innovation, and the exchange of information between parts of
organizations.
The accountant’s emphasis on economic measurement and value
highlights the importance of using chargeback for strategic company-value
purposes. Research into the linkages between the strategic objectives of IT use
and the use of chargeback in enhancing this role would be of value. A key
challenge for researchers is to understand and reconcile the chasm between two
sets of conflicting views from practice: those advocating the use of chargeout for
new information technologies [17, 82], and those criticizing it [27].
10
A final piece of the chargeback puzzle lies with the potential tradeoff
between the encouragement of best IS support practices (i.e. minimal
chargeback) and the decision to outsource, which is driven by gaps in information
quality and IS support quality. Is an unsuccessful chargeback system a precursor
to outsourcing? For these reasons, economists and accounting researchers
recommend that there should be either no cost pricing or marginal cost pricing at
the most [113, 38]. Research into the nature of IT use and the type of chargeback
practices would also be helpful to determine the linkages between the chargeback
and outsourcing of IS.
OUTSOURCING
Outsourcing involves the contracting out of all or part of a company’s
activities or projects to external parties. IS outsourcing decisions are
characterized by their size, complexity, and potential irreversibility. The benefits of
outsourcing IS activities include reduced costs due to the outsourcing vendor’s
economies of scale, immediate access to new technology and expertise, strategic
flexibility, and avoiding the risk of obsolescence [68]. The complexity of IS
outsourcing is characterized by its intangible nature, multiple stakeholders with
varying objectives, and the delivery of the service over time. These factors
highlight the need to consider the value of such services over several (including
post-contractual) stages to evaluate their success or feasibility. In addition, the
post-contractual stages require informal forms of governance that bring the
contracting parties (which are initially at arms length) into a quasi-hierarchical
cooperative relationship.
An extensive body of research on outsourcing in the IS literature dates
back to Eastman Kodak’s pioneering decision in 1989 to outsource its mainframe
computers to IBM. Applegate and Montealagre [5] documented the effect of
Eastern Kodak’s decision in terms of the quality of processes and services, while
Loh and Venkatraman [59, 60] found positive stock market reactions to
outsourcing announcements. Subsequent IS research has found that the
11
determinants of successful outsourcing outcomes (such as quality processes and
services) include the sharing of knowledge, having a detailed formal evaluation
process, using shorter-term contracts, and outsourcing commodity IT on a
selective basis [31, 52]. Lacity and Willcocks [54] identified 43 research articles
that used a variety of methods to study IS outsourcing.
IS researchers have generally applied transaction cost economics to
understand the rationales for outsourcing, such as the avoidance of obsolescence
risk, access to new technology, and vendor economies of scale. However, IS
research has inadequately accounted for the costs that are associated with the
management and completion of IS outsourcing projects. The customer and
supplier/stakeholder relationship that has been portrayed by applying transactions
cost theory is an overly simplistic representation of a phenomenon made complex
by: 1) the expectations of different stakeholders in outsourcing and 2) the
existence of six outsourcing phases: scoping, evaluation, negotiation, and the
transition, middle, and mature phases. Teng, Cheon and Grover [104] examined
the relationships between several strategy-theoretic factors and the IS
outsourcing decision. These factors include, gaps in information quality, IS
support quality and IS cost effectiveness, and the strategic orientation of the firm.
Their results strongly suggest that the outsourcing decision is a means of
compensating for resource deficiencies. In contrast, neither cost considerations
nor the firm’s financial performance were found to significantly affect the
outsourcing decision.
Outsourcing varies in terms of the degree of perceived client conflict in
contracting relations, which is a result that can be explained with agency theory
[36]. In applying agency theory to outsourcing, information asymmetry arises
between the user and the supplier because of the supplier’s expertise and the
inability of the user to effectively monitor and control the project. Only recently
have studies examined post-contract management as well as the middle and
mature stages of the outsourcing lifecycle. For example, Lander et al. [55] and
Lee [56] addressed trust and knowledge sharing issues, respectively.
12
Meanwhile, Miranda and Kavan [73] proposed a theoretical model of the
governance requirements that are needed at different stages of the outsourcing
model. Specifically, in contrast to the market/hierarchy decision that is associated
with the initial decision to outsource (known as a promissory contract), they
suggest that the governance of the outsourcing contract relies on the
psychological contract and social capital, which have the aim of facilitating
cooperation between the contracting parties (see Figure 4). This perspective
provides a holistic understanding of “(1) when governance occurs; (2) what
governance choices are available; and (3) what the consequences of governance
choices are.” Miranda and Kavan used three theories (transactions costs theory,
the embeddedness and knowledge-based perspectives) to model the processes
and outcomes that are involved in the governance of IS outsourcing. They
propose that an IS-specific theory of outsourcing is needed because of specific
factors that “constrain and redefine the governance options available and the
effects of governance choices.” Factors such as the mobility of time and space in
today’s dynamic global environment may weaken the feasibility of the
embeddedness perspective.
Theories on
contracting
Psychological contract
Social capital
Cooperation
Market prices Market
(Transactions costs)
Prior
Research
Conflict (Agency) Reciprocity
In-house Contracting Management
Stages of outsourcing lifecycle
Figure 4. Outsourcing- Areas of prior research
13
Accounting perspective
Accounting researchers have made concerted efforts to understand the
effectiveness of different control and coordination mechanisms in managing the
post-contracting stages of outsourcing [75, 106, 78, 95, 11, 47, 96, 29]. Van der
Meer-Kooistra and Vosselman [107] identified three different patterns of inter-firm
relationship controls based on markets (arms length transactions), bureaucracies
(formal rules and procedures), and trust (common values and maintaining
reputations). Chalos and O’Connor [14] found that socialization and
communication mechanisms were critical for knowledge and technology transfer
within U.S.-Sino alliances. These two mechanisms are closely associated with
bureaucratic and trust control patterns, respectively. Similarly, the notion of clan
control has been proposed as important for cooperative coordination: for example,
the use of boundary spanning workgroups, which cut across traditional business
functions [81].
While the focus on the types of control mechanisms that are used and are
effective in managing the outsourcing relationship has not been limited to the
accounting domain, recent research has focused on the specific role of
accounting in providing (relational) signals about the behavior of each contracting
party [107]. Accounting measures provide formal information about each party’s
actions that helps to keep each party honest, and in so doing preserves the
strength of relational ties between the contracting parties. The objective is for the
parties to the relationship to openly share their knowledge and information.
Another stream of accounting research has examined the different contexts
in which certain types of networks are formed, and then uses this as a basis for
explaining the determinants of the usefulness of different types of controls [34].
For example, the type of network determines the amount and type of information
that each party has about each other that contributes to the common
understanding and stability of the relationship.
Researchers could compare cooperative relationships in a dyadic format
with those in a network format, where there are simultaneous connections to other
14
business entities, such as the firm’s customers. For example, a bank may retain
its fiduciary responsibilities to customers even though they access an online
banking site that is operated entirely by the bank’s outsourcing agent. This type of
relationship is likely to impose greater demands on the supply of information
concerning the delivery on an outsourcing contract, which in turn, may drive the
types of accounting controls required to maintain the outsourcing relationship [4].
MANAGING THE BENEFITS OF INFORMATION SYSTEMS
As mentioned earlier, an accounting measurement system provides
information to influence decisions through mechanisms like chargeback and to
facilitate decisions such as allowing managers to compute the costs and benefits
of outsourcing. The provision of measures to facilitate and influence decisions
extends to other areas of IS practice: decision support (at the management level)
and process improvement (at the operational level). The planning and
development of these types of IS have benefited from the literature on IS success,
and particularly the models proposed by Delone and McLean [18] and Seddon
[97] , which have been widely accepted by academics and practitioners.
DECISION SUPPORT SYSTEMS (MANAGEMENT PROCESS IMPROVEMENT)
Decision support systems (DSS) are “computer based information systems
(CBIS) that support one or more phases of the decision making process from
intelligence to design, choice and implementation” [98].3 Decision support can be
conceptualized to include three parts of a knowledge management system: problem
finding, problem solving, and knowledge base development. IT can support decision
making by collecting, manipulating, and disseminating data and information. Better
decisions, defined as being more accurate and timely, may result from the use of
decision support technologies to the extent that the information on which they are
3 A broad range of technologies that aid decision makers in organizations has evolved
over the past three decades, such as executive information systems, data warehouse
systems, online analytical processing systems, artificial intelligence systems, knowledgebased
systems, data mining, customer relationship management and group support
systems [26].
15
based is accurate, complete, flexible, relevant, simple, verifiable, accessible, secure,
reliable, timely, and economical [102].
The integration of such decision support technologies in the form of a
knowledge management system is rare. The need to cross hierarchical and crossfunctional
boundaries makes it difficult to create appropriate organizational
incentive and support mechanisms. It also encourages resistance due to political
factors, such as protecting one’s turf. While a firm’s capability is a necessary
condition for implementing DSS, it is far from sufficient for DSS success. DSS
implementation depends on the voluntary commitment of personnel [2]. A better
understanding of DSS is critical to improving the design, implementation, and
operational effectiveness of these systems.
A DSS is useful for problem finding, problem solving, and knowledge
development (see Figure 5). Problem finding has been considered by scholars in
various disciplines using terms such as executive information systems (digital
dashboards for top management), business/competitive intelligence, and
environmental scanning, but the timely and accurate identification of management
problems is somewhat fuzzy. Meanwhile, knowledge development has only
recently emerged as an important and highly promising research area. One of the
reasons for this situation is that, typically, IS were traditionally viewed as static
systems that help routine decision making, and not as dynamic learning systems
that can help in organizational learning.
Research into the design of DSS has focused on problem solving rather
than problem finding because it is more tractable and amenable to technology
support in a systematic way. Decision support tools are often defined according to
an understanding of the underlying demands of the task. In modeling the DSS
framework, researchers have commonly examined the relationship between
decision support and outcomes. Researchers are beginning to examine the
complete set of linkages from decision aid, through decision process, to decisionmaking
effectiveness. For example, Todd and Benbasat [105] present a model in
which DSS performance is dependent on DSS strategy, which in turn depends on
16
the interaction of several factors: DSS capabilities, task, perceived effort,
expenditure, and perceived accuracy. Incentives moderate each of the linkages
before DSS performance and DSS strategy.
Problem solving
Benefits of DSS:
1. Greater information about each alternative means
greater accuracy in decision making
2. Audit trail means greater accountability, thus
greater effort in decision making
Amenability to IS support: Emphasis on design,
evaluation, and choice of alternative solutions to a
specified problem/task
Antecedents: Incentives, team support, individual
characteristics (effort and capability)
Uncertainty
Problem finding
Benefits of DSS:
1. Timely feedback for faster decisions
2. Greater access to monitoring (audit
trail)
Amenability to IS support: Depends on
an adequate knowledge base
Antecedents: Incentives, individual
characteristics (effort and capability)
Organizational knowledge base
Benefits of DSS:
Lower knowledge dissemination costs
- Greater search power
- Greater opportunity generation
Examples:
- Knowledge maps—IS can help convert
implicit knowledge of employees into
explicit knowledge for the firm.
- History of prior decisions—specific
decisions routinized and/or past mistakes
reviewed
- Learning from post contracting
investment audit.
Amenability to IS support: Depends on
voluntary sharing of information of
knowledge workers/experts
Antecedents: Incentives, alienability of
individual knowledge restricts incentive to
share
Figure 5. Decision Support System (DSS) Typology
17
Most studies of DSS implementation have identified key factors that enable
or inhibit successful adoption. Kwon and Zmud [53] found that resistance to DSS
is related to poor IS interface design, the proposed system’s functionality (being
inferior to the current system or less than expected), and user lack of aptitude and
motivation to learn new skills and roles. Similarly, Setzekorn et al. [98] reviewed
41 cases of IT implementation and identified six areas of resistance (including,
user, task, development process, system and organizational resistance) and
propose a conceptual model of the relationship between implementation
resistance and success. Compared to other IS, the support of top management
(who are the primary beneficiaries of DSS) appears to be a less critical resistance
factor, while user involvement and change process competency were more critical
to successful DSS implementation.
Accounting perspective
Accounting researchers are interested in quantifying the benefits of the use
of a particular technology. They have thoroughly examined the costs and benefits
of particular DSS such as activity-based costing and multiple performance
measurement systems (e.g. balanced scorecard). Table 1 presents a dichotomy
of the direct and indirect benefits of DSS.
Table 1. Benefits of Decision Support Systems (DSS)
Quantifying the benefits of DSS Examples of DSS examined
in the accounting literature
Direct 1. Decision support system
a. Problem finding—timeliness (e.g. ABC, Balanced
scorecard).
b. Problem solving—accuracy, informated (e.g.
greater information processing capability, more
accurate knowledge maps (ABC, Balanced scorecard).
c. Knowledge management—Reduce knowledge
transfer costs (e.g. leveraging span of control through
more efficient organizational structure)
1. Strategic cost
measurement systems (e.g.
activity based costing and
total quality management)
2. Strategic performance
measurement systems (e.g.
balanced scorecard).
Indirect Learning (competitive advantage)
Transparent culture (innovative capability)
Morale (decreased turnover)
18
Most present value analyses of IT attempt to quantify the indirect benefits
of learning, the development of a culture that fits the company’s competitive
strategy, increasing the quality of products, services, and processes, and high
staff morale. These indirect benefits are likely to translate into more tangible
returns in the future as they help the firm to develop a competitive advantage and
innovative capability (e.g., allow faster responses to changes in the market). The
challenge for management is to quantify these future benefits in terms of their
links to the present indirect benefits of IT investment.
Accounting research can enrich our understanding of the decision
aid/benefits of IS in two specific ways: (i) furthering the work undertaken on
accounting-related judgment and decision-making performance, and (ii) providing
feedback on the benefits that can be gained from having a particular decision
support system in place. First, for a DSS to be effective we need to understand
the limitations and biases that are associated with the storage, search for, and
retrieval of information in the DSS, and the use of that information in decision
making. Accounting researchers have spent considerable time in understanding
how the various characteristics of decision makers and the tasks that they perform
affect accounting-related judgment performance. Differences in the judgment
performance of managers (e.g., accuracy) have been attributed to cognitive
mechanisms such as framing [57] and characteristics of the manager such as
experience [45] and education [6], and the manager's task, such as the type of
feedback [39, 25] incentives [24, 51], time pressure, and the quantity and
dimensionality of information available [100].
More recent studies have examined the cognitive biases that are
associated with the use of multiple performance measurement information in DSS
systems such as the balanced scorecard. This research has found that managers
are prone to several biases such as the tendency to put more weight on common
than unique measures [9, 58] and outcome effects [28]. Aware of these potential
biases, accounting researchers have examined the effectiveness of de-biasing
mechanisms such as accountability, experience, and counterexplanation [50]. IS
19
researchers can expand the focus of this research to include non-accounting
information and to develop a clearer picture of the characteristics of an effective
DSS. The aim is to improve our understanding of the biases in the information
that is provided by a DSS, and to examine the mechanisms that may be effective
in reducing such biases, which will ultimately improve to effectiveness of the DSS.
Second, there is limited work on understanding how a firm knows when it
gains from the use of a DSS. For example, does the use of a DSS speed up the
decision process or produce a superior decision outcome? The large commitment
of resources that is typically required for the development of an organizational
knowledge base requires measures to help management assess the benefits of
an IS. An effective IS is one that grows with increasing volitional use and the
| 