The Interaction of Contract, Control, and Relational Norms as Governance Mechanisms in IS Outsourcing Relationships

Table of Contents

List of Figures

List of Tables

List of Abbreviations

List of Keywords

Abstract

1 Introduction
1.1 Motivation and Objectives
1.2 Structure of the Thesis

2 Theoretical Foundations
2.1 IS Outsourcing
2.2 IS Outsourcing Relationships
2.2.1 IS Outsourcing Relationships as Inter-Organizational Exchanges
2.2.2 Underlying Challenges of IS Outsourcing Relationships
2.2.3 IS Outsourcing Relationship Governance Modes
2.2.3.1 Contractual Governance
2.2.3.2 Relational Governance
2.3 Control
2.4 IS Outsourcing Success

3 Research Model and Propositions
3.1 Research Approach
3.2 Research Model
3.2.1 Research Questions, Objectives, and Approaches
3.2.2 Towards an IS Outsourcing Relationship Framework
3.2.3 Description of the Research Model
3.2.4 Summary of Propositions

4 Case Study
4.1 Methodology
4.2 Setting
4.3 Results
4.4 Analysis

5 Discussion
5.1 Interpretation of Findings and Implications for Theory and Practice
5.2 Limitations
5.3 Further Research

6 Conclusion

Appendix A – Case Study Protocol

References

List of Figures

Figure 1 - Outsourcing phases (Dibbern et al. 2004)

Figure 2 - Outsourcing stages (Alborz et al. 2003)

Figure 3 - Outsourcing relationships: interaction approach (Kern & Willcocks 2002)

Figure 4 - Outsourcing relationship model (Goles & Chin 2002)

Figure 5 - Control framework (Flamholtz 1983)

Figure 6 - Core control system (Flamholtz 1983)

Figure 7 - Portfolios of control mechanisms (Rustagi et al. 2008)

Figure 8 - Research model

List of Tables

Table 1 - Overview of IS outsourcing relationship literature

Table 2 - Challenges derived from TCE

Table 3 - Challenges derived from agency theory

Table 4 - Relationship challenges and mitigation strategies

Table 5 - Contractual elements in IS outsourcing (Chen & Bharadwaj 2009)

Table 6 - Contractual elements in IS outsourcing (Goo 2010)

Table 7 - Relationship constructs in research literature

Table 8 - Control modes and mechanisms

Table 9 - Units of analysis in control literature

Table 10 - Research approach (based on Saunders et al. 2007)

Table 11 - Corresponding relationship constructs

Table 12 - Relationship constructs and related authors

Table 13 - Research constructs and definitions

Table 14 - Overview of case study key facts

Table 15 - Summary of case study results

Table 16 - Overview of empirical support for propositions

Table 17 - Sources of interview questions

List of Abbreviations

Abbildung in dieser Leseprobe nicht enthalten

List of Keywords

Agency theory Theory that describes challenges prevalent to relationships characterized by information asymmetry, uncertainty and risk.

Contractual governance Effect on relationships caused by written forms of terms and expectations: Rules of Conditions.

Control Attempts to influence the behavior of individuals or groups to ensure that they are working towards organizational goals. Comprises formal and informal control modes.

Current mode of operation Here: current design of IT infrastructure in place.

Future mode of operation Here: to-be design of future IT infrastructure.

IS outsourcing Contracting-out of IS functions to an external service provider.

IS outsourcing success Achievement of predefined organization-specific IS outsourcing goals.

Relational exchange theory Theory that considers both contractual and relational dimensions of relationships.

Relational governance Effect on relationships caused by relational norms and behavioral expectations: Rules of Behavior.

Relationship Multi-dimensional phenomenon that embodies the ongoing exchange between two parties. Can be represented by a comprehensive but non-exhaustive set of relationship characteristics.

Relationship characteristics Relationships can be represented by six dimensions or characteristics: trust, flexibility, interdependence, commitment, cultural compatibility, and consensus.

Relationship governance Multi-dimensional phenomenon, represented by a comprehensive but non-exhaustive list of governance modes: control, contractual governance, and relational governance. Relationship governance is an abstract term that embodies the joint effect of single governance modes on relationship characteristics.

Service level agreement Formally defined and agreed level of service.

State of relationship Joint and specific occurrence of relationship characteristics.

Transaction cost theory Theory to explain the existence of firms in relation to market-based exchanges. Suggests that different organizational forms exist due to costs caused by transactions. Describes challenges prevalent to inter-organizational exchanges.

Abstract

The dynamics of the relationship between service recipient and service provider in IS outsourcing relationships recently gained increased attention as relationships are believed to have a considerable influence on IS outsourcing success. This thesis adds to this growing field of interest by developing an IS outsourcing relationship framework in the form of a process model. Three rather disjointed areas of research, namely contractual governance, relational norms, and control, have been set in a common context by interrelating them as the three main governance modes that jointly influence the relationship. One in-depth case study has been conducted in order to provide first empirical evidence and to gain deeper insights into the dynamics of relationship governance. The proposed model could be confirmed in general, revealing the following insights: first, contractual and relational governance modes determine the rules that govern the relationship while control is used to execute and enforce specified rules. All three jointly influence the state of the relationship. Second, relational norms have only been observed at an individual level and not at an organizational one. Third, formal control modes have been used to execute and enforce relational norms. This finding contradicts current control theory. Fourth, while contractual and relational governance are seen as complementary and equally important, relational norms have been left completely unmanaged in the observed organization due to a lack of adequate approaches. These results are discussed in detail to outline opportunities for further research.

1 Introduction

1.1 Motivation and Objectives

This paper attempts to provide deeper insights into a long-discussed yet not satisfyingly explained phenomenon in information systems (IS) outsourcing: the dynamics of the relationship between service recipient and service provider. There is a wide variety of theories in place about what constitutes and characterizes the mutual exchange between organizations. There are, however, hardly any explanations about how these relationships can be systematically influenced. This question is highly relevant to theory and practice.

From an academic point of view, the decision process of IS outsourcing has been thoroughly examined (Dibbern, Goles, Hirschheim & Jayatilaka 2004). Research on the implementation phase, however, has remained fragmentary and has only recently gained increased attention. Although the impact of relationship management on outsourcing success has long been recognized (McFarlan & Nolan 1995), even today a “greater understanding of how to manage IT outsourcing relationships that create and sustain strategic value […] is highly desired” (Goo & Huang 2008, p. 216).

IS practitioners have a strong interest in relationship management as well (Beimborn et al. 2008; Blumenberg 2009). Although failure and dissatisfaction in IS outsourcing projects can often be traced back to challenges during the implementation and operation of outsourcing agreements (Würz & Blankenhorn 2010), existing practitioner literature is regularly limited to the definition and monitoring of performance targets, providing very little advice on how inter-organizational relationships can be developed and maintained.

The purpose of this thesis is therefore to investigate how contract, control, and relational norms interact to influence and govern inter-organizational relationships. Three research questions will be examined:

1. What mechanisms can be used to govern IS outsourcing relationships?

2. How are these mechanisms interrelated?

3. How do organizations employ these mechanisms?

1.2 Structure of the Thesis

The thesis is divided into two parts: one is conceptual and the other is empirical. Chapter 2 starts with a brief outline of the IS outsourcing research context in order to integrate the paper into existing literature and to define the context to which the suggested findings can be applied. Afterwards, an overview of existing theories about inter-organizational relationships in general and IS outsourcing relationships in particular is given, covering organizational, economic, managerial and IS research areas. The underlying challenges of inter-organizational relationships are described from both an economic and a social theory point of view.

Based on the review of theoretical foundations, the research approach and research model are presented in Chapter 3. Prior research is consolidated into a new framework in order to clearly distinguish between relationship characteristics and relationship governance modes. Following the overarching research questions, the research model is developed by describing proposed constructs and respective interconnections. Finally, a set of propositions is derived from the research model to allow for empirical validation.

The suggested research model is empirically evaluated using a case study, which is presented in Chapter 4. After demonstrating the applied methodology, empirical findings are presented and it is analyzed whether the examined propositions can be supported empirically.

In Chapter 5, theoretical and practical implications, as well as the generalizability and limitations of the findings, are discussed. Based on the results of this thesis, possible directions for future research are presented.

Finally, a conclusion is given in Chapter 6.

2 Theoretical Foundations

2.1 IS Outsourcing

IS outsourcing can be defined as “a decision taken by an organization to contract-out or sell the organization's IT assets, people, and/or activities to a third party vendor, who in exchange provides and manages assets and services for monetary returns over an agreed time period” (Kern & Willcocks 2000a, p. 322).

Having a history of more than two decades, IS outsourcing has received a great deal of attention since the seminal article of Loh & Venkatraman (1992). In their remarkable and influential literature review, Dibbern et al. (2004) provide a comprehensive discussion of IS outsourcing definitions, concepts, and theories. A more recent review, identifying today’s research agenda, is given by Chadee & Raman (2009). Focusing on the key concepts that are relevant to this thesis, two perspectives on IS outsourcing will be briefly discussed: the scope or target and the different phases.

Following Lacity & Willcocks (2003), three different layers of IS outsourcing can be distinguished. The first layer comprises IS infrastructure, i.e. networks, hardware provision, and data center facilities. The second layer covers application hosting and management. The top layer, finally, corresponds to business process outsourcing, where entire processes are handed over to the service provider, including physical requirements for the provisioning of the service (Halvey & Melby 2007). In addition, information systems development projects can be outsourced as well. Contrary to the three layers described above, the latter is organized as a project and therefore fundamentally differs from the more operational character of infrastructure, application, or business process outsourcing agreements. As detailed in Chapter 5.2 Limitations, the conceptual model and empirical foundation in this thesis focus on the management of long-term routine operations. Findings are therefore not assumed to be applicable to the outsourcing of projects.

The second perspective on IS outsourcing addresses the chronological sequence of IS outsourcing decisions and implementations. Dibbern et al. (2004) propose a five-stage model, subdivided into two main phases (see Figure 1). Phase 1, referred to as the decision process, encompasses the stages why, what, and which. This phase covers the general decision to outsource; the question of which IS part or function should be outsourced; and finally the procedures and guidelines necessary to decide between different outsourcing alternatives. After the decision to outsource has been made, phase 2, referred to as the implementation phase, becomes relevant. This phase covers the stages of how to outsource and what the desired outcomes of the outsourcing decision are. Relevant to this thesis is the how stage as it covers the building and maintaining of the relationship between service recipient and service provider. The research model and empirical foundation of this thesis are focused on this stage as the relationship during the ongoing operation of the outsourcing agreement is examined. A closer examination of what exactly characterizes this relationship is given below.

Abbildung in dieser Leseprobe nicht enthalten

Figure 1 - Outsourcing phases (Dibbern et al. 2004)

2.2 IS Outsourcing Relationships

2.2.1 IS Outsourcing Relationships as Inter-Organizational Exchanges

As outlined in the previous chapter, Kern & Willcocks (2000a) define IS outsourcing as “a decision taken by an organization to contract-out or sell the organization's IT assets, people, and/or activities to a third party vendor, who in exchange provides and manages assets and services for monetary returns over an agreed time period” (ibid., p. 322). This characterization implies that outsourcing inherently is a form of inter-organizational relationship. Ongoing exchanges between two distinct but interrelated organizations on a contractual basis have been thoroughly investigated in IS and management literature in the past. Important contributions have been made in the fields of organizational theory (Oliver 1990; Ring & van de Ven 1994), marketing (Anderson & Narus 1990; Dwyer, Schurr & Oh 1987), and IS (Bensaou & Venkatraman 1996; Henderson 1990).^[1] Although every theoretical perspective emphasizes certain facets, a relationship can be generally defined as “a long-term commitment, a sense of mutual cooperation, shared risk and benefits, and other qualities consistent with concepts and theories of participatory decision making” (Henderson 1990, p. 8).

The importance of the relationship between service recipient and service provider has been recognized early on in IS outsourcing research (e.g. Klepper 1995; McFarlan & Nolan 1995; Willcocks & Choi 1995). Nevertheless, research in this area has long been restrained by misleading naming (Lacity & Willcocks 1998) and by a lack of conceptual focus leading to a corresponding inconsistent treatment of constructs (Kern & Willcocks 2002; Goles & Chin 2005). As a result, characteristics and determinants of relationships are not always clearly distinguished. These shortcomings in general, and three relationship models in particular, will be discussed below.

First, the understanding and use of the term relationship has been inconsistent (Sargent 2006). Due to the strategic importance of IS, outsourcing arrangements have often been labeled as partnerships (Grover, Cheon & Teng 1996; Klepper 1995; Lee 2001; Lee & Kim 1999, 2003) or strategic alliances (Das & Teng 1998, 2001; McFarlan & Nolan 1995). However, a closer look often reveals that outsourcing agreements rarely satisfy the requirements of a partnership or alliance (Willcocks & Choi 1995). Lacity & Willcocks (1998), for example, found that 45 out of 46 analyzed agreements were fee-for-service contracts and not alliances in the sense of risk sharing and goal alignment. Therefore, the more neutral term outsourcing relationship is used in this thesis, following the definition of Goles & Chin (2002):

“For the purposes of this research an outsourcing relationship is defined as an ongoing, long term linkage between an outsourcing vendor and customer arising from a contractual agreement to provide one or more comprehensive IT activities, processes, or services with the understanding that the benefits attained by each firm are at least in part dependent on the other.” (ibid., p. 227)

Second, IS outsourcing relationships have been analyzed using a variety of theoretical lenses. Dibbern et al. (2004) provide an extensive overview of related IS outsourcing relationship articles from 1988 through 2000. Table 1 gives an overview of articles that investigate IS outsourcing relationships in and of themselves, in contrast to related topics like contract design or relationship governance. A comparison among prior research articles reveals that almost every new effort to build a relationship framework is based on a different theoretical lens. As a consequence thereof, similar named constructs are often used in different ways. Goles & Chin (2005) illustrate the problematic overlap between constructs by comparing the definition of “communication” among five studies. The very same construct is viewed as an element of relationship building (Kern 1997) and contract structuring (Willcocks & Kern 1998), as an attribute of relationship quality (Klepper 1995; Grover et al. 1996), and as a determinant of relationship quality (Lee & Kim 1999). Additionally, a multitude of constructs have been used in different combinations, leading to several attempts of consolidation and categorization (Alborz, Seddon & Scheepers 2003; Goles & Chin 2002, 2005; Lee & Kim 1999). This lack of conceptual focus can at least partly be attributed to an often unsatisfying explanatory value of suggested frameworks. It appears that researchers have yet to find a conclusive approach to the “often messy reality of relational exchange” (Goles & Chin 2005, p. 50).

Abbildung in dieser Leseprobe nicht enthalten

Table 1 - Overview of IS outsourcing relationship literature

Nevertheless, existing research already achieved some very valuable, albeit fragmentary, insights into the nature of outsourcing relationships. Particularly noteworthy for the purposes of this thesis are the contributions of Alborz et al. (2003), Kern & Willcocks (2002), and Goles & Chin (2005) as they are useful in defining the scope and focus of the current study.

Alborz et al. (2003) propose a temporal model comprising three sections: the pre-contract, contract, and post-contract stage (see Figure 2). Every stage is then linked with different factors influencing IS outsourcing relationship efficacy. An insight that can be drawn from this work is that relationships exist from the very beginning and not only after signing the contract. Moreover, they evolve and change over time. A comprehensive relationship model should therefore reflect the dynamics of a relationship or at least clearly state which stage is modeled by the framework. This thesis focuses on the post-contract stage, i.e. the day-to-day operational activities of an outsourcing arrangement.

Abbildung in dieser Leseprobe nicht enthalten

Figure 2 - Outsourcing stages (Alborz et al. 2003)

Abbildung in dieser Leseprobe nicht enthalten

Figure 3 - Outsourcing relationships: interaction approach (Kern & Willcocks 2002)

Kern & Willcocks (2002) draw on the work of the industrial market and purchasing group (IMP group). This group has been particularly influential in the field of buyer-supplier exchange relationships as it was among the first ones to enhance strictly economic relationship models by considering the social embeddedness of relationships (Granovetter 1985; Cunningham & Tynan 1993; Uzzi 1997). Based on this research, Kern & Willcocks (2002) adopt the interaction approach from Hakansson (1982) to provide a more holistic model of outsourcing relationships (see Figure 3). This model is based on four main groups of variables: the environment of the interaction, the atmosphere that characterizes the relationship, the parties and their organizational and individual characteristics, and finally the interaction process, covering service, information, financial, and social exchanges. Kern & Willcocks (2002) found the framework to be too general for a detailed analysis of outsourcing relationships. Nevertheless, it is highly useful in classifying and illustrating the complexity and diversity of relationship factors. It is at least a very valuable macro framework that can be used to position further research. This thesis focuses on the interaction process, i.e. the attributes and influencing factors of relational exchanges.

Abbildung in dieser Leseprobe nicht enthalten

Figure 4 - Outsourcing relationship model (Goles & Chin 2002)

Goles & Chin (2002, 2005) suggest a relationship framework (see Figure 4) that is particularly interesting as it clearly distinguishes and links prerequisites, relationship dimensions, relationship measurements (quality), and relationship outcomes (success). Goles & Chin (2002, 2005) reviewed the existing relationship literature to produce a compilation of relationship factors. The generated list with a total of twenty-three factors was then consolidated into eleven constructs drawing on relational exchange and contract theory. Afterwards, the constructs were grouped into attributes (inherent characteristics or properties) and processes (the means by which the attributes are developed). With the aim of setting the described core relationship dimensions in the context of a more comprehensive view, the vendor and customer capabilities are integrated as input factors. Quality and success are considered to be output factors. The described model will serve as a guiding framework in this thesis as it (a) makes a clear statement on the dimensions representing the relationship itself, instead of investigating peripheral constructs, and (b) distinguishes between attributes as inherent characteristics of a relationship and factors by which these attributes are developed and influenced.

Among the articles in IS outsourcing relationship research listed in Table 1, only Kern (1997), Kern & Willcocks (2000a, 2002), and Goles & Chin (2002, 2005) strive to develop a model of the relationship itself. The unit of analysis in the remaining literature often embraces a construct peripheral to relationship, like quality, efficacy or development. This is somewhat surprising as a common understanding about what attributes and dimensions determine the relationship itself is still to be developed.

To conclude, it can be said that there is a lack of a clear differentiation in IS outsourcing relationship literature between what attributes define a relationship, what determinants influence these attributes, what constructs represent the quality of a relationship, and how a relationship can be managed, developed and governed. The most comprehensive model is the one of Goles & Chin (2002, 2005) as it clearly differentiates between relationship characteristics and determinants.

It is beyond the means of this thesis to solve all discussed shortcomings. Nevertheless, an attempt will be made to integrate existing outsourcing relationship literature in order to build a modified relationship framework. The proposed model does not claim to explain relationships entirely. However, it aims to meet the purposes of this thesis by clearly defining and linking the components necessary to describe the relationship itself and the approaches to govern this relationship.

To this end, an overview of the underlying challenges inherent to relationships will be discussed in the next chapter, followed by an examination of governance modes. Finally, an integrative outsourcing relationship framework will be developed.

2.2.2 Underlying Challenges of IS Outsourcing Relationships

As the main purpose of this thesis is to investigate relationship governance mechanisms, a closer look at the challenges these governance mechanisms should solve is needed. Economic theories, or more specifically transaction cost economics (TCE) and agency theory, have been widely used as a theoretical lens for the analysis of inter-organizational relationships in general (Cannon, Achrol & Gundlach 2000; Claro, Hagelaar & Omta 2003; Ferguson, Paulin & Bergeron 2005; Logan 2000; Rindfleisch & Heide 1997; Sheng, Brown, Nicholson & Poppo 2006; Wathne & Heide 2000, 2004) and IS outsourcing relationships in particular (Aubert, Rivard & Patry 1996, 2003b, 2004; Aubert, Houde, Patry & Rivard 2003a; Gellings 2007a, 2007b). Both theories offer precise and well-known instruments to model the underlying challenges of inter-organizational relationships. Therefore, an overview of challenges is given here, while mechanisms to mitigate these are explored in the subsequent chapters.^[2]

Based on well-known early works (Williamson 1975, 1985) and versions later reviewed and extended (Williamson 1991a, 1991b, 1992, 1993, 1996), TCE has identified a set of governance problems, like safeguarding specific assets, and related governance mechanisms, like markets, hierarchies or hybrids, to manage these problems. TCE is built on the notion that transactions cause costs (Coase 1937), e.g. for drafting and negotiating contracts or monitoring and enforcing agreements. As different governance mechanisms differ in their costs and in their ability to mitigate underlying problems of transactions, the final goal is about finding an optimal – i.e. cost-minimizing – governance structure. In order to solve this problem, Williamson integrates different theoretical perspectives by synthesizing Coase’s (1937) conception of transaction costs, Commons’ (1934) focus on transactions as units of analysis, Simon’s (1957) suggestion of bounded rationality, and Knight’s (1921) argument that individuals behave opportunistically. In sum, transaction cost economics rests on two key attributes of transactions (asset specificity and uncertainty) and two main assumptions of human behavior (bounded rationality and opportunism). Bounded rationality is the assumption that decision makers have constraints on their rationality due to limited information processing and communication ability. In combination with the key attribute uncertainty, this imposes two problems: environmental conditions surrounding a transaction cannot be specified ex ante (environmental uncertainty) and performance is difficult to evaluate ex post (behavioral uncertainty). Environmental uncertainty creates an adaption problem; that is, it leads to difficulties in adapting the agreement to changing environments. The behavioral uncertainty results in a performance evaluation problem; that is, it causes difficulties in deciding whether specified agreements have been fulfilled. Opportunism is the assumption that the actions of decision makers may be guided by their self-interest, which can lead to lying, cheating, or otherwise violating the agreement. As it can be difficult to know beforehand who is trustworthy (Barney 1990), opportunism is especially problematic in combination with specific assets that have limited value outside the relationship. This creates a safeguarding problem to protect the specific asset. These challenges are summarized in Table 2. The described attributes and resulting problems are particularly prevalent in inter-organizational relationships. Asset specificity in IS outsourcing agreements is usually high. While single hardware components and standard software packages are commodities, the entire IS environment is generally tailored to the individual organization. The configuration of the IS environment, the knowledge about the demand and needs, as well as the knowledge of the particular processes of an organization, are highly specific and create a lock-in effect for both service provider and recipient. Environmental uncertainty is typically high as well. Future demand is difficult to forecast in both volume and manner. IS is closely coupled with business needs, which can be highly volatile due to changing factors inside and outside the organization. Furthermore, evolution of technology might have a profound influence on both business needs and IS implementation.

Abbildung in dieser Leseprobe nicht enthalten

Table 2 - Challenges derived from TCE

Behavioral uncertainty is high as well. The service is rarely provisioned on the premises of the recipient and in addition, the provider’s employees are not under the authority of the customer. Therefore, it might be difficult to observe behavior and evaluate performance. In summary, IS outsourcing relationships are highly susceptible to the hazardous assumptions of TCE, which is, in turn, applicable to IS outsourcing contexts (cf. Poppo & Lacity 2006).

The second economic approach relevant to this thesis is agency theory (Arrow 1976, Jensen & Meckling 1976; Ross 1973; Wilson 1968). Following Eisenhardt’s (1989a) excellent review, agency theory is concerned with a situation where one party (the principal) delegates work to another party (the agent). The relationship between the principal and the agent can be modeled by a contract (Jensen & Meckling 1976). Underlying assumptions of human behavior are, similar to TCE, opportunism, bounded rationality, and risk aversion. In addition, two organizational assumptions are made: a partial goal conflict and an information asymmetry between principal and agent. Out of these assumptions, two problems arise: a risk-sharing problem and an agency problem. The risk-sharing problem occurs when principal and agent have different risk preferences and thus prefer different actions. The agency problem occurs when the goals of principal and agent conflict and when it is difficult or expensive for the principal to observe the agent’s behavior. Two appearances are mentioned in the literature for this problem: moral hazard and adverse selection. The former refers to the problem where the agent does not make the agreed effort, i.e. is shirking. The latter refers to the problem where the principal cannot verify the abilities claimed by the agent. An alternative view on the challenges underlying a relationship between a principal and an agent is based on the assumption of information asymmetry. Depending on the type of information that is unavailable to the principal, three different problems can be defined: hidden characteristics, hidden information, and hidden action (cf. Küpper 2008). Hidden characteristics refer to (negative) characteristics of the agent that are covert before the contract is signed. The resulting problem corresponds with adverse selection. Hidden information refers to the moral hazard of the agent to manipulate or only selectively reveal private information for personal advantage. Hidden action is also a form of moral hazard. It refers to the problem of unobservable behavior and therefore possibly undetected opportunistic behavior. The described challenges are summarized in Table 3.

Abbildung in dieser Leseprobe nicht enthalten

Table 3 - Challenges derived from agency theory

The described assumptions and resulting problems are again particularly prevalent in inter-organizational IS outsourcing relationships. As described in the section about TCE, outsourcing services are usually provisioned remotely and behavior is difficult to observe, leading to information asymmetry. The service provider may be tempted to reduce its effort and blame low performance on circumstances outside of his control. Due to the complexity of outsourcing agreements and the often limited experience in conducting outsourcing projects, service recipients suffer from information asymmetry during the vendor selection phase. Furthermore, the outsourcing market is dominated by a small number of global organizations. While these huge companies may have a uniform appearance on the market, the competences of the teams which ultimately execute the outsourcing project may differ. Finally, the objectives of the service recipient and service provider are typically diametrically opposed. While the recipient strives for high service quality and low costs, the service provider is incentivized to invest only a little effort while maximizing payments. Risk preferences may therefore differ considerably between both parties.

Opportunism is a necessary assumption for all problematic outcomes in TCE and agency theory. Wathne & Heide (2000) refine the general construct of opportunism in their insightful article by indicating that actually several forms of opportunism exist. They identify four different types along the dimensions behavior and circumstances. A passive behavior under existing circumstances can lead to evasion or shirking (corresponding to the performance evaluation problem and hidden information form of moral hazard) . A passive behavior under new circumstances can lead to a refusal to adapt (corresponding to the adaption problem). Active behavior under existing circumstances can lead to violation, i.e. engaging in prohibited behavior (corresponding to the hidden action form of moral hazard). The last type, forced renegotiation, is caused by active behavior under new circumstances and refers to the cases where new circumstances are used to enforce concessions.

Both TCE and agency theory share an economic perspective and see (formal) contracts as a way to govern relationships and mitigate presented problems. A good amount of research has thus been undertaken in order to design optimal – i.e. cost minimizing – contracts for various kinds of relationships.^[3]

As discussed, the economically derived challenges of a relationship are supposedly accurate, applicable and relevant in an IS outsourcing context. Likewise, contracts are an appropriate way to mitigate and, to some extent, even solve described problems. Nevertheless, as the discussion about relationship frameworks has shown, inter-organizational transactions are embedded in a social context. Governing mechanisms that are purely based on formal contracts are likely to be insufficient if applied on their own, as are mechanisms that are purely based on social theories, like social exchange theory (Blau 1964; Emerson 1972; Homans 1961).

Thus, some researchers (e.g. Goles & Chin 2002) draw upon relational exchange theory (RET), which is “based on the notion that parties to an exchange are in mutual agreement that the resulting outcomes of the exchange are greater than those that could be attained through other forms of exchange, or from exchange with a different partner” (Goles & Chin 2002, p. 227). In contrast to purely economic and purely social theories, relational exchange theory takes into account the fact that exchanges between two organizations are governed by both contractual and relational dimensions.

Based on the work of Macaulay (1963), Macneil (1974, 1978, 1980) is one of the most influential advocates of relational exchange theory. RET emphasizes the fact that contracts are incomplete as they cannot anticipate or specify every contingency that arises during the life of the contract. TCE and agency theory neither assume nor postulate complete contracts. However, in contrast to economic theories, RET considers relational norms as a way to strengthen the relationship by complementing this incompleteness. As both parties acknowledge the benefits of the mutual exchange, the relationship is valued in and of itself. Relational norms transform the nature of the contract from a contract-in-law (one that is restrained to legally enforceable agreements) to a contract-in-fact (one that includes interactions to preserve and harmonize the relationship) (Goles & Chin 2005).

To conclude, economic theories are helpful and valuable in describing the underlying challenges of IS outsourcing relationships. Nevertheless, (formal) contracts alone are not sufficient to mitigate these challenges due to the social embeddedness of relationships. Thus, both contractual and relational mechanisms to govern relationships are discussed in the following section.

2.2.3 IS Outsourcing Relationship Governance Modes

As outlined in the previous chapter, the understanding of relationships has developed from a purely economic, transaction-based perspective to one of a socially embedded mutual exchange based on relational norms. Together with this change of perspective, the mechanisms used to manage and direct relationships, i.e. relationship governance mechanisms, have adjusted as well.

(IS) governance is often understood in a broad sense (de Haes & van Grembergen 2006). Webb et al. (2006), for example, define IS governance as “the strategic alignment of IT with the business such that maximum business value is achieved through the development and maintenance of effective IT control and accountability, performance management and risk management” (ibid., p. 7). This broad understanding consequently includes elements like strategic alignment, risk management, roles and responsibilities and so forth. For the purposes of this thesis, however, governance refers to “the formal and informal rules of exchange between partners” (Vandaele, Rangarajan, Gemmel & Lievens 2007, p. 237). Following this narrow understanding, the definition of Heide (1994) is used in this thesis:

"Governance is a multidimensional phenomenon, encompassing the initiation, termination and ongoing relationship maintenance between a set of parties. […] Essentially, governance includes elements of establishing and structuring exchange relationships as well as aspects of monitoring and enforcement." (ibid., p. 72)

According to Vandaele et al. (2007), early versions of TCE (Williamson 1979, 1985) focused on single and unilateral transactions. Given this, markets and hierarchies have been appropriate governance forms to mitigate the challenges outlined in the previous chapter. Markets refer to market-based exchanges where transactions are governed by legal and formal terms (David & Han 2004). Hierarchies refer to vertical integration where transactions are internalized by the organization (Rindfleisch & Heide 1997). As the perspective on inter-organizational relationships changed to one of a bilateral, relational exchange, TCE has been further developed to include intermediate, or hybrid, governance modes (Williamson 1991b). In today’s understanding, principles of TCE and social theories are combined, shifting the focus of analysis from the transaction to the relationship itself (Vandaele et al. 2007). Since then, two types of governance strategies have been investigated: economic-based strategies, such as contracts, and relational-based strategies, such as relational norms (Griffith & Myers 2005). A classification into these two categories is prevalent in current research, although different terms are used: hard and soft sides of outsourcing management (Barthélemy 2003), formal contracts and relational/social norms (Cannon et al. 2000; Dahlstrom, McNeilly & Speh 1996), formal and informal management (Heckmann 1999), and formal contracts and psychological contracts (Koh, Ang & Straub 2004; Miranda & Kavan 2005). This thesis adopts the terms contractual governance and relational governance (Chaudhary & Kishore 2010; Ferguson et al. 2005; Vandaele et al. 2007) for two reasons: first, the similar naming emphasizes the complementarity and equality of the two concepts; second, the word governance clearly distinguishes the function as a governing mechanism from underlying principles like social norms.

According to Ferguson et al. (2005), contractual governance can be defined as “the degree to which the formal contract is currently implemented in established service exchanges” (ibid., p. 219) . It refers to formal, explicit and typically written contracts. Contracts are legally binding agreements that detail the roles and obligations of both parties. They can therefore be considered as substitutes for the formal governance mechanisms of hierarchies (Ferguson et al. 2005; Vandaele et al. 2007).

Relational governance can be defined as “a social institution that governs and guides exchange partners on the basis of cooperative norms and collaborative activities” (Poppo, Zhou & Zenger 2008, p. 1197) while relational norms are “the bilateral expectations that exchange partners will act in ways that assist each other during the course of the relationship” (Vandaele et al. 2007, p. 241). Thus, compared with contractual governance, relational governance is rather informal and social.

A good amount of research has been invested into the question of whether contractual and relational governance function as substitutes or complements (Lazzarini, Miller & Zenger 2004; Poppo & Zenger 2002). The substitution view (Adler 2001; Bernheim & Whinston 1998; Bradach & Eccles 1989; Dyer & Singh 1998; Granovetter 1985; Gulati 1995; Uzzi 1997) argues that extensive formal agreements might undermine trust between both parties (Ghoshal & Moran 1996; Macaulay 1963). Other authors (Hill 1990; Uzzi 1997) suggest that relational governance might act as a safeguard in complex and uncertain environments that are less costly than extensive contracts. The complementary view, on the contrary, considers the joint use of contractual and relational governance mechanisms to be more effective than each one on its own. The main argument is based on the notion that incomplete contracts facilitate the self-enforcement of informal agreements as incentives or punishments reduce gains from short-term opportunism while increasing the value of respecting informal agreements (Baker, Gibbons & Murphy 1994; Klein 1996; Pearce & Stacchetti 1998; Poppo & Zenger 2002; Schmidt & Schnitzer 1995). Recent conceptual and empirical research provide strong support for the complementary view (Lazzarini et al. 2004; Poppo & Zenger 2002).

This thesis therefore follows the notion that contractual and relational governance function as complements. More specifically, the interrelation between both governance modes and control as a shared implementation mechanism is investigated. Both modes are therefore examined in more detail in the subsequent chapters while control is examined thereafter.

2.2.3.1 Contractual Governance

As one of the main objectives of transaction cost economics and agency theory is the design of optimal contracts, formal agreements are a natural choice for governing relationships. Thus, contractual governance has often been the focus of research about IS outsourcing relationships (Aubert et al. 2003a, 2003b; Chen and Bharadwaj 2009; Gellings 2007a, 2007b; Goo 2009, 2010; Goo & Huang 2008; Goo, Huang & Hart 2008; Goo, Kishore, Rao & Nam 2009).

Contracts mitigate the underlying challenges of relationships outlined in Chapter 2.2.2 Underlying Challenges of IS Outsourcing Relationships in several ways. Opportunism and information asymmetry can be countered in two principal ways. First, the reasons for opportunistic behavior can be reduced by aligning the goals of the service provider and service recipient. Rewards can be used to incentivize beneficial behavior while penalties reduce the payoff for unfavorable actions (Gellings 2007a). Additionally, long-term contracts can protect against short-term profit maximization (Aubert et al. 2003b). Wathne & Heide (2000) give an interesting overview of contractual and non-contractual mechanisms to avoid different types of opportunism. Second, creating transparency inherently reduces information asymmetry and can also reveal opportunistic behavior, making it less likely. Transparency can be achieved by clearly defined performance standards, ongoing monitoring, and external benchmarks. Different risk preferences can be aligned through risk and profit sharing or by using incentives. Contracts can be an effective instrument to implement these mechanisms. Nevertheless, outcome-based measures are limited if behavior cannot be separated from non-influenceable environmental conditions. Furthermore, contracts have particular difficulties in mitigating uncertainty. Due to bounded rationality, it is not possible to foresee every contingency in dynamic environments and long-term agreements like IS outsourcing. Uncertainty can be partly compensated for through specification or by defining contingency clauses that define processes to cope with change. However, the limited use of contracts regarding these points is one of the reasons for the increasing interest in relational governance. Table 4 provides an overview of relevant challenges and contract-based strategies to mitigate them.

Abbildung in dieser Leseprobe nicht enthalten

Table 4 - Relationship challenges and mitigation strategies

In IS outsourcing contract research, two articles provide a notably comprehensive view of contract content and structure. Chen & Bharadwaj (2009) derived a contract structure from 112 IS outsourcing contracts. They identified 16 major clauses among the four dimensions monitoring, dispute resolution, property rights protection, and contingency provisions (see Table 5). Their work allows a valuable insight into how outsourcing partners structure contractual governance provisions in practice.

Abbildung in dieser Leseprobe nicht enthalten

Table 5 - Contractual elements in IS outsourcing (Chen & Bharadwaj 2009)

Another perspective is given by Goo (2010), whose work is based on contract, control, and relational exchange theory. Goo (2010) theoretically derives and empirically validates 11 contractual elements in three categories using grounded theory (Strauss & Corbin 1990) and quantitative analysis. The three categories, or characteristics as they are called, reflect common underlying themes in governing outsourcing relationships. Foundation characteristics define performance standards, common objectives, and roles and responsibilities of the parties. Change characteristics outline processes for adapting the contract to future needs and changing objectives. Governance characteristics define management mechanisms such as communication protocols, reporting policies, and administrative procedures. An overview and definition of contractual elements and characteristics is given in Table 6.

The work of Chen & Bharadwaj (2009) might be somewhat more comprehensive compared to Goo (2010), comprising practically relevant clauses like property rights, staffing, or training. Nevertheless, the framework of Goo (2010) is chosen to define the contractual governance dimensions in this thesis as it is a conceptually sound approach based on the theories that also underlie this thesis, i.e. contract, control, and relational exchange theory.

Nevertheless, it is important to state that it is not the contract itself that serves as a governance mechanism. It is the application, not the existence, of contractual clauses that governs the relationships. Contractual governance is therefore not measured using the specificity or completeness of the contract. Instead, what is relevant is the degree of implementation of, or reliance on, the contract (Ferguson et al. 2005).

Abbildung in dieser Leseprobe nicht enthalten

Table 6 - Contractual elements in IS outsourcing (Goo 2010)

2.2.3.2 Relational Governance

Due to the social character of inter-organizational relationships, relational norms as governance mechanism have received broad attention in organizational research in general (Claro et al. 2003; Greenberg, Greenberg & Lederer 2008; Joshi & Stump 1999a, 1999b; Larson 1992; Noordewier, John & Nevin 1990; Poppo et al. 2008; Sheng et al. 2006; Wathne & Heide 2004) and IS outsourcing research in particular (Beimborn et al. 2008; Blumenberg 2009; Blumenberg, Beimborn & Koenig 2008; Lee, Huynh & Hirschheim 2008; Zaheer & Venkatraman 1995).

In contrast to contractual governance, relational norms are endogenous in relational governance (Joshi & Stump 1999a). Behavior is not controlled through incentives, but regulated through moral control (Larson 1992). As Joshi & Stump (1999a) remarkably point out, relational norms evolve from description (“this is the way we do things in the relationship”) to expectation (“this is the way things should be done in the relationship”). Thus, relational norms become a moral control mechanism that promotes behavior that is beneficial for the relationship (e.g. cooperation) and inhibits adverse behavior (e.g. opportunism). Opportunism can then be hindered either by social pressure in case of exposure or by goal alignment through socialization processes (Wathne & Heide 2000). The social context itself then functions as a protection and a governance mechanism to mitigate the challenges underlying the relationship (Macneil 1980; Wathne & Heide 2000, 2004; Joshi & Stump 1999a). Nevertheless, even after relational norms are established, they require continuous maintenance and development (Dwyer et al. 1987; Larson 1992).

A review of existing relational governance literature reveals that many different constructs exist to represent relational norms, relational governance, relationships or relationship quality. While trust and commitment are often used elements, most articles use neither congruent sets nor even define the constructs in a homogenous way. RET and thus the constructs defined by Macneil (1980) are a common base, but they are only selectively chosen among different articles. Table 7 gives an overview of reviewed articles and used constructs.

Abbildung in dieser Leseprobe nicht enthalten

Table 7 - Relationship constructs in research literature

Abbildung in dieser Leseprobe nicht enthalten

Table 7 - Relationship constructs in research literature (continued)

Abbildung in dieser Leseprobe nicht enthalten

Table 7 - Relationship constructs in research literature (continued)

The diversity of constructs shows that no common understanding about relational norms exists. The reviewed articles use a mix of characteristics (e.g. trust), activities (e.g. information sharing), and qualities (e.g. communication quality). The perception of linkages between and antecedents of different constructs vary as well. Greenberg et al. (2008), for example, suggest that trust is a determinant for TCE characteristics like asset specificity and uncertainty, while Poppo et al. (2008) found the relationship to be vice versa. However, due to the complexity of social realities, these findings are not necessarily contradicting. As Noordewier et al. (1990) point out, relational governance “does not exist in a context-free vacuum. It must be operationalized in the context of a specific exchange.” (ibid., p. 84)

For the sake of a more consistent and uniform understanding of used constructs in IS outsourcing relationship research, an integrative IS outsourcing relationship framework is developed in this thesis, bringing together the examined components of relationship characteristics as well as contractual and relational governance. In order to do this, a third relationship governance mechanism, control, is examined in the following chapter.

2.3 Control

“Control theory attempts to explain how one person or group in an organization can ensure that another person or group works toward and attains a set of organizational goals.” (Kirsch 1996, p. 1)

Control has long been of interest to researchers and practitioners alike, resulting in numerous definitions, theoretical foundations, and implementation designs. Two often-used perspectives on control are organizational and economic theory, especially agency theory (Eisenhardt 1985). The underlying theme to the organizational control approach is the question of which mechanisms can be used to manage an organization so that it moves towards its goals (Ouchi 1979). Agency theory, on the other hand, provides a theoretical foundation to describe and model the relationship and its inherent challenges between two contracting parties with divergent objectives (Ross 1973; see Chapter 2.2.2). Both approaches are concerned with how to influence the actions of individuals or groups and therefore jointly provide the foundation for a behavioral understanding of control (Eisenhardt 1985). Following this view, control can be defined as an attempt to influence the behavior of individuals or groups to ensure that they are working towards organizational goals (cf. Jaworski 1988; Kirsch 1996).

Abbildung in dieser Leseprobe nicht enthalten

Figure 5 - Control framework (Flamholtz 1983)

Flamholtz (1983) suggests a meta-framework that can be used to elaborate different dimensions of a control system and to contrast the behavioral understanding with other perspectives on control. In Figure 5, the control system is represented by three concentric circles, surrounded by the organizational environment. The innermost circle, the core control system, consists of an integrated structure based on four basic organizational processes: planning, operations, measurement, and evaluation-reward (see Figure 6). This structure represents the “classic” control process where the results of operations, i.e. the day-to-day organizational activities, are measured against a predefined set of goals and performance standards. If performance deviates, corrective actions can be taken by either adapting operations or goals. Additionally, rewards can be granted for motivational reasons. Flamholtz (1983) argues for a broader understanding and suggests that a control system is not only determined by the outcome-oriented mechanisms just described. Additionally, organizational structure represents an independent component of a control system. The structural dimensions of an organization – like the degree of centralization, specialization or integration – influence the behavior and the decision-making process of individuals and groups (Flamholtz 1983; Otley & Berry 1980; Perrow 1977). The third component of a control system is the organizational culture, which is understood as shared beliefs, values and social norms that influence and guide the actions of organizational members (Flamholtz 1983; Ouchi 1979). Flamholtz’ (1983) article is a very valuable contribution as its ideas serve as a foundation for several concepts and constructs developed by successive researchers referred to in this thesis. First of all, its framework can be used to distinguish between two common perspectives on control theory. The core control system (Figure 6) corresponds to the cybernetic understanding of control, emphasizing the process of standard setting, monitoring, evaluating, and taking corrective actions if desired performance is not achieved (Flamholtz 1983; Green & Welsh 1988; Merchant 1988).

Abbildung in dieser Leseprobe nicht enthalten

Figure 6 - Core control system (Flamholtz 1983)

This thesis adopts the broader perspective of a behavioral understanding as it integrates the structural, cultural, and environmental dimensions of a control system as well (Choudhury & Sabherwal 2003; Eisenhardt 1985; Jaworski 1988; Kirsch 1996; Ouchi 1979). Two main implications arise from this: first, the emphasis of control is on influencing behavior, not on evaluating performance standards; second, control is not limited to outcome-oriented methods, but also comprises behavior-oriented or culture-based mechanisms, as described later on.

A good amount of research has been done to investigate the determinants for the design of a control system. A common framework for explaining the choice of control governance structures is transaction cost economics. As outlined in Chapter 2.2.2 Underlying Challenges of IS Outsourcing Relationships, governance structures, which can either be based on markets, hierarchies, or a hybrid form, are determined by the transaction characteristics uncertainty and asset specificity (Williamson 1979, 1991a). These characteristics can be applied to an intra-firm, inter-firm or outsourcing context (Birnbirg 1988; Geringer & Hebert 1989; Jap & Ganesan 2000; Speklé 2001; van der Meer-Kooistra & Vosselman 2000). However, TCE-based frameworks have been criticized for insufficiently taking into account the social context of a transaction. Broader contingency frameworks have thus been developed to integrate contextual factors like attitudes or personal relationships (Harmanciouglu 2009; Kumar & Seth 1998; Langfield-Smith & Smith 2003). Nevertheless, as this thesis does not investigate the determinants of control system design, a closer examination of mentioned theories is not required.

Control is ultimately exercised via control mechanisms that can be described as measures taken to regulate behavior. Based on the work of Ouchi (1979) and Eisenhardt (1985), Jaworski (1988) introduced the nowadays widely acknowledged distinction between formal and informal modes as a way to classify control mechanisms. Formal controls are “written, management-initiated mechanisms that influence the probability that employees or groups will behave in ways that support the stated […] objectives” (ibid., p. 26). Informal controls can be defined as “unwritten, typically worker-initiated mechanisms that influence the behavior of individuals or groups” (ibid., p. 26). Examples for formal control include performance requirements or project plans while self-set project milestones or dinner meetings are typical informal control mechanisms.

The concept of control modes was advanced in Kirsch’s (1996) seminal paper about control in software development projects. Kirsch (1996) consolidated prior research into a framework with two formal and two informal control modes: outcome and behavior control as formal modes and clan and self control as informal modes. In outcome control, the controller focuses on the outputs of a process while it is not of interest how this output is achieved. Hence, rewards are granted if goals are met, assuming that the controllee can be made fully accountable for process results. As a prerequisite, desired outcomes must be known and measurable. Behavior control, however, focuses on the process itself. The controller influences and rewards the behavior of the controllee by specifying rules and procedures to follow. Thus, the controllee is made responsible for compliance, but not for outcome. To exercise this mechanism, the behavior has to be specifiable and observable. In clan control, on the contrary, neither outcome nor behavior can be specified. A clan, i.e. a group of individuals, defines and enforces acceptable behavior by sharing a common set of beliefs, values, and social norms. In cases where behavior is not observable and outcome is not measurable, clan control can at least maintain the integrity and adherence to shared norms of group members. In self control, task goals and procedures are set, monitored and evaluated by the controllee himself and can therefore be either outcome or behavior oriented.

Choudhury & Sabherwal (2003) further examined the implementation of control modes. They suggested that two types of mechanisms are necessary to implement formal modes: specifying and evaluating mechanisms. Applied to an outsourcing context, performance standards like service level agreements specify desired outputs in outcome control and need complementary monitoring and evaluating mechanisms. Detailed job descriptions specify desired behaviors while client personal on vendor premises are a way to evaluate compliance to these descriptions. Clan control, on the other hand, is implemented by aligning controller’s and controllee’s preferences, values and beliefs. A typical mechanism is socialization through regular joint meetings. Self control can, due to its nature, not be implemented systematically. Nevertheless, the controllee can be encouraged to implement measures on his own by valuing his intrinsic motivation and rewarding his ability to self manage.

Abbildung in dieser Leseprobe nicht enthalten

Table 8 - Control modes and mechanisms

Abbildung in dieser Leseprobe nicht enthalten

Figure 7 - Portfolios of control mechanisms (Rustagi et al. 2008)

Due to the multifaceted nature of day-to-day organizational activities, control mechanisms are rarely implemented in isolation. Rather, several mechanisms and modes are combined to achieve the desired effects. These combinations, referred to as portfolios of control, have been subject to extended research (Choudhury & Sabherwal 2003; Daityari et al. 2008; Harmancioglu 2009; Jaworski 1988; Jaworski, Stathakopoulos & Krishnan 1993; Kim 1984; Kirsch 1997; Rustagi, King & Kirsch 2008). Thus, within a portfolio of control, each control mode is implemented through several control mechanisms (see Figure 7). Choudhury & Sabherwal (2003) point out that some control mechanisms can support more than one control mode. For example, a meeting can be used to evaluate outputs in outcome control, to specify rules in behavior control, to socialize in clan control or to encourage self control. Another subject to research has been the process of constructing a portfolio of control. According to Kirsch (1997) this process comprises three steps, each one being influenced by task characteristics, role expectations, and knowledge and skills. Primarily, controllers judge the appropriateness of pre-existing formal control mechanisms. If pre-existing formal mechanisms are not sufficient, new mechanisms of formal control are defined. In most cases, formal controls are then supported and complemented with informal control mechanisms. Choudhury & Sabherwal (2003) and Daityari et al. (2008) further investigated the dynamics in the construction of a control portfolio in an outsourcing context. Both works add a longitudinal perspective to Kirsch’s (1997) framework. It was found that initial portfolios were mainly based on existing formal mechanisms. As time evolved, different encounters, like negative experiences or malperformance, triggered a re-evaluation and adjustment of control portfolios. Relevant influencing factors for the reassessment have been the client’s perception of vendor’s knowledge, role expectations, difficulties in monitoring behavior and prior vendor performance (Choudhury & Sabherwal 2003), respective partnership quality, relationship encounters, and learning curve benefits (Daityari et al. 2008).

Control theory reviewed so far has been independent of the organizational context and can be applied to intra-firm and inter-firm environments. However, of particular interest to this thesis is whether the discussed control theory is applicable to an outsourcing context, and if so, to what extent.

Control in inter-firm relationships has long been of interest to researchers. Pioneering work has been done at the organizational level, focusing on cooperation, joint ventures, and strategic alliances (Birnbirg 1998; Das & Teng 1998, 2001; Geringer & Hebert 1989; Jap & Ganesan 2000; Kumar & Seth 1998; van der Meer-Kooistra & Vosselman 2000). These works have shown the applicability of previously reviewed control theory in an inter-firm context and have shed some light on the particularities of these relationships. Another research stream has investigated control in outsourcing relationships outside of an IS context (Daityari et al. 2008; Harmancioglu 2009; Langfield-Smith & Smith 2003; Linder & Sawyer 2003). The explanatory value of reviewed control theory has been shown in this context as well. Regarding control theory in an IS outsourcing context, three broad streams can be distinguished. Practitioner literature (Gómez, Junker & Odebrecht 2009; Kendrick 2009; Kütz 2005) is governed by a cybernetic understanding of control, emphasizing the process of setting, measuring and evaluating performance standards in terms of service level agreements (SLAs). In academic literature, most work focuses on project-based contexts, like in-house or outsourced software development projects (Boland 1979; Choudhury & Sabherwal 2003; Henderson & Lee 1992; Kirsch 1996, 1997, 2004; Kirsch, Sambamurthy, Ko & Purvis 2002). Until now, little attention has been paid to control in a non-project-based IS outsourcing context, like infrastructure outsourcing (Kern & Willcocks 2000b; Rustagi et al. 2008). Kern & Willcocks (2000b) analyzed the role of contracts in supplier-buyer relationships and found that contracts, beyond their legal nature, outline a number of control dimensions that are then enforceable during post-contract management. Rustagi et al. (2008) investigated the amount of formal control, defined as the “variety of mechanisms used by a client to exercise control over a vendor and the extent to which the mechanisms are used” (ibid., p. 126). Empirical evidence suggests that major antecedents for the amount of control are task uncertainty (positively correlated), trust, and technical and relationship management knowledge on the client’s side (all negatively correlated). Table 9 provides an overview of the discussed works and their respective units of analysis.

[...]

^[1] An extensive overview of influential literature in these fields can be found in Kern (1997).

^[2] The summary for TCE is based on Rindfleisch & Heide 1997, who give a great review of transaction cost theory and related research in the field of marketing.

^[3] Gellings (2007b) is as an example for contract-based governance in IS outsourcing relationships.