Sage Journals: Discover world-class research

Abstract

This research derives testable hypotheses from property rights theory regarding the allocation of decision authority in contracts. Government-based professional service contracts provide the context for study. Using content analysis we code 361 contracts to identify the main contributors of knowledge assets to the work product along with the relevant decision authority, which is reflected in provisions that stipulate who must accomplish the work and how it must be done. Results of various regression models show that the distribution of knowledge assets does influence the assignment of decision authority, specifically: (1) a higher fraction of decision rights is allocated to government when government is also identified as a source of expertise; (2) authority is less centralized when both parties contribute knowledge inputs; (3) bargaining power (lobbying) leads to a lower fraction of decision rights allocated to government.

Keywords

Governance structures outsourcing decision rights contracts knowledge governance

Introduction

Academics and pundits often use the term “knowledge economy” to emphasize the value of human capital and skilled labor to the economy. Human capital and organizational capital are viewed as key organizational assets (Williamson, 1996). Yet, knowledge needed to create value is increasingly dispersed either geographically (e.g. Doz et al., 2001) or in terms of specialization (e.g. Matusik and Hill, 1998). Consequently, knowledge assets are often outsourced. The distributed nature of knowledge and production requires new methods of coordination (Michailova and Foss, 2009: 7), and more extensive use of delegation (Mendelson and Pillai, 1999). According to property rights theory (PRT) the efficient and effective coordination of knowledge can be accomplished through the allocation of decision authority. In short, PRT prescribes the co-location of knowledge and decision authority; the party who possesses the knowledge most important to the outcome should also have authority to make key decisions (Alchian, 1965; Hart and Moore, 1990).

Previous research on the allocation of decision rights has produced mixed evidence, in part because different subsets of rights are often treated as functionally equivalent. Studies tend to be industry specific and survey based, due to the challenges associated with coding actual contract documents (see Windsperger and Yurdakul, 2007 for a review). A study is needed that accounts for the functional differences in decision rights; one that also examines whether the party contributing knowledge assets is also provided the relevant decision authority to leverage that knowledge. Studies able to confirm or disconfirm previous findings using different methods and contexts will also help advance the cumulative bodies of work on knowledge governance, contracts and decision rights.

Our main research question is derived from PRT: Does the distribution of knowledge assets influence the assignment of decision authority? We examine the content of 361 government-based professional service contracts along with other government reports to identify the main contributors of knowledge assets to the work product. We focus on contracts from two social service agencies so that our sample includes a range of service types (delivery of health services, engineering, legal services, etc.) but also achieves a desirable level of contract standardization. The decision authority of relevance to the analysis is reflected in provisions that stipulate who must accomplish the work and how it must be done. Regression results are generally consistent with PRT: (1) a higher fraction of decision rights is allocated to government when government is also identified as a source of expertise; (2) authority is less centralized when both parties contribute knowledge inputs; (3) bargaining power (lobbying) leads to a lower fraction of decision rights allocated to government.

Section one provides the theoretical framework; here we summarize the main concepts of PRT, link PRT to the contractual features of professional service contracts and then develop our hypotheses. Section two explains our research application, including our sample, coding, measures, and regression equations. Section three provides results. Section four concludes with a discussion of research implications, limitations and avenues for future study.

Theory

The main concepts of property rights

Property rights “encompass the individual’s ability, in expected terms, to consume the good (or service the services of an asset) directly…or indirectly…” (Alchian, 1965; Barzel, 1997: 3). Property rights are neither absolute nor static. They are subject to one’s own efforts of protection as well as to the recognition and enforcement by others. For example, you may own an apartment and have the right to profit from a sale, but your profits will likely diminish if someone breaks into the apartment or if the apartment is otherwise damaged. As a landlord you can hold tenants responsible for damage in an agreement, presumably enforceable in court. In the language of PRT, those who can increase or decrease the value associated with an asset share “residual claimancy” (Barzel, 1997: 3–4). Thus, PRT is concerned with the association between the ownership of an asset and control over any potential benefits that can be derived therefrom; the key question is how to allocate rights so that the value of resources is maximized.

The ability of others to diminish the value of an asset arises because contracts are incomplete. This premise follows Coase (1960), who instructed decades ago that resource allocation is efficient when rights are well defined and the costs of transacting are zero. However, because costs of transacting are positive and delineating and enforcing rights is costly, we need a system for rights assignment. PRT also assumes rights assignment is cost-prohibitive “if done to perfection” (Barzel, 1997: 7). It follows that in constructing agreements parties must consider both the cost of specification and the potential gains or losses associated with clarifying rights. PRT assumes individuals seek to maximize value; thus, contracting parties will clarify rights or seek reallocation of a previous rights assignment when doing so will lead to appreciable gains.

PRT provides a couple of basic propositions for the allocation of rights to maximize surplus. To the extent possible, decision authority should concentrate rewards and benefits with persons responsible for them. The appropriate allocation is expected to motivate self-interested decision makers to exercise judgment consistent with organizational objectives (Jensen and Meckling, 1992: 251–253). Yet, the proposition presupposes the person with decision authority is also capable of making decisions that lead to value maximization. To ensure the co-location of knowledge with decision authority PRT also proposes, “authority should be allocated to derive any advantages of specialization” (Alchian, 1965: 140).

Before developing our hypotheses we summarize the features of professional service contracts and how they fit the PRT framework.

Professional service contracts

Professional service contracts are, at their core, agreements for the use of human capital. In contrast to agreements for the purchase of office supplies, for example, these contracts explicitly involve the engagement of knowledge workers to accomplish work tasks. The work itself is characterized by a degree of uncertainty. The focal contracts involve the implementation of government programs, for example Medicaid, or the implementation of an initiative, such as a new computer system or marketing campaign, or consultation on a problem, perhaps legal or engineering. The parties are unaware of all the possible issues that may arise so they are unable to fully specify all contract features. Numerous contracts provide evidence of this uncertainty. For example, one contract stipulates: “the Contractor shall prepare program reports if requested by the State. ” Another stipulates: “the Contractor may be required to submit monthly reports.” In short, the parties opt for vague language when outputs or outcomes are difficult to measure, verify, or expected to change. Contracts that do specify performance metrics often leave room for the parties to adjust standards or report timing if necessary.

Under the PRT framework, the need to allocate decision rights is a function of the challenges associated with specification. Many contracts for professional services involve specialized tasks, which are especially difficult to anticipate and/or articulate and thus also difficult and costly to specify. Consequently, professional service contracts are an ideal sample for testing PRT. The focal contracts cover a range of services and by extension, varied degrees of specialized knowledge. For example, a contract for accounting services involves a different knowledge domain and level of training compared with a contract for temporary clerical work. Many contracts (about 22 percent) reference the supplier’s expertise somewhere in the main document or in an exhibit. In general, expert knowledge has a tacit and intangible quality, which makes it more difficult to communicate and to absorb, and thus more costly to transfer (Polanyi, 1958; Von Hippel, 1998). For example, an accountant who is very capable of completing a financial report may nevertheless have difficulty articulating the precise tasks involved in its completion. A study by Kogut and Zander (1993) supports this logic, finding knowledge that is difficult to codify is less likely to be contracted.

The contracts include functionally different sets of decision rights, for example rights to compensation, monitoring rights, termination rights, and rights involving the direction and management of the individuals involved in the work processes.¹ Given our reliance on PRT and the importance of intangible assets to rights allocation, the rights most relevant to this study relate to authority over human resources. Thus, in this study we focus on contract provisions that give either party authority to decide who will complete the work or how the work will be done. We refer to these rights as authority over knowledge assets. In sum, professional service contracts are characterized by a degree of uncertainty and involve specialized (human) investments. Their features also fit the PRT framework: they are characterized by incompleteness and they involve an exchange of knowledge assets, which have an intangible, non-contractible quality.²

Hypotheses

Early PRT literature focuses on the value and rights associated with physical assets, such as buildings or equipment (e.g. Grossman and Hart, 1986). However, extensions to the theory acknowledge “human assets” can also be important production inputs (Hart and Moore, 1990: 1121). To apply PRT theory to our contract setting requires: (1) the identification of the party or parties contributing valuable knowledge inputs to the project or initiative under contract; (2) the identification of the relevant authority/decision rights associated with valuable inputs; and (3) an assessment of costs associated with the transfer of authority/ decision rights.

To determine if the government (the buyer) possesses knowledge important to the contract outcome we examine the work description, which by Indiana state rules must be fully detailed in the contract document (Indiana Department of Administration, 2014a). We also examine the mission of the contracting agency. In other words, we assume that an agency is competent in areas directly related to its mission. Consider a contract with Family Social Services Administration (FSSA) involving the administration of a program under its jurisdiction. According to FSSA’s 2014 mission (Indiana Family Social Service Agency, 2014), the agency develops finances, and administers various programs to provide healthcare and other social services to Indiana residents. Thus, on its face, the contract work is closely aligned with the agency’s mission. In contrast, when the FSSA enters into a contract for janitorial services or for the maintenance of its copiers the agreement is not directly connected to the agency mission, although it may still be important to the functioning of the agency. Likewise, the Indiana Department of Child Services (DCS) may enter into a contract with an organization to provide parenting classes. According to the DCS website, the agency provides direct attention and oversight in two critical areas, “protection of children and oversight and child support enforcement” (Indiana Department of Child Services, 2014). Thus, the agreement for parenting classes is directly related to the agency’s mission. By contrast, the DCS may contract for engineering services or hire a realtor to assist with securing office space. In such cases, the outsourcing decision is based in part on the fact that the services needed are outside the agency’s bailiwick. The engineering firm or real estate firm under contract provides services important to the agency, but the work under contract is outside the scope of the agency’s social service mission. Contracts for work closely aligned with the agency mission reflect the importance of government expertise to the contract outcome.

To determine the locus of relevant authority over knowledge assets, we look to other contract provisions. In addition to specifying the work to be undertaken, the standard professional service contract contains numerous provisions that specify decision authority. In the perspective of PRT, authority is a property right. By definition, property rights provide individuals or organizations control over resources owned (Alchian, 1965: 129). These property rights take the form of the contract provisions describing who will do the work and how it must be done. To give examples of relevant decision authority in our contract setting, some contracts cap the number of hours or hourly rate for services. Other contracts specify key persons to perform the work. Alternatively, a contract may provide the supplier a wide range of discretion to determine how to get the job done; the supplier determines how many hours to dedicate to the work or whether or not to subcontract specific tasks. In general, decision authority given to one party restricts the authority of the other party; the more government directs or controls the work effort, the less discretion given the supplier and the more the supplier is constrained in the use of its own resources. Thus, the analysis of constraints is also incorporated into the study of property rights (Barzel, 1997: 11).

The question becomes: Is it more costly to transfer decision authority to the government or to the supplier? We propose it is more efficient for government to possess relevant property rights (decision authority over the knowledge assets) when government contributes valuable knowledge inputs. Prahalad’s and Hamel’s (1990) description of how organizations develop around their mission is helpful for understanding why this knowledge is both a valued input and costly to transfer: Specifically, organizations develop a set of core competencies around their functional areas. Core competencies involve collective learning, for example how to coordinate diverse production skills and integrate multiple streams of technologies so that an organization can accomplish its mission. This type of knowledge often involves the structure of decision processes, organizational routines, procedures for finding answers, and unwritten rules about handling different situations. Core competencies often involve many levels of people and multiple functions (Prahalad and Hamel, 1990).

Knowledge embedded in the organization’s general social fabric is not easily codified (Rosen, 1972). Since it involves organizational expertise it also has a tacit quality (Polanyi, 1958). Tacit knowledge possesses a “stickiness” or intangibility (Von Hippel, 1998). Two studies further support our logic. Zander and Kogut (1995) find knowledge that less common knowledge is difficult to codify and takes more time to transfer. Similarly, Contractor and Ra (2002) find intangible knowledge can be difficult (costly) to transfer because it is very specific.

PRT prescribes allocating authority to derive any advantages of specialized application (Alchian, 1965: 140). As Jensen and Meckling (1992) point out, allocation should address the “rights assignment problem” by moving knowledge and the decision authority together so that specialized knowledge can lead to better decisions. From the government’s perspective, maintaining decision authority for services that directly involve the agency’s mission will lead to better decisions. As the experts in social service programs they are best positioned to make decisions as to how these programs should be carried out. For example, they possess important knowledge about federal reimbursement procedures that the contractor is not likely to have. Therefore, the agency they should decide how some implementation tasks should be performed because it will affect the budget available for a project and ultimately their ability to compensate the supplier. In this respect, the co-location of knowledge and decision authority can maximize joint surplus.

The agency will also have an added incentive to minimize the cost of poor information by retaining control over decision authority in areas where it is most likely to be held accountable, i.e. when contracts involve programs directly related to its mission. In Indiana all state agencies are held to performance standards.³ Each agency reports annually on its performance metrics and must provide a correction plan when it falls short of benchmarks. In recent years the state has also increased requirements for agency transparency (Indiana Office of the Governor, 2014).

These arguments underscore the incentive properties of property rights that can be found in the literature. According to Hart and Moore (1990), ownership (in the form of decision rights) gives the holder control over its attributes, which provides a residual claim to income that can be generated from an asset. And Jensen and Meckling (1992: 251–253) argue that the appropriate allocation can motivate self-interested decision makers to exercise judgment consistent with organizational objectives. From these observations it follows that:

H1: contracts involving work directly related to the mission of a government agency will be associated with a higher fraction of decision rights allocated to government.

We also observe contracts that describe a more collaborative setting, in which both parties contribute an expertise. One case exemplifies the situation: having come under recent criticism, an agency needed an image boost and hired a marketing firm that is experienced in public relations. Yet the contract also requires the marketing firm to work with agency personnel to learn about agency culture, procedures and the nature of the criticism in order to complete the work. We describe this situation as one in which both parties must contribute knowledge inputs: the agency has insider knowledge necessary to complete the work; the marketing firm possesses at least some knowledge that the government does not possess; contract success requires both parties to contribute important know-how. Moreover, knowledge inputs from both parties are more than just incidental to the contract outcome.⁴ The question becomes: How should decision making be allocated in such a situation?

The knowledge literature provides reasons to expect a different allocation of decision authority in the presence of jointly contributed knowledge. Specifically, when contract parties must work closely together to accomplish contract objectives we expect them to develop a capacity for understanding each other. The knowledge literature refers to this as “absorptive capacity,” which is defined as “the ability to recognize the value of information to assimilate and apply it” (Cohen and Levinthal, 1990: 120). Absorptive capacity is considered an organizational phenomenon (Gupta and Govindarajan, 2000). It includes prior knowledge and basic skills, as well as the ability to apply knowledge to meet organizational needs (Cohen and Levinthal, 1990; Szulanski, 1996). The existence of a common language or shared vision and ability to solve problems all increase absorptive capacity and lower knowledge transfer costs (Szulanski, 1996).

Notwithstanding, scholars provide different theoretical logics on the effects of blending knowledge on the allocation of decision authority. For example, Arrow’s (1974) work implies that under shared knowledge we should observe a higher fraction of decision rights associated with government as “authority, the centralization of decision-making serves to economize on the transmission and handling of information” (p. 69) Yet, Hayek (1945) argues that the market is superior in the presence of dispersed knowledge, especially when knowledge is context-specific and tacit (p. 521). Following Hayek we should observe a lower fraction of decision rights allocated to government. Given the opposite claims of Arrow and Hayek, we provide alternative hypotheses as to the effect of joint knowledge on the assignment of decision rights:

H2a: contracts involving joint-knowledge inputs will be associated with a higher fraction of decision rights allocated to government.

H2b: contracts involving joint-inputs knowledge will be associated with a lower fraction of decision rights allocated to government.

According to PRT, the value of an asset is subject to change, for example based on the behavior of others, which is expected to prompt changes in property rights allocations (Barzel, 1997). We argue that the value of the supplier’s knowledge is diminished in the presence of competition. As we see it, the value of the supplier’s knowledge is closely related to the concept of agency costs. Agency costs include the sum of costs associated with structuring contracts, monitoring and bonding, as well as the inevitable (residual) losses due to the inability to completely align the interest of the agent doing the work with the interest of the principal (Jensen and Meckling, 1976). Indeed, knowledge is less valuable when it is replaceable. If there is a competitive market for the service, the supplier’s knowledge is less critical (and less valuable) to the contract. Thus, the government will not be as likely to give the supplier decision authority in the presence of competition.

As Barzel (1997: 10) indicates, the exchange value of an asset includes both the potential income it can generate but also the cost of measuring and policing. The cost of measuring and policing decreases in the presence of competition. Competition at the bidding stage can give the government access to more information, for example about the supplier’s costs to complete the work. Similarly, Williamson (1996) discusses the concept of “information impactedness,” pointing out the hazard when information is known to one or more parties but cannot be costlessly discerned by or displayed for others (p. 65). Even after the contract is awarded, potential competition for future work may give government the means to verify the supplier’s work efforts over the course of the relationship. Thus, competition can lower the value of the supplier’s expertise, in which case the government will be less likely to defer to the supplier’s decisional authority. Accordingly, we provide a third hypothesis:

H3: the presence of contract competition will be associated with a lower fraction of decision rights allocated to government.

Our final hypotheses owe to the institutional stream of PRT and the work of Libecap (1989) and North (1990). The institutional stream of PRT is fitting to our contract setting for at least two reasons. First, we know state guidelines (formal institutions) play a role in contract specification and we presume agency practices (informal institutions) also affect contract choices. Institutions are “humanly devised constraints that shape interaction” (North, 1990: 3). Second, public organizations are often differentiated from private organizations on the basis of their political environment (Rainey, 1989). We expect politics and bargaining to influence contracts. In fact, we can identify at least one state regulation that may directly affects the ability of supplier’s to influence agency officials and in turn contract specifications, i.e. lobbying regulations. The institutional stream of PRT explains the mechanism by which these institutions affect changes in a property rights.

According to the institutional stream of PRT, precedent plays an important role in how property rights develop (North, 1990). We expect government agencies to develop standard practices or defaults for drafting contracts. They will choose standard contract models allocating decision rights with boilerplate options recommended by the Indiana Department of Administration (Indiana Department of Administration, 2014a) unless there is a clear value to changing the status quo. The same holds true for the suppliers; they will prefer entering into standard-type contracts and will be reluctant to agree to contract language that is inconsistent with their past practices. Notably, one provision in government contracts, including all contracts in this study, does stand out from typical business-to-business contracts, i.e. termination of convenience clauses. A termination of convenience clause, also referred to as a termination-at-will-clause, trumps any other controls. The clause gives government the right to terminate the contract for any reason; it does not need to claim performance failure. We suspect most businesses would not agree to include this clause when contracting with another business. On the other hand, the fact that businesses generally accept this term in government contracts is consistent with PRT’s view of institutions as stable and resistant to change (North, 1990).

What prompts a change from norms? First, a change must be permitted under the rules. As noted, government agencies follow state guidelines for contract drafting. In addition, Indiana-registered lobbyists are permitted to make contact with an executive branch agency “for the purpose of trying to influence the outcome of a contract, a lease, another financial arrangement, or a rule” (Indiana Administrative Code, 2012: 25 IAC 6).⁵ Thus, in the context of the focal study, Indiana Department of Administration (IDOA) guidelines as well as state rules and regulations for lobbying define the range of feasible contracting solutions.

PRT assumes utility maximization, which implies property rights will change when potential benefits associated with a change make the effort worthwhile (Barzel, 1997: 10; North, 1990: 19–20). The greater the size of anticipated benefits from a change in property rights the more likelihood that change will be adopted (Libecap, 1989: 28). Nevertheless, not all parties have the requisite bargaining power to bring about change. Organized interests are better positioned in the negotiation process and more likely to receive an allocation of decision rights they perceive as favorable. Accordingly, North (1990) describes how various competing interests bargain over the distribution of property rights and how legal rules and regulations, which are considered property rights themselves, can either facilitate or block change. In the end, the division of gains depends on relative bargaining power (North, 1990: 49). This leads to two related hypotheses, the first positing a direct connection between lobbyists and a favorable allocation of decision rights and the second positing an interaction effect between the contract amount and lobbying.

H4a: the identification of a lobbyist with a contract will be associated with a lower fraction of decision rights allocated to government.

H4b: lobbyists will succeed in negotiating a favorable allocation of decision rights (lower fraction of decision rights allocated to government) when they perceive their efforts will yield sufficient benefits.

Application

Sample

When studying contract design choices, the sample should include contracts of similar purpose and structure (see also Arruñada et al., 2001; Brickley et al., 2003; Elfenbein and Lerner, 2009). Accordingly, our sample is drawn from two state agencies in Indiana with similar social services missions: FSSA and DCS.⁶ To obtain our sample, we identified an initial set of 500 contracts from the state contract repository (Indiana Department of Administration, 2014b). After removing amendments, intergovernmental agreements, and renewal contracts our final sample comprised 361 contracts covering 10 years (2004 through 2014).

The contracts in our sample meet the criteria of “structured diversity” (see also Anderson and Dekker, 2005; Joskow, 1987), which is to say they are characterized by both a degree of standardization and a degree of heterogeneity. The contracts are standardized in that many provisions are boilerplate and required by state guidelines. This characteristic facilitates coding. Yet, there is also heterogeneity with respect to important contract elements: the contracts cover a range of services, for example accounting; legal; program development; and clerical; among other services. They also vary in duration. Most contracts (about 84 percent) bind the parties for more than one year; some contracts (about six percent) extend for over three years, and almost all contracts (about 90 percent) have an option for renewal. There is also variation in contract amounts. The median amount is about $200,000. The bottom quartile ranges from $68,000 and below, while the top quartile is at or above $880,000. If we consider contract length to be a proxy for completeness, the degree of completeness also varies; some contracts include only two pages, while others exceed 200 pages plus exhibits.

Coding

We took several steps to objectively and reliably code relevant contract language. We carefully reviewed the agreements along with other state documents to identify key features and variations in the agreements. The Indiana Department of Administration’s manual of contract guidelines was a key source for understanding the contract language (Indiana Department of Administration, 2014a). After several iterations we developed a coding form to capture key contract features. We followed guidelines in Neuendorf (2002) for coding practices, operationalizing measures, reconciling coding inconsistencies, and establishing inter-coder reliability. Two different individuals coded key features of the contracts and information from the agencies’ mission statements; each coded all content at least twice.⁷ To reduce coding errors, the coding task was confined to manifest content, recording values of either 0 (absence) or 1 (presence) for most provisions.⁸ The latent meaning for coded values was left to the authors.

Measures

Dependent variable

Our dependent variable measures the extent of decision authority, or conversely the extent of limits on discretionary authority, over knowledge assets. We construct the measure by coding contract clauses requiring specific individuals to do the work and clauses specifying work methods, including: clauses naming key (key) persons; clauses restricting the supplier’s use of subcontractors (subcontractor limits) ; clauses capping remuneration either in hours or units (caps) ; clauses specifying the order in which tasks are to be performed (work stages) ; clauses requiring a formal contract amendment to change the scope of work (work changes) . Each indicator clause is coded dichotomously (1 = presence; 0 = absence).

Our measure is an additive index, valued from 0 to 5. Higher values indicate decision authority is more centralized with government (central) (see Arruñada et al., 2001 for a similar approach). There are tradeoffs associated with using a simple additive index. On the one hand it provides an inclusive measure of decision rights that have the same functional purpose and that are most relevant to the use of knowledge resources. It also presents a measure of authority asymmetry—higher values imply more governmental authority and less discretion for the supplier. The subset is also distinguishable from monitoring rights, for example, which more directly address opportunism (Mesquita and Brush, 2008).⁹ The disadvantage of using an additive index is that we assume each item should be given equal weight. Items that form the index are summarized in Table 1.

Table 1.

Description of dependent variable clauses.

Clauses	Obs	Mean	SD	Max
Caps	361	0.094	0.292	1
Key employees	361	0.828	0.378	1
Work changes	356	0.188	0.391	1
Work stages	353	0.329	0.470	1
Subcontractor limits	360	0.628	0.484	1

Independent variables

Our main independent variable (core) measures the knowledge assets contributed by government, and specifically whether the contract tasks directly involve the agency mission. We assume such inputs reflect a valuable contribution by government. We construct the measure from the contract itself, which describes the work to be undertaken along with the agency mission statements. Our measure (core) is valued at 1 if the task descriptions match any part of the agency mission, and 0 otherwise. The measure is similar to intangible knowledge assets constructed by Gurcaylilar-Yenidogan and Windsperger (2013) and Windsperger (2009).

Joint

Joint knowledge refers to situations where both the supplier and government provide knowledge inputs. Scholars argue that the advantages of centralized decision authority may be a function of how dispersed knowledge is and whether it is shared (Arrow, 1974; cf. Hayek, 1945). The level of interaction and coordination may also impact the allocation of decision authority and the value of the alliance (Elfenbein and Lerner, 2003: 361). Our measure (joint) is valued at 1 where there is evidence in the contract document that the government and the supplier must work together to complete the contract work, and 0 otherwise.

Competition

Bargaining power may be important to decision allocation (Elfenbein and Lerner, 2003). The method of source selection (request for proposal, bidding, negotiation, or special procurement) indicates whether there was contract competition. We code this variable from the Executive Summary Document (EDS) associated with the contract. Contracts subject to competition are valued at 1, 0 otherwise.

Lobby

State rules permit organized interests to lobby executive agencies. Lobbyists have the ability to affect contract allocation decisions. Consistent with principles of utility maximization, incentives for lobbying will increase with the importance of the contract. Lobbyist information comes from two sources: the database of state registered lobbyists and the vendor code on the contract EDS. Lobby is valued at 1 for contracts in which the supplier or an employee of the supplier is registered with the state to lobby the executive agency under contract, and 0 otherwise.

Control variables

To control for other effects and to test for alternative explanations we included various other measures in our models. The importance of the service under contract (i.e. larger contract) is associated with the need to avoid decision errors and thus may affect allocation decisions (Jensen and Meckling, 1992). The dollar amount is a proxy for contract importance; the variable is in logarithm form to correct for its skewness. Agency staffing levels and agency contract experience may affect the agency’s ability to negotiate, specify, and/or monitor the contract. Employees is a proxy for agency resources, measured as the number of agency staff for the year prior to the relevant contract year; data for the measure is obtained from IDOA reports (Indiana Department of Administration, 2014c). The number of contracts executed during the year prior to the relevant contract year is a proxy for contract experience and also obtained from IDOA reports (Indiana Department of Administration, 2014c). We also coded for the presence (1) or absence (0) government equipment or files used in the course of contract work; government equipment is coded directly from the contract document. Following Gulati (1995) we coded for previous working relationship (previous relationship) between the government and the supplier. To determine if the government and supplier had a previous contract relationship we matched the vendor number on the contract with vendor numbers in the IDOA contract repository (Indiana Department of Administration, 2014b). Table 2 provides summary statistics for all variables used in our analysis.

Table 2.

Descriptive statistics.

Variable	Obs	Mean	SD	Min	Max
Central	352	2.085	0.75	0	4
Joint	361	0.357	0.48	0	1
Core	361	0.33	0.471	0	1
Competitive	361	0.507	0.501	0	1
Lobby	361	0.296	0.457	0	1
Employees (000)	361	3.329	0.546	2.872	3.98
Contract experience	361	385.413	89.476	191	498
Previous relationship	361	0.54	0.499	0	1
Government equipment	353	0.839	0.368	0	1
Amount (ln)	360	12.401	1.698	7.923	16.829

Regression analysis

The nature of the dependent variable determines the appropriate statistical modeling strategy. In our case, the dependent variable can be conceptualized in two ways. First, our construct can be viewed as an ordinal measure. Ordinal measures assume a rank ordering but also assume the true distance between two rights and three rights, for example, is unknown. Secondly, our construct can be viewed as an event count. Similar to the ordinal construct, a higher count represents more centralization of decision rights. However, unlike the ordinal conceptualization, a count represents a discrete, positive integer where the distance between outcomes is constant. Both strategies are appealing as they both provide a comprehensive measure of contract asymmetry with respect to decision rights in the domain of interest authority related to the management and control of human assets. We show the count model to corroborate results and to cover the possibility that different contracting parties may also view decision rights differently. Importantly, the results of both approaches are entirely consistent. Our analysis will focus primarily on an ordinal conceptualization of the dependent variable, as it allows for a more nuanced discussion of results.

When treating the dependent variable as an ordinal measure, applying a proportional odds model (ordered logit) is appropriate. The model assumes a variable Y with j categories of the ordered response, where,

P (Y \geq j) = e x p (α j + β x) (1 + e x p (α j + β x))

with j having possible values from zero (i.e. no decision rights) to five. However, very few contracts exhibited extreme values: six contracts had no provisions and three contracts had five provisions. Therefore we combined contracts from these categories with the next highest and lowest levels (one and four) in order to improve estimation.¹⁰ Thus j has a maximum value of 4 and minimum value of 1.

Ordered logistic regression assumes that the independent variables’ effect is consistent across the categories of the dependent variable—also known as the parallel regression assumption.¹¹ Table 3 presents the results of our statistical models. Regressions 1 and 2 represent ordered logistic estimation and regression 3 utilizes Poisson regression for count data. As noted above, we will focus primarily on the ordered logistic regressions.¹²

Table 3.

Regression results.

	1		2		3
	Ordered logistic				Poisson
VARIABLES	coef	se	coef	Se	Marginal effects	se
Core	0.248**	(0.126)	0.251**	(0.112)	0.0674*	(0.0382)
Joint	−0.284**	(0.140)	−0.316**	(0.139)	−0.0911*	(0.0467)
Competitive	−0.515**	(0.232)	−0.436	(0.277)	−0.135*	(0.0749)
Lobby	−2.199***	(0.305)	−2.232***	(0.296)	−0.612***	(0.0713)
Employees			0.0817	(0.258)	0.00460	(0.0881)
Contract experience			0.00355*	(0.00215)	0.00122*	(0.000713)
Previous relationship			0.121	(0.245)	0.0185	(0.0813)
Government equipment			1.114**	(0.471)	0.311**	(0.126)
Amount (ln)			0.117	(0.0875)	0.0281	(0.0260)
cut1	−4.066***	(0.358)	−0.148	(1.189)
cut2	−0.731***	(0.233)	3.321***	(1.201)
cut3	1.606***	(0.449)	5.700***	(1.287)
Constant					0.561***	(0.190)
Observations	352		351		351

Robust standard errors in parentheses.

***p < 0.01, **p < 0.05, *p < 0.1.

Regression 1 presents the results of a parsimonious model that includes our primary independent variables, while regression 2 presents a more fully specified model that includes a set of control variables. By scanning coefficients across models, we observe statistically significant effects for three variables important to our theoretical arguments. We can make some general statements with confidence. First we have support for hypothesis 1: contracts for core professional services increase the probability of more centralized decision rights, on average. We have support for hypothesis 2b: when both parties contribute (joint) knowledge inputs (joint) valuable to the contract, decision authority is less centralized. We have support for hypothesis 4a: registered lobbyists appear to be effective in limiting the number of decision rights granted to government. We also have tentative support for hypothesis 3: in our parsimonious model competition decreases the likelihood of additional decision rights. Although the coefficient no longer meets the threshold for statistical significance in the fuller model, the negative direction of the effect is maintained. We also see statistically significant effect on two control variables: contract experience and government equipment, both increase the probability of observing additional central decision rights. We discuss the possible meaning of effects associated with control variables in the next section.

While these general findings provide a useful overview, the ordered logistic model allows for more nuanced prediction and assessment of substantive significance. For example, while contracting for core services has a positive, statistically significant effect on the dependent variable, the substantive effect is small; the average change in the predicted probability across all of the categories is 0.02. A closer inspection reveals that core contracts decrease the probability of outcomes 1 and 2 by 0.02 and 0.01, respectively, and increase the combined probability of higher levels of centralization (outcomes 3 and 4) by 0.035. This pattern is consistent with the logic supporting hypothesis 1. The impact of joint action is also substantively small, yet supportive of hypothesis 2b. Joint action increases the probability of outcomes 1 and 2 by approximately by 0.05 and decreases the probability of outcomes 3 and 4 by 0.04. Adding to our overall confidence in the results, the marginal effects from the Poisson estimation (regression 3) are consistent with the ordered logit. Specifically, core contracts increase the predicted count of central decision rights from 2.07 to 2.14, roughly a 3 percent change. Joint action decreases the expected count from 2.07 to 1.98.¹³

Suppliers who are registered lobbyists have a pronounced impact on the allocation of decisions rights. Specifically, the predicted probability increases from 0.16 to 0.41 for the lowest category of the measure (outcome = 1). Similarly, lobbyists are able to decrease the likelihood of outcomes 2, 3 and 4 by 0.09, 0.22 and 0.03, respectively. These results suggest the lobbyists successfully negotiate for more decision-making discretion when contracting with government agencies. This is strong support for hypothesis 4a. The marginal effect of the lobby variable in the Poisson regression (regression 3) provides further confidence in our result—lobbyists are able to negotiate for 0.61 fewer central decision rights. Given the limited range of our dependent variable, we believe this constitutes a substantively significant change.

Lastly, we hypothesized both a direct effect and a conditional effect for our lobbying variable. As to the latter, we hypothesized that lobbyists are likely to negotiate for even fewer central decision rights when the contract is more lucrative (hypothesis 4b). We constructed a variable to test whether effectiveness of lobbying was moderated by the contract amount. We found no support for an interactive effect. We do not show the coefficient in our results table because inclusion in the model causes the variable to fail the parallel regression assumption. Notwithstanding, we ran additional tests to consider possible differences in the effectiveness of lobbyists at different contract amounts. A difference in means test also indicates that lobbyists have 1.56 relevant decision clauses on average compared with 2.18 for the non-lobbyists in the sample. Overall we are confident that lobbyists are effective regardless of the amount of the contract, i.e. their impact is not moderated by the contract amount.

Discussion and conclusion

The allocation of decision rights has been investigated in various inter-organizational contexts (for example, Arruñada et al., 2001; Baker et al., 2008; Brickley et al., 2003; Elfenbein and Lerner, 2003; Lerner and Merges, 1998; Windsperger and Yurdakul, 2007). However, this is the first study that we are aware of focusing on a range of professional service agreements or on the allocation of decision rights in agreements between government and the private sector. We are reluctant to directly compare our findings with previous work due to the use of different measures. Nevertheless, the present research adds to the evidence that PRT propositions are just as relevant to knowledge (intangible) assets as they are to physical assets (see Gurcaylilar-Yenidogan and Windsperger, 2013).

The present research has limitations. Although our findings are robust to various specifications, our modeling strategy is based on how we conceive of and measure decision authority in our contract setting. Our dependent variable is conceived as an explanatory combination of indicators. We intentionally considered a small subset of decision rights so as to match the assets of importance in our study to the most relevant decision authority. Failure to include all facets of relevant decision authority is always a potential problem with measures of this type. In addition, we assume that more contract provisions associated with directing human resources mean more decision authority to one party, and conversely more constraints to the other party. And in fact our treatment is consistent with what contracting officials and suppliers tell us. Nevertheless, different parties may view some decision rights as more important or in some way functionally different than other decision rights, and this may have consequences for the measurement and modeling strategy.

Viewing our findings as a whole, we are confident that parties do consider the source of knowledge and costs associated with knowledge coordination when assigning decision rights. When government contributes important knowledge assets, in the form of agency expertise, a higher fraction of decision rights is also likely to be allocated to government. Moreover, parties adjust their allocation decisions when knowledge contributions come from more than one source as well as when contributions are both shared and also involve the government’s core mission. Thus, contracting parties appear to understand the payoffs associated with relying on experts to make decisions. And, they consider the costs associated with obtaining specialized knowledge, as they are more likely to assign decision authority to the experts when knowledge is costly to transfer.

It is one thing to say that parties adjust decision authority consistent with understanding the value and costs associated with the contribution of knowledge, yet it is quite another thing to say the advantages and disadvantages of centralized and decentralized decision authority are the only factors or even the most factors considered by the contracting parties. For example, lobbying efforts do appear to sway decision makers even when contract amounts are not significant (lobbying effects are not moderated by amount). Lobbying efforts are consistently associated with lower fractions of decision rights allocated to government.

We also consider our findings in light of Hayek’s argument that decentralization is optimal when knowledge is dispersed. On the surface our results provide cautious support for Hayek’s arguments on the benefits of decentralization. However, this interpretation assumes that knowledge contributed by more than one source is by definition dispersed. If we look more closely at the contracts that involve knowledge contributions from both the supplier and buyer, those same contracts also tend to detail a high level of interaction between the parties. We suspect this because there is a level of uncertainty about the problems to be solved. When solutions are highly dependent on interactions among individuals with different knowledge sets, the problems are said to be of a non-decomposable type (Nickerson and Zenger, 2004). Non-decomposable problems require heuristic searches and a collective development of a solution landscape, which means that people have to agree on solutions. Thus, it is possible that knowledge under this scenario is not dispersed in Hayekian terms. Rather, the parties may be choosing to come together and to interact because they are dealing with complex problems of a non-decomposable type, in which case they may be substituting coordination for decision authority. Compared with other studies, our findings are inconsistent with research by Conner and Prahalad (1996) and Zollo and Winter (2002), which support the comparative advantages of a more authoritative and centralized structure to facilitate information sharing and to blend expertise.

Our findings also indicate significant effects on two control variables: contract experience and government equipment. Higher levels of contract experience (measured as the number of contracts executed in the prior year) are associated with a higher fraction of decision rights allocated to government (more centralization). Interpreting the result optimistically, it appears government takes a responsible approach to accepting decision authority, allocating more rights to itself when it also presumably has more resources to handle that responsibility.

We also find interesting effects relevant to the contribution of physical assets in these contracts. Since PRT also does not offer clear guidance on the expectations with regard to physical assets that may accompany human assets in contracts, we did not have any expectations regarding the effects of physical assets (when accompanied by knowledge assets) on the allocation of decision rights. In fact, some PRT scholars strongly disagree about the importance of physical assets to the boundaries of the firm. Hart (1989: 1771–1773) argues for a definition that emphasizes physical assets, while Cheung (1983) argues that physical assets are irrelevant. Klein (1988) discusses the importance of ‘organizational assets,’ which seems to acknowledge the importance of both tangible and intangible (human) assets. Our results show a robust effect for the contribution of government tangible assets on decision allocations; when government equipment, files, or buildings are used in the production process decision, a higher fraction of decision authority is allocated to government. This result suggests government officials successfully negotiate for consequential decision-making authority when their agency is also contributing value-tangible resources to outsourced projects. The result is consistent with the view that government officials can make responsible decisions in the same way that business owners do. The finding is inconsistent with the view that government employees, including agency heads and contracting officials, make poor choices because they do not bear the full economic consequences for bad decisions.

Yet, a brief explanation of the nature of these physical assets is in order. In general the supplier’s use of government-owned assets to fulfill contract obligations diverges from the prototypical market where a supplier owns all of the key assets required for the work. Also, contracts that mention the use of government-owned physical assets also include provisions that specify that government retains ownership once the contract expires. Thus, we have no evidence to suggest these physical assets are dedicated in the sense suggested by transaction cost economics. In other words, government equipment such as a computer system may be used over the course of the relationship, yet we assume it retains its value for future projects. Thus, in transaction cost economic terms we do not expect physical assets to lock-in the parties or to make one side more vulnerable to the other.

Interestingly, previous contracting relationships between the parties do not appear to alter the assignment of decision rights, suggesting the parties design the contracts for coordination more than for control. In this respect our results add to the recent stream of research highlighting governance as a coordination mechanism that goes beyond the costs of controlling opportunism (see Schepker et al., 2014 for a review).

Finally, we see numerous avenues for future research. The significant effects of lobbying suggest the need to incorporate a range of additional political factors into property rights models, for example election years and media exposure (salience) of the contract. Future research could also explore how rights allocations simultaneously affect other elements of the contract structure, such as incentives or contract duration. The most fruitful research will explore the effects of contract structure on performance. Some possibilities include two-stage models that consider overall satisfaction with the contract outcomes, and the likelihood of contract amendments or renewals. We leave these ambitious endeavors to future research.

Footnotes

Notes

Acknowledgement

The authors gratefully acknowledge Atlantis Richter for her diligent coding and help with content analysis. We also thank the anonymous referees for their helpful comments and insights. Usual disclaimers apply.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

References

Alchian

(1965) Some economics of property rights. Il Politico 30: 816–829.

Anderson

Dekker

(2005) Management control for market transactions: The relation between transaction characteristics, incomplete contract design, and subsequent performance. Management Science 51: 1734–1752.

Arrow

(1974) The Limits of Organization. New York: Norton.

Arruñada

Garicano

Vázquez

(2001) Contractual allocation of decision rights and incentives: The case of automobile distribution. Journal of Law, Economics, and Organization 17: 257–284.

Baker

Gibbons

Murphy

(2008) Strategic alliances: Bridges between “islands of conscious power”. Journal of the Japanese and International Economies 22: 146–163.

Barzel

(1997) Economic Analysis of Property Rights. New York, NY: Cambridge University Press.

Brickley

Linck

Smith

Jr (2003) Boundaries of the firm: Evidence from the banking industry. Journal of Financial Economics 70: 351–383.

Brown

Potoski

Slyke

DMV

(2013) Complex Contracting: Government Purchasing in the Wake of the US Coast Guard’s Deepwater Program. New York, NY: Cambridge University Press.

Cheung

SNS

(1983) The contractual nature of the firm. Journal of Law and Economics 26: 1–21.

10.

Coase

(1960) The problem of social cost. Journal of Law and Economics 3.

11.

Cohen

Levinthal

(1990) Absorptive capacity: A new perspective on learning and innovation. Administrative Science Quarterly 35: 128–152.

12.

Conner

Prahalad

(1996) A resource-based theory of the firm: Knowledge versus opportunism. Organization Science 7: 477–501.

13.

Contractor

(2002) How knowledge attributes influence alliance governance choices: A theory development note. Journal of International Management 8: 11–27.

14.

Doz

Santos

Williamson

(2001) From Global to Metanational: How Companies Win in the Knowledge Economy. New York: Harvard Business School Press.

15.

Dyer

Singh

Prashant

(2008) Splitting the pie: Rent distribution in alliances and networks. Managerial and Decision Economics 29(2/3): 137–148.

16.

Elfenbein

Lerner

(2003) Ownership and control rights in Internet portal alliances, 1995-1999. RAND Journal of Economics 34: 356–369.

17.

Elfenbein

Lerner

(2009) Exclusivity, contingent control rights, and the design of internet portal alliances. Journal of Law, Economics, and Organization: ewp035.

18.

Foss

Nicolai

(2009) Managerial authority when knowledge is distributed: A knowledge governance perspective. Chapter 5 In: Foss

Michailova

(eds) Knowledge Governance. pp : 108–137.

19.

Greene

Hensher

(2010) Modeling Ordered Choices: A Primer. New York: Cambridge University Press.

20.

Grossman

Hart

(1986) The costs and benefits of ownership: A theory of vertical and lateral integration. The Journal of Political Economy 94: 691–719.

21.

Gulati

(1995) Does familiarity breed trust? The implications of repeated ties for contractual choice in alliances. Academy of Management Journal 38: 85–112.

22.

Gurcaylilar-Yenidogan

Windsperger

(2013) Centralization of decision making authority in inter-organizational networks: Evidence from the Australian automotive industry. Journal of Global Strategic Management 13: 184–194.

23.

Gupta

Govindarajan

(2000) Knowledge flows within multinational corporations. Strategic Management Journal 21: 473–496.

24.

Hart

(1989) An economist’s perspective on the theory of the firm. Columbia Law Review 89: 1757–1774.

25.

Hart

Moore

(1990) Property rights and the nature of the firm. Journal of Political Economy 48: 1119–1158.

26.

Hayek

(1945) The use of knowledge in society. The American Economic Review 35: 519–530.

27.

Indiana Department of Administration (2014a) Professional services contract manual. Adopted June 11, 2014. Indianapolis, Indiana. Available at: http://www.in.gov/idoa/2349.htm (accessed 13 June 2014).

28.

Indiana Department of Administration (2014b) Contract repository. Indianapolis, Indiana. Available at: http://www.in.gov/idoa/2448.htm (accessed 8 October 2014).

29.

Indiana Department of Administration (2014c) Yearly contract reports. Indianapolis, Indiana. Available at: http://www.in.gov/idoa/2527.htm (accessed 8 October 2014).

30.

Indiana Department of Child Services (2014) About DCS. Indianapolis, Indiana. Available at: http://www.in.gov/dcs/ (accessed 18 November 2014).

31.

Indiana General Assembly (2014) Indiana Administrative Code. Indianapolis, Indiana. Executive Branch Lobbying. 25 IAC 6-2 -1, July 2012. Available at: http://www.in.gov/legislative/iac/ (accessed 1 December 2014).

32.

Indiana Family Social Service Agency (2014) About FSSA. Indianapolis, Indiana. Available at: http://www.in.gov/fssa/ (accessed 12 November 2014).

33.

Indiana Office of the Governor (2014) Agency performance measures. Transparency Portal. Indianapolis, Indiana. Available at: https://mybi.omb.in.gov/analytics/saw.dll?Dashboard (accessed 22 November 2014).

34.

Jensen

Meckling

(1976) Theory of the firm: Managerial behavior, agency costs, and ownership structure. Journal of Financial Economics 3: 305–360.

35.

Jensen

Meckling

(1992) Specific and general knowledge and organizational structure. In: Werin

Wijkander

(eds) Knowledge Management & Organizational Design. Oxford: Blackwell, pp. 251–274.

36.

Joskow

(1987) Contract duration and relationship-specific investments: Empirical evidence from coal markets. The American Economic Review 77: 168–185.

37.

Klein

(1988) Vertical integration as organizational ownership: The Fisher Body–General Motors relationship revisited. Journal of Law, Economics, & Organization 4: 199–213.

38.

Kogut

Zander

(1993) Knowledge of the firm and the evolutionary theory of the multinational corporation. Journal of International Business Studies 24: 625–645.

39.

Lerner

Merges

(1998) The control of technology alliances: An empirical analysis of the biotechnology industry. The Journal of Industrial Economics 46: 125–156.

40.

Libecap

(1989) Contracting for Property Rights. Political Economy of Institutions and Decisions. New York, NY: Cambridge University Press

41.

Long

Freese

(2005) Regression Models for Categorical Outcomes using Stata. College Station, TX: Stata Press.

42.

Matusik

Hill

CWL

(1998) The utilization of contingent work, knowledge creation, and competitive advantage. Academy of Management Review 23: 680–697.

43.

Mendelson

Pillai

(1999) Information age organizations, dynamics and performance. Journal of Economic Behavior & Organization 38: 253–281.

44.

Mesquita

Brush

(2008) Untangling safeguard and production coordination effects in long-term buyer-supplier relationships. Academy of Management Journal 51: 785–807.

45.

Michailova

Foss

(2009) Knowledge governance: Themes and questions. In: Michailova

Foss

(eds) Knowledge Governance. New York, NY: Oxford University Press. 1–24.

46.

Neuendorf

(2002) The Content Analysis Guidebook. Thousand Oaks, CA: Sage.

47.

Nickerson

Zenger

(2004) A knowledge-based theory of the firm—The problem-solving perspective. Organization Science 15: 617–632.

48.

North

(1990) Institutions, Institutional Change and Economic Performance. New York, NY: Cambridge University Press.

49.

Polanyi

(1958) Personal Knowledge: Towards a Post Critical Philosophy. Chicago, Illinois: University of Chicago Press.

50.

Prahalad

Hamel

(1990) The core competence of the corporation. Harvard Business Review 1990: 79–91.

51.

Rainey

(1989) Public management: Recent research on the political context and managerial roles, structures, and behaviors. Journal of Management 15: 229–250.

52.

Rosen

(1972) Learning by experience as joint production. The Quarterly Journal of Economics 86: 366–382.

53.

Schepker

W-Y

Martynov

(2014) The many futures of contracts: Moving beyond structure and safeguarding to coordination and adaptation. Journal of Management 40: 193–225.

54.

Szulanski

(1996) Exploring internal stickiness: Impediments to the transfer of best practice within the firm. Strategic Management Journal 17: 27–43.

55.

Von Hippel

(1998) Economics of product development by users: The impact of “sticky” local information. Management Science 44: 629–644.

56.

Williamson

(1996) The Mechanisms of Governance, Clinquet

Hendrikse

Tuunanen

Windsperger

(eds) New York: Springer. pp. 69–95.

57.

Windsperger

(2009) Allocation of decision rights in joint ventures. Managerial and Decision Economics 30: 491–501.

58.

Windsperger

Yurdakul

(2007) The governance structure of franchising firms: A property rights approach. In: Clinquet

Hendrikse

Tuunanen

Windsperger

(eds) Economics and Management of Networks. Franchising, Strategic Alliances and Cooperatives. New York: Springer, pp. 69–95.

59.

Zander

Kogut

(1995) Knowledge and the speed of the transfer and imitation of organizational capabilities: An empirical test. Organization Science 6: 76–92.

60.

Zollo

Winter

(2002) Deliberate learning and the evolution of dynamic capabilities. Organization Science 13: 339–351.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.19 MB

Deciding who’s in charge

Abstract

Keywords

Introduction

Theory

The main concepts of property rights

Professional service contracts

Hypotheses

Application

Sample

Coding

Measures

Dependent variable

Independent variables

Joint

Competition

Lobby

Control variables

Regression analysis

Discussion and conclusion

Footnotes

Notes

Acknowledgement

Funding

References

Supplementary Material