rat-m: “a”? 1:255:21: :\-:~ Y m. .- M, ~k an ,‘ ~ fiat)?- i vii" *3;}f§:=1‘ . , t - . ‘ :w» o..- - 3'43: ~ «A n-«A. a. p...” n “a. 'I ‘V" 3"; .,, 47",”, v:-.‘.V In. Eu” [‘3 “w. .. r max: .u i I Q LIBRARY 2307 Michigan State 3“: University This is to certify that the dissertation entitled ESSAYS ON HOW TO MITIGATE HOLD-UPS presented by FABIENNE MILLER has been accepted towards fulfillment of the requirements for the Ph.D. degree in Accounting and Information Systems Loam ¢ Mice Major Professor’s Signature 8/ /7 / Zoo? Date MSU is an affinnative-action, equal—opportunity employer l-I-l-l-.-I-I_LlA-_.—.v—I—‘~ —«-.-—o----a----..---t-n.-“I-tI-O-'-'- PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. MAY BE RECALLED with earlier due date if requested. DATEDUE DATEDUE DAIEDUE 6/07 p:/ClRC/DateDue.indd-p.1 ESSAYS ON HOW TO MITIGATE HOLD-UPS By F abienne Miller A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Accounting and Information Systems 2007 ABSTRACT ESSAYS ON HOW TO MITIGATE HOLD-UPS By Fabienne Miller The first essay examines whether aggregated cost information provided by suppliers to buyers can mitigate the hold-up problem that occurs with idiosyncratic investments. Hold-up implies that, given the opportunity, buyers pursue a self—interested strategy and do not reimburse suppliers for their idiosyncratic investment. In essence, buyers seek to maximize their own trade surplus at the expense of suppliers. However, evidence suggests some firms pursue a fair strategy. With such firms, research suggests that hold-up can be avoided. I propose and provide empirical evidence that the level of aggregation of the supplier-provided cost information (i.e., fine or coarse) will interact with buyer strategy (i.e., self-interested or fair) to affect investment in idiosyncratic assets and trade opportunism. Results show that coarse information leads to an increase in (no change in) self-interested (fair) buyers’ offers and in an increase in suppliers’ investments. Thus, suppliers are better off when they disclose coarse rather than fine cost information. Findings also suggest that buyers will benefit from requesting coarse instead of fine cost information from suppliers. The second essay presents a review of analytical and empirical hold-up literature that spans economics, accounting, finance, and supply chain by examining strategic decisions made by firms. Relationship-specific investments render parties vulnerable to potential opportunism in dyads between firms (e. g., organizational design and make or buy decisions, and inter-firm trade decisions), between divisions (e.g., intra-firm trade and transfer pricing decisions) and between managers and their firm (e. g., resource allocation decisions). While overall results of analytical and empirical literature show that, consistent with theoretical predictions, integration, contractual terms, allocation of property rights, incentives and relational contracting are effective at mitigating hold-ups, review of the research conducted suggests two significant limitations of past research (and resulting opportunities for additional investigations). First, empirical literature has largely focused on inter-firm decisions and analytical research has targeted intra-firm decisions, but, to a large extent, resource allocation research seems to assume away the idiosyncratic nature of the investment made by managers who acquire firm-specific project information. Second, considerations of firm strategy and of private information of the investor have remained until recently absent from analyses of hold-ups and warrant firrther development. Cepyright by FABIENNE MILLER 2007 ACKNOWLEDGMENTS Many hands, hearts, and minds have taken part in this dissertation and deserve recognition for helping me achieve my dream. I would like to thank my committee, Susan Haka, Ranjani Krishnan, Joan Lufi, and Alex von Eye, for their insightful comments and their counseling. I feel truly privileged to have had such a competent and caring committee. I would like to express my gratitude to Susan Haka, my committee chair, for caring about my development as a researcher, a teacher, and well-rotmded contributor to our profession. Her advice was invaluable and extends much beyond this document. Thank you to fellow doctoral students, especially Alex Woods and Christian Mastilak, for your encouragement and for the stimulating discussions. I also gratefully appreciate the financial assistance I received from the Department of Accounting. Finally, I would like to express my gratitude to my family for being flexible, supportive, patient, and not asking (too often) whether I was done yet. Special thanks to Marjorie for brightening my day with her art, Xavier for bringing some humor in my life, and Tiffany for reminding me of the importance of relationships. My husband, Keith, deserves the biggest thank you for supporting me in my belated career choice and for keeping me grounded. I thank you all for taking over household chores or waiting patiently for me to pick you up afier school. You all give my life purpose and joy and I could not have done it without you. “Un grand merci” to my parents and grandparents for teaching me the value of continuous learning and for letting me spread my wings many years ago. TABLE OF CONTENTS LIST OF TABLES ........................................................................................................... viii LIST OF FIGURES ........................................................................................................... ix 1 INTRODUCTION AND OVERVIEW ...................................................................... 1 l . 1 INTRODUCTION .............................................................................................. 1 1.2 OVERVIEW OF THE FIRST ESSAY ............................................................... 3 1.3 OVERVIEW OF SECOND ESSAY .................................................................. 4 2 INFORMATION AGGREGATION, STRATEGY, AND INVESTMENT DECISION IN BUYER-SUPPLIER RELATIONSHIPS .................................................. 6 2.1 INTRODUCTION .............................................................................................. 6 2.2 LITERATURE REVIEW ................................................................................. 10 2.2.1 Limits of Vertical Integration, Contracts and Reputation ......................... 11 2.2.2 Increasing Information Difference between Supplier and Buyer ............. 12 2.2.3 Trading with Buyer with Fair Firm Strategy ............................................ 16 2.2.4 Increasing Information Difference in the Presence of Different Firm Strategies ................................................................................................................... 19 2.3 HYPOTHESIS DEVELOPMENT .................................................................... 20 2.3.1 Trade Predictions — Hla-d ........................................................................ 25 2.3.2 Investment Predictions — H2 and H3a-d ................................................... 30 2.4 RESEARCH DESIGN ...................................................................................... 34 2.4.1 Experimental Design and Variables .......................................................... 34 2.4.2 Experimental Materials and Procedure ..................................................... 36 2.4.2.1 Trade Task ............................................................................................ 36 2.4.2.2 Investment Task ..................................................................................... 39 2.5 RESULTS ......................................................................................................... 40 2.5.1 Descriptive Statistics ................................................................................. 40 2.5.2 Manipulation Checks ................................................................................ 43 2.5.3 Hypotheses Tests ...................................................................................... 45 2. 5. 3. 1 Trade Decisions -- H1 a-d ..................................................................... 45 2.5.3.2 Investment Decisions -- H2 and H3a-d ................................................. 48 2.5.4 Supplementary Analysis ........................................................................... 51 2.6 DISCUSSION AND CONCLUSIONS ............................................................ 54 APPENDD( A ............................................................................................................... 57 APPENDIX B ............................................................................................................... 63 APPENDIX C ............................................................................................................... 69 APPENDD( D ............................................................................................................... 80 vi 3 REVIEW OF THE HOLD-UP LITERATURE IN THE CONTEXT OF STRATEGIC DECISION-MAKIN G ............................................................................. 136 3.1 INTRODUCTION .......................................................................................... 136 3.2 HOLD-UP AND OVERVIEW OF THEORIES ............................................. 139 3.2.1 Description of Hold-ups .......................................................................... 139 3.2.2 Transaction Cost Economics ................................................................... 142 3.2.3 Property Rights Theory ........................................................................... 144 3.2.4 Agency Theory ........................................................................................ 144 3.2.5 Relational Contracting Theory ................................................................ 145 3.3 STRATEGIC DECISION-MAKING ASSOCIATED WITH IDIOSYNCRATIC INVESTMENTS ......................................................................... 147 3.3.1 Organizational Design ............................................................................ 152 3.3.2 Trade Decisions ...................................................................................... 155 3. 3. 2. I Inter-firm Trade Decisions ................................................................. 156 3. 3. 2. I . 1 Analytical models .......................................................................... 156 3. 3. 2. 1.2 Archival empirical evidence ......................................................... 159 3. 3. 2. 1.3 Experimental empirical evidence .................................................. 161 3.3.2.2 Intra-firm Trade Decisions and Transfer Pricing .............................. 165 3. 3. 2. 2. I Contracting and Transfer Price Mechanisms ............................... 165 3. 3. 2. 2.2 Incentives ...................................................................................... 167 3.3.3 Resource Allocations .............................................................................. 168 3.4 CONCLUSIONS ............................................................................................. 172 REFERENCES ............................................................................................................... 174 vii LIST OF TABLES Table 2-1 Summary of Hla-d ........................................................................................... 30 Table 2-2 Sumary of I12 and H3a-d .................................................................................. 34 Table 2-3 Test of Subjects’ Comprehension of Experimental Setting ............................. 38 Table 2-4 Descriptive Statistics ........................................................................................ 41 Table 2-5 Manipulation Checks ........................................................................................ 44 Table 2-6 Tests of Hla-d .................................................................................................. 47 Table 2-7 Tests of H2 and H3a—d ................................................................................ ' ...... 4 9 Table 2-8 Sample Demographics ...................................................................................... 69 Table 2-9 Analysis of Pre-decision Questions for Comparisons between Conditions ..... 71 Table 2-10 Analysis of Decision-related Questions for Differences between Treatments72 Table 2-11 Post-decision Investigation of Trade and Investment Task Perception .......... 74 Table 2-12 Investigation of Decision-making Perceptions between Conditions .............. 78 Table 3-1 Analytical Papers and Hold-ups Mitigating Mechanisms ............................ 148 Table 3-2 Empirical Papers and Hold-ups Mitigating Mechanisms .............................. 150 viii LIST OF FIGURES Figure 2-1 Timeline of the Hold-up Problem ................................................................... 22 Figure 2-2 Investment and Trade Stages .......................................................................... 23 Figure 2-3 Predicted Outcomes under Various Information and Strategy Conditions ..... 26 Figure 2-4 Distribution of Buyer Offers ........................................................................... 42 Figure 2-5 Suppliers’ Mean Investment ........................................................................... 50 Figure 2-6 Evolution of Testing ........................................................................................ 63 Figure 3-1 Simplified Timeline of the Hold-up Problem ............................................... 140 ix 1 INTRODUCTION AND OVERVIEW 1.1 INTRODUCTION This dissertation investigates remedies to an important economic problem: hold- ups. Hold-ups occur in small numbers bargaining when a party does not make a socially optimal relationship-specific investment because he expects that, in an environment characterized by uncertainty, his bargaining counterpart will appropriate the return generated by the investment when trade ensues. In sum, expectation of ex post opportunism affects the firm’s ex ante investment decision negatively. The importance of the hold-up problem in the economy has been widely documented. However, since suboptimal investment is difficult to measure, researchers have argued that two types of evidence show that hold-ups are a problem: First, firms provide guarantees to potential customers they will not hold them up in the future; second, trade opportunism follows relationship-specific investments. Recent examples of the guarantees suppliers provide to encourage customers to invest in new technologies (i.e., make an idiosyncratic investment) can be found when suppliers give away their technology to competitors. When switching costs are high, customers might be hesitant to invest in technology because the supplier can hold them up by choosing prices or quality unilaterally (Farrell and Shapiro 1989). Consequently, voluntarily inviting competitors into the market by licensing a product to a competitor at low royalties, or by using an open architecture, provides evidence of innovative firms’ commitment not to behave opportunistically in the future (Shepard 1987; Farrell and Gallini 198 8). Examples of trade opportunism following a relationship-Specific investment can be found in the automobile industry (e.g., Klein 1988, 1996; Anderson, Glenn, and Sedatole 2000), in the railroad industry (e.g., Pittman 1991), and with franchises (e. g., Milgrom and Roberts 1992). More recently, Chang and Ive (2007) conducted a case study of the English Channel Tunnel (i.e., Chunnel). Spanning over six years, this construction project was characterized by high uncertainty. Chang and Ive find that one of the main contractors was able to threaten to delay the construction to extract compensation above and beyond the agreed amount as Eurotunnel’s switching costs increased as the result of sinking more funds in the project. Hold-up of Eurotunnel by the contractor led the Chairman of Eurottmnel to conclude soon after completion and settlement (Chang and Ive 2007, 9): “I have paid £100 million more than we think they’re entitled to, but it saves us a fortune in legal costs and means we don’t have friction in the future.” A stylized illustration of the hold-up problem follows. Suppose that an American automobile supplier must decide whether to invest in R&D that can lead to valuable improvements in a component of a specific model of automobile produced by one of the American Big Three original equipment manufacturers (OEM). R&D is costly and does not have any value outside the relationship with this American OEM.l The supplier’s R&D investment decision is based on expectations of payoffs from eventual purchase of the component by this American OEM. Additionally, contracts are incomplete and the l In this setting, R&D constitutes an idiosyncratic asset, i.e., relationship-specific asset. In the hold-up literature, acquisition of assets, increase in production eapacity, relocation of production, and investment in human capital have also been presented as examples of idiosyncratic assets. OEM cannot credibly commit to reimburse the supplier for the R&D investment.2 The supplier realizes that once the R&D investment is made, it becomes a sunk cost that may be ignored in trade negotiations with the OEM. In other words, the supplier anticipates that the American OEM will only pay for the component’s marginal production costs and not the sunk investment in R&D (i.e., self-interested strategy). Foreseeing the potential for the American OEM’s self-interested behavior during trade negotiations, the supplier under-invests in R&D in relation to the socially optimal investment (Henke 2006). Description of the hold-up problem assumes that the buyer has firll knowledge of the supplier’s marginal production costs (i.e., there is no private information related to marginal production costs between buyer and supplier) and that buyers are self-interested. Mechanisms aiming to mitigate hold-ups are presented in the following two essays. In the first essay, I relax the aforementioned two assumptions (namely, full information sharing and self-interest) and test aggregation of supplier-provided cost information and buyer firm strategy as mitigating mechanisms to hold-ups. In the second essay, I review the analytical and empirical literature on mitigating mechanisms to hold- ups. 1.2 OVERVIEW OF THE FIRST ESSAY Economics and accounting research has recently examined the role that information plays in buyer-supplier relationships, while supply chain research has focused on characteristics of the relationship (i.e., relational contracting). I build on those 2 It would be socially optimal for the buyer to commit to reimbursing the supplier for his investment cost. Yet, such a commitment might not be credible because renegotiation is possible. Alternatively, promises to reimburse investment costs may not be honored by buyers, or may lead to opportunistic behaviors by suppliers (instead of buyers) (Roberts 2004). literature streams to construct the first essay (i.e., Chapter 2). More specifically, I relax the full information sharing and self-interest assumptions traditionally made by hold-up research. First, based on recent investigation from economics and accounting regarding the role played by private information (cf. Arya et al. 2000; Gul 2001), I investigate how private information of the supplier (in the form of aggregated marginal production costs and investment cost) helps mitigate hold- ups in so far as it removes information that a self-interested buyer might use opportunistically during trade to appropriate the return on investment. Second, I build on the evidence from supply chain that some firms are motivated by strategies based in part on fairness and propose that fair firm strategies eliminate the hold-up problem. Finally, I examine how the level of aggregation of the supplier-provided cost information and the firm strategy of the buyer interact to affect the trade and investment dimensions of the hold-up problem. Results from experiments support the predictions that aggregated cost information and fair firm strategy reduce trade opportunism and, accordingly, encourage investment in idiosyncratic assets (thus, mitigating hold-ups). Moreover, findings from this research suggest that, contrary to the general belief that more detailed information is better, buyers are better off asking only for aggregated cost information as it encourages suppliers to make socially optimal idiosyncratic investments. 1.3 OVERVIEW OF SECOND ESSAY The hold-up problem has attracted considerable research directed at investigating remedies and the first essay fits in a long stream of research that originated in the 19703. The second essay (i.e., Chapter 3) provides a comprehensive review of this literature highlighting how the first essay complements the existing literature. In Chapter 3, I introduce the theories that have provided support for analyses of hold-ups and present a literature review of mitigating mechanisms to hold-ups that spans the fields of economics, accounting, finance, and supply chain. I highlight how Chapter 2 fits in the overall research stream related to hold-ups and suggest avenues for future development. This literature review is organized around the strategic decisions of choice of organizational design, trade decisions and resource allocations. While the empirical literature provides support for predictions fi'om transaction cost economics, property rights and agency theory, it has also recently offered evidence of the importance of relational contracting in buyer-supplier relationships. The review points to two limitations of past research. First, hold-up problems have been largely ignored in resource allocation research despite firm-specific investment made by managers who conduct capital project searches. Second, the role played by firm strategy in guiding behavior of individuals and the role of private information have been little investigated and constitute fruitful avenues for future research. 2 INFORMATION AGGREGATION, STRATEGY, AND INVESTMENT DECISION IN BUYER-SUPPLIER RELATIONSHIPS 2.1 INTRODUCTION Firms enter into collaborative arrangements that span their boundaries to improve the efficiency of their value chain. Often, a collaborative arrangement presents an opportunity for one firm to make an investment that only has value with a specific trading partner, i.e., an idiosyncratic investment. This investment creates a surplus that needs to be divided between the trading partners. Fair division of this surplus is an important concern of collaborating firms (Kumar 1996; Dekker 2003). However, the information sharing necessary to collaborate (Dyer 1997; Ittner et al. 1999; Seal et al. 1999; 2004; Cooper and Slagrnulder 2004) leaves firms vulnerable to the potential of self-interested behavior by their counterpart (Parkhe 1993; Roberts 2004; Anderson and Dekker 2005; Das 2006). More specifically, following an idiosyncratic investment by the supplier, a more powerful buyer might choose to appropriate the surplus generated by the investment, thus leaving the supplier worse off than before his investment. Anticipating the self-interested behavior of the buyer during trade, the seller chooses not to make the socially optimal investment (i.e., hold-up problem) (Williamson 1985). I examine, first, whether, in the context of supplier idiosyncratic investments, supplying aggregated instead of disaggregated cost information to the buyer can help curtail the self-interested behavior of the buyer during trade. If the supplier provides disaggregated information about the marginal production costs and the sunk investment cost, then an opportunistic buyer can divide the joint surplus in such a way that only the marginal production costs are reimbursed. However, if the supplier provides only aggregated costs by combining marginal and investment costs, the buyer will find it more difficult to pursue a self-interested strategy. As a result, the supplier should invest and the joint surplus increases. Second, I examine how the buyer’s strategy (i.e., buyer follows a self-interested or fair strategy)3 affects buyer trade offers and whether a signal sent to the supplier about the buyer’s strategy will increase the likelihood of investment. Hence, I address the following two research questions. First, does aggregated supplier-provided cost information mitigate the hold-up problem? Second, how does the level of aggregation of the supplier-provided cost information (i.e., fine or coarse) interact with buyer firm strategy (i.e., self-interested or fair) to affect buyer trade offers and supplier investment in idiosyncratic assets? I investigate two manners in which the hold-up problem might be mitigated. Irnportantly, since suppliers base their investment decisions on expectations of the payofl‘s they will derive from trade, both mitigating mechanisms aim to encourage investment in idiosyncratic assets by reducing the likelihood that buyers will behave opportunistically during trade. First, when the supplier cannot be confident ex ante that the buyer will not behave in a self-interested manner ex post, reducing the fineness of the information provided to the buyer will mitigate the risk of buyer opportunistic behavior during trade. I build on accounting research on aggregation4 as a way of limiting 3 1 define as fair a strategy that instructs buyers to reimburse the marginal production and investment costs incurred and to share the net surplus. Conversely, I define as self-interested a strategy that instructs buyers to only reimburse marginal production costs. I do not propose to investigate the motivations for choosing a fair or self-interested strategy. " Ijiri (1975, 109) suggests, "Generally, the reduction of n-dimensional data x = {x1, x2, ....,xn} into m- dimensional data y = {y1, y2, ...,ym} by a function y = f(x) is called aggregation when m < n." In the simple hold-up example I use in this study, aggregation reduces cost information from two dimensions (investment cost and marginal cost) to one dimension (total cost). opportunism (cf. Demski and F rimor 1999; Arya et al. 2000) and examine whether increasing the level of aggregation of supplier-provided cost information helps mitigate the hold-up problem. I predict that aggregation of supplier-provided cost information reduces the potential for self-interested behavior because aggregation removes information (namely, exact marginal production costs) that self-interested buyers might opportunistically use during trade. Thus, I propose that suppliers are more likely to invest and hold-ups are mitigated when aggregated instead of disaggregated cost information is provided to buyers. Second, practice suggests that the hold-up problem can be eliminated by trading with a partner who has a reputation for following a fair strategy instead of a self- interested strategy (Parkhe 1993). For instance, Henke (2006) provides practice evidence that American automobile suppliers who do not invest in R&D for American Original Equipment Manufacturers (OEM) invest in R&D for Japanese OEMs because they expect that the latter will reimburse suppliers for their investment and will share some of the profits generated by the investment. Accordingly, I examine how buyer strategy affects the trade (through its effect on buyer offer) and investment (via a signal to the supplier) dimensions of the hold-up problem. I predict that buyers with a fair (self- interested) strategy will cover (not cover) the suppliers’ marginal production costs and investment cost and will share (not share) the net surplus. Since suppliers anticipate the surplus they generate by investing will not be appropriated (be appropriated) by fair (self- interested) buyers, suppliers choose to make (do not choose to make) the socially optimal idiosyncratic investment. Thus, I propose that the hold-ups are avoided when suppliers believe buyers will follow a fair strategy. Additionally, I investigate how the level of aggregation of supplier-provided cost information (i.e., fine or coarse) interacts with the firm strategy of the buyer (i.e., self-interested or fair) to affect hold-ups. To investigate the questions of interest, first, I examine the effects of the level of aggregation of supplier-provided cost information and the buyer’s trade strategy on the buyer’s trade decision following an idiosyncratic investment by the supplier. The buyer learns the supplier’s investment cost and expected marginal production costs either disaggregated (i.e., fine information) or aggregated (i.e., coarse information) and his actions are guided by his firms strategy. An ultimatum game, with the buyer making a take-it-or-leave-it offer to the supplier, is used to divide the surplus generated by the investment. Second, the supplier receives a signal of the buyer’s strategy and learns the fineness of the cost information given to the buyer. Then, the supplier makes his investment decision. I find that supplying buyers with aggregated (disaggregated) cost information results in self-interested buyers making higher (the same) offers and suppliers investing more than (the same as) predicted by economic models. Accordingly, giving buyers aggregated cost information results in an increase of the joint surplus. Additionally, contrary to previous results fi'om studies investigating other forms of information asymmetry (e.g., Sloof, Oosterbeek, and Sonnemans 2007), I show that aggregation of cost information interacts positively with buyer strategies to affect trade offers and investment in idiosyncratic assets. My contribution is twofold. First, this dissertation makes a contribution to the accounting literature on incomplete contracting and opportunism in buyer and supplier relationships (e. g., Edlin and Reichelstein 1996; Baiman and Rajan 2002a; Drake and Haka 2007). While accounting research typically suggests that obtaining more detailed information increases firm profits by reducing moral hazard and by improving the decision-making process, I suggest that aggregated cost information (in other words less detailed information) increases the joint surplus (via investment in idiosyncratic assets). Second, this dissertation contributes to the accounting literature on aggregation by highlighting and isolating the effects of buyer strategy (i.e., self-interested or fair) and by documenting an interaction between buyer strategy and levels of aggregation of supplier- provided cost information (i.e., fine and coarse) when inter-firm hold-up is of concern. The remainder of this chapter is structured in five sections. Following the introduction, Section 2.2, a literature review, introduces the various mitigating mechanisms that have been investigated highlighting their limitations. Out of those mitigating mechanisms, I focus on buyer-supplier information difference and trading with buyers with a reputation for fairness to conclude the literature review. The third section develops hypotheses. Section 2.4 presents the research design. Experimental methods and results are described in Section 2.5. The dissertation concludes with a discussion, limitations of this research and suggestions for future development. 2.2 LITERATURE REVIEW This section presents, first, the various mitigating mechanisms of the hold-up problem that have been investigated by economics, accounting and marketing, highlighting their limitations. Second, I suggest how investigation of aggregation of supplier-provided cost information and of buyer strategy can help inform the search for a remedy to hold-up problems. The hold-up problem presented by transaction cost economics theory is based on 10 the premise that, given the right circumstances and buyer’s lack of commitment not to behave opportunistically, self-interested behavior is likely to occur during the trade that follows an idiosyncratic investment (i.e., trade opportunism) (Williamson 1975, 1985). Irnportantly, all the mechanisms discussed below have been proposed as a means to encourage idiosyncratic investment by reducing the likelihood that buyers behave opportunistically during trade. 2.2.1 Limits of Vertical Integration, Contracts and Reputation Vertical integration (e.g., Heide and John 1990; Baiman and Rajan 1995; Anderson, Glenn, and Sedatole 2000), contracts (e.g., Klein 1980; Joskow 1985; Edlin and Reichelstein 1995, 1996; Che and Hausch 1999; Baiman and Rajan 2002a), and reputation (e.g., Anderson and Weitz 1992) have been presented as potential ways to encourage investment by protecting investors’ return on investment against the self- interested behavior of the buyer. Although the protection mechanisms discussed above can be effective in many circumstances, they also suffer from limitations. For instance, the costs of formal mechanisms such as vertical integration and contracts might exceed their potential benefits. Furthermore, vertical integration is not always possible and hold-ups can also exist between divisions of a same firm (cf. Baldenius 2000). Moreover, contracts are inherently incomplete and cannot always offer suppliers full protection from buyers’ self- interested behavior. Finally, informal mechanisms such as reputation are not always effective either. Attributing responsibility for self-interested behavior might be difficult and potential partners might not believe that past self-interested behavior is predictive of future behavior (Bensaou and Anderson 1999). Consequently, self-interested behavior 11 might be observed and return on investment might be appropriated by buyers without affecting the buyers’ reputation. Since vertical integration, contracts, or reputation do not offer full protection fiom hold-ups, economics literature has recently turned to increasing the information difference between supplier and buyer as a mechanism to alleviate trade opportunism, and consequently encourage idiosyncratic investments.5 2.2.2 Increasing Information Difi'erence between Supplier and Buyer To limit the potential for trade opportunism by buyers, Gul proposes that suppliers can augment their bargaining power by increasing the information difference between supplier and buyer instead of disclosing all the supplier’s information (2001). This can be achieved in two ways. It can be accomplished, first, by not communicating whether an idiosyncratic investment was made, or, second, by supplying buyers with aggregated information. Truthful reporting by the supplier is assumed. Both types of information asymmetry are detailed below. Building on Tirole’s investigation of unobservable investment (1986), Gul (2001) presents a model where the supplier makes his investment decision so as to maximize the profits he will derive fi'om trade. The supplier does not communicate to the buyer whether he made any idiosyncratic investment. Thus, the supplier possesses private information about the size of the surplus and can extract information rents during trade (i.e., the buyer is unable to appropriate the total surplus generated by the investment and trade opportunism is limited). Gul concludes that, compared to the perfect information 5 Irnportantly, as with other mitigating mechanisms, increasing the information difference is not always effective. For instance, buyers might have other means of obtaining suppliers’ private information or increasing the information difference might result in loss of benefits that could be derived from knowing the suppliers’ marginal production costs. 12 condition, investment inefficiency6 is reduced when the supplier’s investment decision is unobserved by the buyer. Conversely, unless negotiation is costless, trade inefficiency7 increases because, since the buyer does not know whether an investment was made, he does not know the supplier’s reservation price. As a result, the supplier does not benefit from his private information and overall efficiency does not improve. However, Gul suggests that keeping other information private might solve the hold-up problem. Some information can be kept private through aggregation of information (Baiman 1975). Baiman and Rajan’s (2002b) and Arya et al.’s (2000) models show that, by curtailing opportunistic behavior during trade, aggregation of supplier-provided information can serve as a potential mitigating mechanism to hold-ups. Baiman and Rajan (2002b) model the effects of disclosing know-how in a buyer- supplier network. A buyer makes a non-relationship—specific R&D investment decision and chooses the level of disclosure of the resulting innovation to the supplier. Disclosure of innovation to the supplier helps increase the size of the joint surplus by revealing a new technology that increases the quality of the product, but also facilitates misappropriation of that innovation by the supplier who could sell the improved product directly to the outside market. Baiman and Rajan conclude that disclosing the coarse know-how generated by the innovation is optimal for certain cases of innovation realization because aggregated know-how reduces the potential for self-interested 6 In this paper, investment efficiency occurs when the supplier makes an investment that maximizes the potential smplus from trade net of investment costs. 7 In this paper, trade inefficiency occurs when negotiations breakdown because only one round of negotiation is conducted. Conversely, trade is efficient if trade occurs (i.e., a joint benefit to the dyad is created and at least one party benefits). Alternatively, lengthy negotiations are another form of trade inefficiency when several rounds of negotiations are conducted. 13 behavior by the supplier while allowing the supplier to improve the quality of the product. Irnportantly, Baiman and Rajan’s paper does not address whether aggregation of information would also be effective to encourage investment in idiosyncratic assets. Additionally, it is unclear whether coarsening information would encourage idiosyncratic investment when cost information, instead of know-how, is provided. Expanding on prior literature on coarse information as a comrrritrnent device between principal and agent, Arya et al. (2000) examine the effects of aggregated accounting information on investment in idiosyncratic assets in a capital budgeting and monitoring context. In this setting, an agent needs to make an idiosyncratic investment in project search for the principal. An information system that provides only coarse and late information restricts the information available to the principal; thus, it can serve as the principal’s commitment not to appropriate, ex post, the benefits generated by the agent’s investment in project search. As a result, the agent is motivated to increase his search for a profitable project (i.e., make an idiosyncratic investment) by the prospect of obtaining budgetary slack. In other words, in a capital budgeting setting, aggregation of monitoring information reduces the potential for opportunism by the principal and, as a result, strengthens the agent’s incentives to make an idiosyncratic investment. Drake and Haka (2007) examine the effect that the information system (fine with Activity Based Costing or coarse with Volume Based Costing) can have on buyers and suppliers’ propensity to share cost information under different market conditions. They find that subjects with fine cost information are less likely than subject with coarse cost information to share their information with their counterpart even though such information-sharing could benefit both parties. Drake and Haka propose that subjects’ 14 information-Sharing decisions are motivated by inequity aversion and fear of being held- up during trade. Thus, an information system that provides coarse cost information encourages sharing of information by reducing the risks the other party will appropriate trade efficiencies. Extant literature suggests that aggregated information may help mitigate hold-ups because aggregation removes information (namely, exact marginal production costs) that self-interested buyers might opportunistically use during trade. My dissertation differs from Baiman and Rajan’s research (2002b) in so far as I examine aggregation of cost information in the context of a relationship-specific investment8 instead of aggregation of know-how. Additionally, I expect to provide different insights from Arya et al. (2000) because the purpose of the transaction I investigate is trade between buyer and supplier instead of monitoring of project search effort. Finally, whereas Baiman and Rajan and Arya et al. assume self-interested behavior, I expand the analysis to include strategies that induce fair behavior. My dissertation also differs from Drake and Haka (2007) for two main reasons. First, the focus of my study is the two stages of the hold-up problem (investment and trade) whereas the choice to share and its impact on trade efficiency is of interest for Drake and Haka. Additionally, I manipulate the buyers’ behavior via their firm strategy. Accordingly, I can assess whether aggregation of cost information affects the buyers who follow a fair strategy differently from those who follow a self-interested strategy. 8 Such aggregation is likely to be possible when idiosyncratic investments take the form of investments in human capital or R&D. With these types of investment, it might be too costly for the information system to disaggregate the marginal and investment cost information. Hence, even a powerful buyer might not be able to obtain disaggregated cost information. It would, however, be more difficult if the supplier invested in relationship-specific equipment. 15 2.2.3 Trading with Buyer with Fair Firm Strategy In contrast to the underlying pursuit of self-interest assumption necessary for opportunistic behavior (Williamson 1985), trading with fair buyers has been proposed as a way to avoid hold-ups. Two streams of literature have investigated the effect of fairness on hold-ups. Supply chain literature, the first stream of research, associates fair strategy with sharing of the benefits created by the relationship (e.g., Ring and Van De Ven 1992; Kumar 1996; Bensaou 1999; Dyer 1997; Dekker 2003; Langfield-Smith and Smith 2003; Cooper and Slagrnulder 2004), but does not investigate the motivation for such strategic choices. Sharing of the benefits created by the relationship (e.g., cost savings or improvement in product quality) is made possible by the multiperiod nature of the exchanges between the firms. Supply chain literature observes that, whereas some firms pursue strategies that are arm’s length in nature and focus on purchasing products for the lowest price possible by only reimbursing marginal production costs (i.e., self-interested strategy), others pursue more collaborative strategies (Bensaou and Venkatraman 1995; Bensaou 1999; D02 and Baburoglu 2000) characterized in part by considerations of fairness (i.e., fair strategy) (Helper 1991; Carr and Ng 1995; Kumar 1996; Bensaou and Anderson 1999; Cooper and Slagrnulder 2004; Henke 2006). Fair strategies can lead to different outcomes depending on reference points and industry norms (Kahneman, Knetsch, and Thaler 1986), but, in general, a buyer with fair strategies will reimburse suppliers for their marginal production costs and past investment cost, and share some of surplus created by the investment (of. Kumar 1996; Sako 2004). The extent of such fair strategies in the economy has, however, not yet been determined. Experimental economics, the second stream of research, examines the behavior of 16 individuals and attempts to distinguish the motivation behind fair behaviors (e.g., social preferences such as reciprocity, inequity aversion, or altruism). Experimental economics goes as far as queStioning the seriousness of the hold-up problem because of the extent of observed fair behavior in the laboratory. Experimental economics researchers (e.g., Berg, Dickhaut, and McCabe 1995; Ellingsen and Johannesson 2004a; 2004b; 2005) observe that, in contradiction with economic predictions, investment in relationship-specific assets is significantly greater than zero (i.e., 16% to 35% of participants invest at the socially optimal level) and offers fiom the party who benefited from the investment are greater than zero. Based on the offers they observe, these researchers conclude that some individuals intend to reciprocate9 and that expectations of fairness play a role in their counterparts’ investment decisions. Yet, as pointed out by Roth and Murninghan (1982) and Ochs and Roth (1989), it is likely that such sharing of the surplus created by the investment is motivated by one of two factors: the self-interest of the individual who wants to make an offer that is likely to be accepted or the individual’ 8 actual preferences for fairness. Those studies suffer from two important limitations. First, those studies are unable to distinguish between self— interest and actual preference for fairness as the studies do not manipulate buyer fairness. Second, they study the behavior of individuals and abstract from the fact that firms attempt to control the behavior of their employees through corporate culture and strategies. Strategic management research investigates the relation between the guidance 9 Alternatively, buyers might have offered more than zero and covered the investment cost not for fairness reasons, but because they suffer from the sunk cost bias, intend to signal competitors, or have other strategic reasons (Parayre 1995; Diekmann et al. 1996; Troeger 2002). 17 provided by the firm (i.e., culture, values and strategy) and the actions of the firm’s managers. Research has focused on the creation and formulation of strategies, the congruence between individual and firm values (cf. Kristof 1996) and the association between strategy, top executives characteristics and firm performance (e.g., Thomas, Litschert, and Rarnaswarny 1991). Several studies propose that strategies guide actions of managers (e. g., Floyd and Wooldridge 1992; Bourgeois and Brodwin 1984). Additionally, research proposes that for such guidance to be effective, the strategy needs to be communicated clearly, and should be associated with performance measurement and incentive compensation (Kaplan and Norton 1996). For instance, Kumar (1996) finds that Procter & Gamble motivate their managers to follow a strategy of working closely with distributors by compensating managers based on the profits of Procter & Gamble and of the distributors. Accounting literature has focused its investigation on the relationship between accounting systems and generic firm strategies such as those presented by Gupta and Govindarajan (1984) or Porter (Porter 1985). Research has pointed to the importance of adaptability of the accounting systems to strategies (Langfield-Smith 1997) and to the life cycle of the firm (Granlund and Taipaleenrnaki 2005), but little focus has been given to more specific firm-wide strategies such as strategies for dealing with suppliers. Thus, we know little about how strategies might interact with accounting system characteristics to affect individual decision-making and behavior. Still, Liedtka (1989) finds that most managers surveyed comply with a strategy even when they experience some conflict with their values provided the strategy is strong. Thus, there is some evidence that a strong firm strategy that is clearly 18 communicated might induce employees to follow the firm strategy instead of their own preferences. As a result, when an employee, who has individual preferences for fairness (self-interest), is guided by a self-interested (fair) strategy, it is likely that he will follow the guidance provided by the firm instead of his own preference. Consequently, one cannot conclude, as suggested by Ellingsen and Johannesson (2004), that fairness preferences of some individuals makes the hold-up problem a rare occurrence. In this dissertation, I present evidence that a clearly communicated firm strategy can induce subjects to behave in a self-interested or fair manner in an experimental setting resulting in exacerbating or mitigating hold-ups. 2.2.4 Increasing Information Difference in the Presence of Different Firm Strategies Building on extant experimental economics research, Sloof et al. (2007) test Gul’s finding that private information in the form of unobservable investment boosts investment efficiency. Sloof et al. examine the interaction between imperfect information (i.e., unobservable investment), and fair behavior of individuals. Their results are consistent with Gul’s model when investment cost is high and there is little room for fairness (i.e., suppliers are more likely to invest when investment is unobservable than when it is observable). Sloof et al. show, however, that although individuals’ fairness mitigates hold-ups when investment is observable, the positive effect that buyer fairness has on the hold-up problem is negated when investment is unobservable and investment cost is low. Sloof et al. point out that buyers do not know how much suppliers gave up in order to create the gross surplus when investment is unobservable. Consequently, fair buyers are unable to reimburse suppliers for the investment they made or to share the 19 surplus when investment is unobservable. Irnportantly, Sloof et al. do not provide measures of buyer’s fairness. Instead, holding the gross surplus fixed, they propose that fair (self-interested) behavior can be expected when there is a large (small) net surplus to share, i.e., small (large) investment for a fixed surplus. They also suggest suppliers might invest (not invest) because they expect reciprocity (self-interested behavior) fiom buyers. Alternatively, considerations of risk and returns could also explain why suppliers refiain fiom investing when investment costs are high. Findings presented by Baiman and Rajan (2002) and Arya et al. (2000) lead me to propose that, with buyer-supplier relationships, an information difference that arises from aggregated supplier-provided cost information might reduce the likelihood of buyer self- interested behavior, hence encouraging idiosyncratic investment. Additionally, observations from behavioral economics (cf. Sloof, Oosterbeek, and Sonnemans 2007) suggest that the effect of aggregation of supplier-provided cost information (i.e., fine and coarse) on trade and investment is likely to depend on buyer strategy (i.e., self-interested or fair). 2.3 HYPOTHESIS DEVELOPMENT In this section, I develop the intuition behind the predictions first for the trade decision (as expectations from trade drive investment decisions), then for the investment decision. A more formal analysis can be found in Appendix A. Important assumptions follow. First, I examine a one-period transaction and do not take into account reputation considerations. Irnportantly, one-period games are appropriate to examine hold-ups because, in many alliances, one partner has a short-term 20 horizon (Das 2006) and because, as a result of technology or environmental uncertainty, it is often difficult to assess whether future benefits will derive from current transactions. Second, I assume that the buyer-supplier contract is incomplete and renegotiation cannot be avoided. Third, buyers have greater bargaining power than suppliers and buyers are risk averse. The following two assumptions are not central to this analysis, but are made for expositional purposes. Suppliers are assumed to be self-interested, i.e., to accept any offer that at least covers marginal costs. This simplifying assumption was made because informing buyers that they will be dealing with self-interested suppliers removes their strategic uncertainty about what constitutes an offer that will be acceptable to the supplier.10 Thus, knowing that they are dealing with self-interested suppliers assures buyers that offers that cover marginal production costs will not be rejected. As a result, buyers who offer to reimburse marginal production costs and investment cost and to share the net surplus truly intend to follow the fair strategy and are not motivated by the expectation that offering to reimburse costs and to share the net surplus is necessary to avoid rejection of their offer. Finally, consistent with other experimental and analytical research on hold-ups, neither supplier, nor buyer has an outside option should disagreement about trade occur. The timeline illustrating the various stages of the hold-up problem is detailed in Figure 2-1. lo Brenner and Vriend (2006, 629) provide empirical evidence that proposers in ultimatum games are unable to learn their part of the subgame perfect equilibrium even after 100 iterations “unless the players. ..behave exactly as in the subgame perfect equilibrium without ever rejecting any offer.” 21 Figure 2-1 Timeline of the Hold-up Problem Stage 1 Stage 2 0 1 2 3 l I I I Incomplete Supplier Buyer learns Trade contract signed investment supplier costs decision decision Adapted from Hart and Moore (1988) Figure 2-1 shows a contract is negotiated at time 0 for a predetermined quantity of a product. The product price cannot be specified with certainty ex ante because it depends on the resolution of certain parameters that cannot be predicted (e. g., success of R&D efforts). At time 1, the supplier must decide whether to make a non-contractible idiosyncratic one-period investment that will create a potential surplus from trade. Time 1 represents the investment stage. The supplier is motivated to maximize his payoff from trade (i.e., buyer’s offer less marginal and investment costs) and his investment decision is based on expectations of his share of the surplus from trade to be conducted at time 3. In other words, the supplier uses backward induction to make the investment decision. The buyer learns the supplier’s cost information at time 2. An ultimatum game,11 where the buyer makes a take-it-or-leave-it offer, is used to divide the net surplus from trade at 11 Alternatively, Nash demand games (e.g., Gantrrer, Gueth, and Koenigstein 2001), dictator games (e.g., Sloof, Oosterbeek, and Sonnemans 2007) and infinite horizon games with forced breakdown (e.g., Hackett 1994) have been used to model the trade stage of the hold-up problem. 22 time 3. Time 3 represents the trade stage. Although ultimatum games are parsimonious representations of the negotiation process (Gueth, Schmittberger, and Schwarze 1982), they capture the bargaining power differential between the buyer and supplier, an important assumption of this study. The various outcomes of this two-stage game are detailed below and shown in Figure 2-2. Figure 2-2 Investment and Trade Stages Trade stage Respective payoffs of buyer and supplier Supplier (G-T, T-M-F) accepts Investment stage Supplier invests T Supplier (0, -F) rejects Supplier does not invest ------— ----——---——-— ——-----—----—————— (0,0)* Where F = Investment cost T = Buyer trade offer G = Gross surplus from trade such that G > M + F M= Marginal production costs * While most papers (e.g., Berg et al. 1995; Ellingsen and Johannesson. 2004a, 2004b; Cox 2004) assume that trade doe not take place if the supplier does not invest, a few papers (e.g., Sloof et a1. 2007) assume that trade occurs when the supplier does not invest and that a smaller surplus is divided. 23 Let G represent the gross surplus generated by the supplier’s idiosyncratic investment (i.e., the surplus from trade), M the predicted marginal production costs, and F the supplier’s investment cost. G > M + F. In other words, investment is socially efficient Let I represent the supplier’s investment decision: I = 1 if investment takes place, I = 0 otherwise. X, represents the net monetary payoffs of player i. At time 3, the surplus fiom trade is as follows: G if I = 1 0 if I = 0 Let T equal the trade offer of the buyer to the supplier. I normalize the costs incurred by the buyer (in addition to the offer he makes to the supplier) to zero. The supplier and buyer monetary payoffs are respectively XS and Xb with: r T - F - M if I = l and supplier accepts buyer’s offer Xs= t - F if I = 1 and supplier rejects buyer’s offer ( 0 if I = 0, and Xb= G - T if I = l and supplier accepts buyer’s offer 0 otherwise I assume that firms follow a self-interested or a fair strategy. I define a self- interested strategy as a strategy that instructs buyers to cover only marginal production costs and a fair strategy as a strategy that instructs buyers to reimburse supplier for marginal production costs and investment cost, and to share the net surplus created by the investment. Firm strategies are instituted to guide the actions of individual decision makers. Accordingly, a self-interested (fair) strategy should lead to self-interested (fair) 24 behavior. Thus, I conduct the analysis at the individual level. Recall that supplier’s expectation of self-interested behavior by the buyer during trade drives the hold-up problem. This suggests that reducing the likelihood that the buyer behaves in a self-interested manner signals that the supplier will be reimbursed for his idiosyncratic investment. As a result, the supplier should invest. I examine two ways of mitigating hold-ups (namely, providing buyers with aggregated supplier-provided cost information and trading with buyers with fair firm strategy) and how they interact to affect buyer trade offers and supplier investment decisions. 2.3.1 Trade Predictions — Illa-d First, I examine the effect of the level of aggregation of supplier-provided cost information. Williamson’s (1985) predictions suggest that the self-interested buyer who knows the supplier’s marginal production costs, M, and past idiosyncratic investment cost, F, (i.e., fine information) will offer just enough to cover marginal production costs plus a” (or M + e) as presented in Figme 2-3 under item A. below. Anticipating this, the supplier will not make the socially optimal idiosyncratic investment. 12 e is an immaterial amotmt offered so that a self-interested supplier is not indifferent between accepting and rejecting the offer. 25 b55325 com «dogma Fe @555 ,3 3262 m mo 03.? vetoes mLoEm u m “85:8 3588:: n m 380 eouozuoa 33334 NE ouch 8on 5:93 v.85 n 0 Eco Segmgfi u m 22:5 6 .9 “met: “on moot com—magi 38w: comma—am m $32: Q- .8 3:35 \2 + S + 3 ”aeomaoo d S: - S - no .QE - E - Q9 d 0 Q2 + S + so ”a”: 8E .o S: - S - so .QE - S - NE .0 a + a - A: + e “emanatcsaamamaoe .m a. + a t a. t a + x2 + e - 9 .m sass .... + 2 necessaeae 8E .< Q + m .. .w - 2 - g .< cow—n55 “202280 anameozmanE wage—Em 05 Hope: cube 3er anthem which .3536 «each. .8386 «5:38.»:— aeezmcaeu 38.25 ...—a zeta—Eon:— mseta> .335 883.59 33:3...— 3 25e— 26 I propose that a self-interested buyer who possesses coarse cost information c in the form of the sum of the supplier’s marginal production costs and past idiosyncratic investment cost (i.e., (F + M)c) will offer more than a self-interested buyer who possesses fine cost information. Knowing that a self-interested supplier has an acceptance threshold of M, a self-interested buyer would like to offer M + s, but he has no prior information about the supplier’s true marginal production costs, M, or investment cost, F. The buyer wants to make a low offer so as to maximize his payoffs, but, being risk averse, does not want to risk losing his share of the surplus from trade (i.e., G - T) by making an offer that does not cover marginal production costs. In essence, uncertainty about the true marginal production costs and fear of losing the surplus from trade will cause a self-interested buyer to include a substantial portion of the cost of the supplier’s past investment in his ofler and offer close to (F + me. In sum, the information loss associated with coarse information limits the buyer’s potential self-interested behavior; thus, resulting in higher offers than fine information. Thus, in the coarse/self-interested condition, the buyer’s predicted offer is (F + M), - 6 + 8'3 with (F + M)c - 6 > M as shown in Figure 2-3 item B. above and detailed in Appendix A. Second, I examine the effect of buyer strategy on the trade decision. Fehr and Schmidt’s inequity aversion theory (1999) provides the fiamework to predict behaviors of buyers. Carmichael and MacLeod (2003), Ellingsen and Johannesson (2004b), and Ewerhart (2006) suggest that considerations of fairness lead some buyers to include investment cost in their payoff calculation and, as a result, in their offer calculation. In l3 5 represents the buyer’s expected value of F reduced by the adjustment the buyer makes for his uncertainty. The higher the buyer’s uncertainty about M and F’s true values, and the higher the potential surplus from trade, the lower 8. 27 other words, fair buyers reimburse marginal production costs and investment cost, and share some of the net surplus fiom trade. Thus, buyers who follow a fair strategy do not need to identify investment cost, F, and marginal costs, M, separately. Consequently, offers of buyers assigned a fair strategy should not be affected by whether investment cost and marginal production costs are disclosed disaggregated (in the form of F and M) or aggregated (in the form (F + M)c). As a result, the offers made by buyers with fair firm strategy who have supplier-provided coarse cost information are not expected to be significantly different fi'orn the offers made by buyers who possess fine cost information. As presented in Figure 23 (items C. and D.) and detailed in Appendix A, offers are predicted to be G/2 + 17/2 + M/2.“ Accordingly, contrary to the case of unobservable investment presented above in Section 2.2. (of. Sloof et al. 2007), information asymmetry in the form of aggregated cost information is not expected to lead to greater trade inefficiency than fine information. In sum, I propose that buyers will make the following offers as shown in Figure 2-3: (i) With fine information, buyers assigned a self-interested strategy will offer M + 6, whereas buyers assigned a fair strategy will offer up to G/2 + F/2 + M/2. (ii) With coarse information, buyers assigned a self-interested strategy will offer (F + M)c - 6 + c with (F + M)c - 6 > M, whereas buyers assignedafair strategy will offer up to G/2 + F/2 + M/2. As detailed above, an ordinal interaction between level of aggregation of supplier- provided cost information and buyer strategy is predicted. Thus, I propose that coarse 1“ G/‘2+F/2+M/2=M+F+(G-M-F)/2. Inotherwords, itisequivalenttoreimbursingmarginal production costs and investment costs and sharing the net surplus. 28 information increases the offers buyers who follow a self-interested strategy, but does not change the offers of buyers who follow a fair strategy. This prediction leads to the following hypotheses: Hla: An interaction between level of aggregation of supplier-provided cost information and assigned buyer strategy is predicted to affect buyers’ offers. The specific predictions associated with this interaction are detailed in the hypotheses below. Hlb: When buyers are assigned a self-interested strategy, buyers who possess coarse cost information make, on average, higher offers than buyers with fine cost information. ch: When buyers are assigned a fair strategy, the offers of buyers do not change based on the level of aggregation of the cost information they possess. Hld: When buyers possess coarse cost information, the mean offer of buyers assigned a fair strategy is greater than that of buyers assigned a self-interested strategy. Hypotheses Hla-d are summarized in Table 2-1. 29 Table 2-1 Summary of Hla-d [Buyer trade offer Self-interested strategy Fair strategy [Fine information A B anrse information C D Predictions: Hla: Information x Strategy Hlb: Coarse information > Fine information if Self-interested strategy C > A ch: Coarse information == Fine information if Fair strategy B = D Hld: Fair strategy > Self-interested strategy if Coarse information D > C In sum, providing buyers with coarse instead of fine cost information can serve, for the supplier, as a protection against the buyer’s potential bade opportunism. H1 a-d suggest that suppliers who are somewhat uncertain about whether they will be dealing with a buyer who has a fair or self-interested firm strategy are better off disclosing only coarse cost information. Recall that the hold-up problem has two dimensions: investment and trade. The effect of the supplier’s knowledge of buyer strategy and of the level of aggregation of the supplier-provided cost information on the supplier’s investment decision is examined below. 2.3.2 Investment Predictions — H2 and H3a-d Expectations of buyer self-interested behavior during the trade stage drive under- investment by suppliers. Therefore, I predict that reducing the likelihood that buyers behave in a self-interested manner will lead suppliers to invest, thus increasing investment efficiency. Suppose that, since buyers cannot commit to not behave 30 opportunistically, suppliers do not know the buyers’ strategy (i.e., self-interested or fair) with certainty, but can assign probabilities to each strategy. The expected utility of self- interested suppliers includes the expected offer net of marginal production costs and idiosyncratic investment cost. (i) When the buyer possesses fine cost information, the supplier expects to receive M + s from a buyer assigned a self-interested strategy, and up to G/2 + F/2 + M/2 from a buyer assigned a fair strategy (see Figure 2-3). The supplier will invest if his expected utility is positive. I predict that that the higher the probability of the buyer strategy being self-interested and the higher the investment cost, the less likely it is the supplier will invest. (ii) When the buyer possesses coarse cost information, the supplier expects to receive (F + M) - 6 + a from a buyer assigned a self-interested strategy, and up to G/2 + F/2 + M/2 from a buyer assigned a fair strategy (see Figure 2-3). Since 8 is predicted to be significantly smaller than the supplier’s investment because of the risk aversion of the buyer, the expected utility of a supplier in the coarse information condition is greater than the expected utility of a supplier in the fine information condition. Consequently, the supplier is more likely to invest in the coarse than in the fine information condition. H2: Suppliers are more likely to make idiosyncratic investments when buyers possess supplier-provided coarse cost information than when buyers possess fine cost information. An example illustrates supplier’s expected utility calculations and the above predictions. Assume the supplier believes there is a 75% chance that the buyer is assigned a self-interested strategy and surplus from trade, G, equalsl70 for an idiosyncratic 31 investment, F, of 40 and marginal production costs, M, of 10. With fine cost information, supplier’s expected utility calculation is as follows: .25 * (170/2 + 10/2 + 40/2) + .75 * 10 - 40 - 10 = -15. Since the supplier’s expected utility is negative, he will not invest. However, with coarse cost information, supplier’s expected utility calculation is as follows: .25 * (170/2 + 10/2 + 40/2) + .75 * (10 + 40 - 8) - 40 -10 =15 - .755. The supplier’s expected utility will be positive if 8 is smaller than 20. Hence, the supplier invests if he expects that the offer made by the buyer will be greater than 30 (i.e., 50 - 20). The supplier should invest because, as a result of the buyer’s risk aversion, 8 is predicted to be small. Additionally, Hlb-c suggest that increasing the level of aggregation of supplier- provided cost information results in an increase in the offers made by buyers with self- interested strategies, but does not change the offers of buyers with fair strategies. Since expectations of trade behavior drive suppliers’ investment decisions, an interaction between level of aggregation of supplier-provided cost information and buyer strategy is also predicted in relation to investment. H3a—d follow. H3a: An interaction between level of aggregation of supplier-provided cost information and buyer strategy is predicted to affect supplier’s investment in idiosyncratic assets. The specific predictions associated with this interaction are detailed in the hypotheses below. H3b: When suppliers know they are likely to be matched with a buyer who is 32 assigned a self-interested strategy, increasing the level of aggregation of the cost information provided to buyers will, on average, result in significantly larger supplier investment. H3c: When suppliers know they are likely to be matched with a buyer who is assigned a fair strategy, supplier’s mean investment does not change based on the level of aggregation of the cost information buyers possess. Irnportantly, Hld predicts buyers with a fair firm strategy will make higher offers than buyers with a self-interested firm strategy when buyers possess coarse cost information. Because of this, I predict that buyers with coarse information cover a substantial portion of the marginal production costs and investment costs.15 Accordingly, suppliers matched with buyers in the self-interested/coarse condition do not expect that buyers will be able to appropriate the entire surplus generated by their investment. It follows that suppliers in the self-interested/coarse condition are likely to invest. H3d follows. H3d: When buyers possess coarse cost information, the mean investment of suppliers who know they have a higher probability of being matched with buyers who are assigned a fair strategy is equal to the mean investment of suppliers who have a higher probability of being matched with buyers who are assigned a self- interested strategy. Hypotheses H2 and H3a-d are summarized below in Table 2-2. 15 In the coarse/self-interested condition, the buyer’s predicted offer is (F + M)c - 8 + s as shown in Figure 23 item B. 33 Table 2-2 Summary of H2 and H3a-d Fupplier investment Self-interested strategy Fair strategy [Fine information A B Foarse information C D Predictions: H2: Coarse information > Fine information C + D > A + B H3a: Information x Strategy H3b: Coarse information > Fine information if Self-interested strategy C > A H3c: Coarse information = Fine information if Fair strategy B = D H3d: Fair strategy = Self-interested strategy if Coarse information D = C 2.4 RESEARCH DESIGN Following protocol studies, pre-pilot tests, and pilot tests,16 106 students (88 MBAs, two Master students and 16 undergraduate students) were recruited on a voluntary basis from managerial accounting classes (see Table 2-8 in Appendix C for additional descriptive statistics of the participants). Participants received performance contingent compensation in addition to a $5 participation fee. Overall payoffs ranged from $5 to $15 with an average compensation of $10 for a thirty-minute session. 2.4.1 Experimental Design and Variables A 2 x 2 between-subject design was used for the investment and trade tasks. Two separate experiments were conducted, one for buyer subjects in the trade task and the other for the supplier’s investment decision. Two levels of aggregation of cost '6 See Appendix B for more details on the pilot tests and the evolution of the experiment. 34 information possessed by the buyer (viz. coarse or fine information) and two levels of buyer strategy (viz. self-interested or fair) were used. In the fine cost information condition, buyers knew the marginal production costs and the past idiosyncratic investment cost of the suppliers separately. Coarse cost information was operationalized by giving buyers only the sum of the suppliers’ marginal production costs and past idiosyncratic investment cost. Firm purchasing strategy provided the means for manipulating buyer strategy. Given that firms employ individuals of varying types ranging from self-interested to fair, fum strategy can serve as a means for providing guidance to employees concerning the expected behavior. More specifically, a strategy that instructs buyers to cover only the supplier’s marginal production costs (i.e., self-interested strategy) encourages buyers to behave in a self-interested manner. Conversely, a purchasing strategy that instructs buyers to cover marginal production costs and investment cost and to share the net surplus from trade17 (i.e., fair strategy) encourages fair behavior. Recall that expectations that buyers will appropriate the surplus generated by the investment drive the hold-up problem. Since the surplus from trade is fixed once the investment has been made, the offer made by the buyer is a measure of how the surplus is divided. Accordingly, buyer’s offer is the main dependent variable of interest for the trade task and Hla—d. Likelihood of investing and mean investment level are the criteria of interest for the investment task. The instructions suggest that making an idiosyncratic investment is socially optimal. Accordingly, consistent with recent experimental economics hold-up 17 Instructions suggested that buyers should share the surplus equally to remove noise. Instructing buyers to share the surplus equally was chosen to be consistent with prior research (Camerer and Loewenstein 1993). 35 papers (Ellingsen and Johannesson 2004b; 2004a; 2005; Sloof, Oosterbeek, and Sonnemans 2007), participants made a binary choice. They could choose between investing $0 and a specified. amount F which was chosen so that M + F < G. Hence, choosing to invest is efficient by construction. The likelihood of investing is the criterion for H2, and investment level is the criterion for H3 a-d. They are measured by suppliers’ investment choice and the resulting investment amount. 2.4.2 Experimental Materials and Procedure The trade and the investment tasks were run as separate experiments to facilitate backward induction. '8 The intent of this study is not to capture how well suppliers can do backward induction. Instead, first, I test the effects of the level of aggregation of the supplier-provided cost information and the buyer’s firm strategy on buyer’s offer in the trade stage. Second, given that suppliers are informed of the buyers’ level of information aggregation and firm strategy, I measure how this knowledge affects suppliers’ investment decision in the investment stage. The trade and investment tasks were conducted as separate experiments. F ifty-two subjects took part in the trade task. Fifty- four subjects participated in the investment task. Subjects were randomly assigned to conditions for both tasks. Detailed experimental materials can be found in Appendix D. 2. 4. 2.1 Trade Task Participants were assigned the role of buyer. Consistent with prior hold-up research (e.g., Gul 2001; Ellingsen and Johannesson 2004a), an ultimatum game 18 Additionally, since it is predicted that suppliers in the fine condition do not invest (thus, trade does not occur), the effect of the level of aggregation of cost information on the trade negotiation could not be assessed for all conditions if an investment decision were to be followed by a trade decision. 36 represented the trade stage. The game was not repeated in order to mitigate reputation concerns that are associated with repeated exchanges. Buyers learned they would trade with self-interested suppliers and the trade task was limited to a one-time offer. Participants made offers with the expectation that suppliers would be able to reject their offers (as in an ultimatum game) if the offers did not cover marginal production costs. Although no participant was assigned to the role of supplier in this experiment, any offer that covered marginal production costs was deemed to be accepted because self-interested suppliers would derive positive utilities fiom such offers. Each participant was presented with a scenario where s/he had the opportunity to purchase parts fi'om a supplier. The participants learned that this self-interested supplier had made an idiosyncratic investment and that they would need to make a take-it-or- leave-it offer on behalf of their firm to purchase parts from that supplier. They learned the supplier’s cost information (i.e., fine or coarse). They knew the expected revenue they could obtain from selling the product to the outside market. Their offer was to be guided by their firm strategy (i.e., self-interested or fair). Introductory materials detailed the buyer and supplier payoffs calculations from trade given that an idiosyncratic investment had been made. The calculation of incentive compensation as a percentage of the buyer’s share of the surplus from trade was also included. Following an example, pre-trade decision questions were administered to gauge the participants’ understanding of the key elements in the materials (namely, existence of an idiosyncratic investment and effect of firm strategy on payoffs as detailed in Table 2-3, Panel A). Participants were encouraged to refer back to the instructions if they were unsure of their answer. 37 Table 2-3 Test of Subjects’ Comprehension of Experimental Setting Questions were measured on a 9-point Likert scale with 1 representing strongly disagree and 9 representing strongly agree. Panel A — Trade Decision Mean Q: My firm’s purchasing strategy affects 8 12 the cash I will derive from trade. ' Q: The Seller can use the project-specific 2 17 machine to make parts for another buyer. ' Q: If the Seller rejects my offer, my 7 87 incentive compensation will be $0. ' Panel B — Investment Decision Mean Q: The Buyer’s firm purchasing strategy 8 37 affects the net cash I will derive from trade. ' Q: I can use the project-specific machine to l 83 make parts for another buyer. ' Q: If I do not purchase a project-specific 6 48 machine, my firm’s net cash is $0. ' After returning the pre-experiment questionnaire, participants made an offer decision and gave their best estimate of the lowest offer the supplier might accept. Then, participants completed a post-experiment questionnaire that included manipulation 38 checks and questions designed to help understand their decision process and help eliminate alternative explanations (see Appendix C). Finally, participants’ personal preference for fairness were elicited with an instrument created by Messick and McClintock ( 1968) and updated by Liebrand (1984). 2.4.2.2 Investment Task For the investment task, participants were assigned the role of supplier. Participants were presented with a scenario where they could purchase, on behalf of their firm, an idiosyncratic machine to manufacture parts for a specific buyer. Before making their investment decision, participants learned that, subsequent to the purchase of the machine, the buyer would make a take-it-or-leave-it offer for the parts. Participants received introductory materials detailing supplier and buyer payoff calculations should investment occur or not. Incentive compensation calculation as a percentage of the supplier’s net cash was also included. Participants knew their costs and had complete knowledge of the level of aggregation of the buyer’s information (i.e., whether buyer had fine or coarse supplier- provided cost information). The materials then further explained the potential effect of buyer’s firm purchasing strategy on trade by means of examples. Following these examples and pre-investment decision questions similar to those of the trade task (See Table 2-3, Panel B), participants were asked to make a one-time investment decision faced with a 75% chance of working with a buyer who is assigned a self-interested (fair) firm strategy and a 25% chance of working with a buyer who is assigned a fair (self- interested) firm strategy. The participants’ expectations of the buyers’ offers were also collected. As in the trade task, post-experiment questions followed (see Appendix C). 39 Finally, the buyer’s actual strategy was selected by drawing from the distribution specified in the instructions; thus, determining suppliers’ net cash and resulting incentive compensation. 2.5 RESULTS 2.5.1 Descriptive Statistics Economic theory predicts that buyers who follow a self-interested strategy will offer just enough to cover marginal production costs plus a. The offers of buyers in the self-interested/fme information condition are consistent with this prediction: Based on marginal production costs of $10,000 in this experiment, Table 24, Panel A shows that offers ranged from $10,000 to $55,000 (mean $16,857 and mode $11,000). In the self- interested/coarse information condition, offers ranged fi’om $13,500 to $90,000 (mean $38,625 and mode $30,000).19 19 Responses from one participant were omitted from the analysis. One participant in the coarse information condition used the incentive compensation formula and his estimate of the incentive pay to calculate the exact marginal production costs (in essence resulting in fine information). 40 Table 2-4 Descriptive Statistics Panel A: Mean (standard deviation) Buyer Offers Self-interested strategy Fair strategy Marginal means Fine $16,857 $107,500* $58,692 information ($15,341) ($17,650) ($48,815) n = 14 n = 12 n = 26 Coarse $38,625 $105,000 $73,140 information ($19,609) ($21,409) ($3 9,381) n = 12 n = 13 n = 25 Marginal $26,904 $106,200 $65,775 means ($20,348) ($19,326) ($44,599) n = 26 n = 25 n = 51 Panel B: Mean (standard deviation) Supplier Investment Self-interested strategy Fair strategy Marginal means Fine $5,714 $36,923 $20,741 information ($14,525) ($1 1,094) ($20,367) 2/ 14 or 14% invest 12/13 or 92% invest 14/27 or 52% invest Coarse $34,285 $36,923 $35,556 information ($14,525) ($1 1,094) ($12,810) 12/ 14 or 86% invest 12/ 13 or 92% invest 24/27 or 89% invest Marginal $18,571 $36,923 $28,148 means ($20,3 15) ($10,970) ($1 8,436) 14/28 or 50% invest 24/26 or 92% invest 38/54 or 70% invest Self-interested strategy: Buyer’s firm strategy is to cover supplier’s marginal production Fair strategy: Fine information: Coarse information: cost. Buyer’s firm strategy is to share the net surplus equally with the supplier and cover supplier’s marginal production costs and investment cost. cost. Buyer knows supplier’s marginal production costs and investment Buyer knows the sum of supplier’s marginal production costs and investment cost. * The mean offer of $ 107,500 observed in the fine information/fair strategy condition is equivalent to reimbursing marginal production and investment costs in the amount of $50,000 and sharing the net sun'plus (i.e., $170,000 less $50,000) almost equally. These results are consistent with prior literature where buyers who possess fine information and behave fairly offer to suppliers about half of the net surplus above and beyond the costs incrured by the suppliers (cf. Berg, Dickhaut, and McCabe 1995; Ellingsen and Johannesson 2004; Sloof, Oosterbeek, and Sonnemans 2007). 41 Figure 24' Distribution of Buyer Offers Level of aggregation Fine information Coarse information assessors Frequency of offers 14 ;~ 12 _ 1O ON‘GG 7 j 7 fir 1’ f '9'9‘9’PC9NNQQ9 14 --«-._-..-_.--.. 12 _ 10 ~— 8 2. a 4 i 2 ‘ J 0 HI], I], ‘ ' [L ,[L Offer amount in thousands . Variables are defined in Table 2-4. 42 Self- interested Buyer strategy Fair Additionally, economic predictions suggest that self-interested suppliers paired with buyers who are assigned a self-interested strategy will not make the socially optimal idiosyncratic investment (i.e., investment of $40,000 per experimental instructions). In the self-interested/fine information condition, two out of fourteen (or 14%) suppliers invest and mean investment amount is $5,714 as detailed in Table 2-4, Panel B. On the other hand, twelve out of fourteen (or 86%) suppliers invest in the self-interested/coarse information condition with a mean investment amount of $34,285. In essence, the observed behavior of participants in the self-interested strategy and fine information conditions is broadly consistent with economic predictions20 (i.e., offers of $10,000 and no investment). Conversely, participants in the self-interested strategy and coarse information conditions make significantly higher offers and investment inefficiency is significantly lower than predicted by economic models; thus, providing initial support for Hlb and H3b (t = 5.06, p < .001 and t = 8.83, p < .001, respectively). 2.5.2 Manipulation Checks Level of aggregation of cost information possessed by buyers was measured with post experiment questions by asking participants about their perception of the buyer knowledge of exact marginal production costs. Perceived buyer knowledge of exact marginal cost was rated significantly higher in the fine information condition than in the coarse information condition as detailed in Table 2—5, Panel A for both trade and investment tasks (t = 7.78, p < .001 and t = 4.64, p < .001, respectively). 20 Irnportantly, these results differ from prior empirical research (cf. Berg, Dickhaut, and McCabe 1995; Ellingsen and Johannesson 2004a; 2004b; Sloof, Oosterbeek, and Sonnemans 2007) because, in those experiments, buyer and supplier types were neither manipulated nor measured. Additionally, prior research did not inform participants of their counterparts’ type (i.e., self-interested or fair). Hence, in these studies, sharing of the Stu-plus was commonly observed as participants did not know the acceptance threshold of their counterparts. 43 Table 2-5 Manipulation Checks? Level of Aggregation of Information and Buyer Strategy Panel A: Level of Aggregation of Information Questions Mean Trade decision Fine Coarse Q: I knew exactly what Seller Z’s expected production costs were. Q: I knew with certainty what was the lowest offer Seller Z would 6.67 2.93 accept. Q: I knew the total of Seller Z’s expected production costs and machine costs but did not know Seller Z’s exact expected (t = 7.78, df= 48, production costs (reverse coded). P < .001) Investment decision Fine Coarse Q: I believed Buyer B knew exactly what my expected production costs were. 6 76 4 07 Q: Even if Buyer B’s firm policy was to offer just enough to ' ' cover production costs, there was a chance I might get reimbursed for the cost of the machine I purchased (reverse coded). Q: I believed Buyer B knew the total of my expected production (t = 4.64, df = 48, costs and machine costs but did not know my exact expected p < .001) production costs (reverse coded). Panel B: Buyer Strategy Questions Mean Trade decision Self- F air interested Q: My firm’s purchasing strategy is to offer sellers just enough to . 7.75 2.79 cover expected production costs. Q: My firm’s purchasing strategy is to share the net cash fi'om (t = 9.32, df = 48, trade evenly between Seller and Buyer firms (reverse coded). p < .001) Investment decision Self- Fair interested Q: Buyer B’s firm purchasing strategy is to cover expected . 6.0 4.04 production costs. Q: Buyer B’s firm purchasing strategy is to share the net cash (t = 2.74, df = 48, evenly between both firms (reverse coded). p < .01) . Questions were measured on a 9-point Likert scale with 1 representing strongly disagree and 9 representing strongly agree. Responses to the questions in Panels A and B were averaged for each decision. Variables are defined in Table 2-4. Additional questions and their responses are analde in Appendix C. 44 Additionally, participants were asked to state their understanding of the buyer strategy. As detailed in Table 2-5, Panel B, participants in the self-interested strategy condition were more likely to agree with the statement that buyers would offer just enough to cover marginal production costs than participants in the fair strategy condition for both trade and investment tasks (t = 9.32, p < .001 and t = 2.74, p < .01, respectively). Irnportantly, I propose that the supplier’s investment decision is driven by the level of aggregation of the information possessed by the buyer and the strategy assigned to the buyer, not by trust considerations. Consistent with March and Olsen (1989) and Williamson (1993) analyses of trust, I posit that the investment decision can be equated to a gamble with given probabilities of different outcomes and is therefore calculative in nature. In support for this opinion, while several participants mentioned they conducted expected utility calculations, none referred to trusting intentions (cf. Mayer, Davis, and Schoorrnan 1995). 2.5.3 Hypotheses Tests 2.5.3.1 Trade Decisions - H] (H! Hla predicts that the effect of level of aggregation of supplier-provided cost information on buyer offers depends on buyer strategy. To test H] a, the interaction between level of aggregation of cost information and firm strategy is examined. AN OVAs are run on buyers’ offers with aggregation level and buyer strategy as between- subject factors. As detailed in Table 2-6, Panel A, the interaction between aggregation level and buyer strategy is significant (F = 5.42, p = .02).21 I conducted additional tests to determine whether results were driven by the buyer strategy specified in the instructions 21 Results remain qualitatively the same when demographics variables (gender: F = 5.06 and p = .029, GPA: F = 4.86 and p = .033, nationality: F = 5.53 and p = .023) are included as a covariate. 45 or by personal preferences of the participants for fairness. Personal preferences for fairness was measured by using an instrument that uses decomposed games to measure social value orientation and classify individuals as self-interested or fair (Messick and McClintock 1968; Liebrand 1984; De Dreu and van Lange 1995). Whether these preferences matched the buyer strategy was used as a covariate in the analysis and results were substantially the same (F = 4.89, p = .032). Hence, I conclude that the effect of level of aggregation of cost information on buyer offers depends on buyer strategy. 46 Dependent variable: Buyer offer Table 2-6 Tests of Hla-d Panel A -dee Decision — ANOVA Test of Hla Sum of F- Significance Eta Source Squares df Mean Square statistic of F squared afigegauo“ 1178495181 1 1178495181 3.42 .071 .068 Buyer 78263973411 1 78263973411 226.82 .000 .828 strategy Aggregation 024 level x buyer 1 178495181 1 1 178495181 5.42 ' .103 Test of Hla strategy Explained 83240282242 3 27746760747 80.43 .000 .837 Error 16214263072 47 344984320 Total 99454545314 50 Adjusted R2 = .827 Panel B - Trade Decision — Planned Contrast Tests Hypothesis Comparison Statistic df p-value Hlb Coarse versus Fine Information (Self- .002 interested Strategy condition) t = 3.18 24 (one- tailed) ch Coarse versus Fine Information (Fair = Strategy condition) t '32 23 '75 Hld Farr versus self-mterested Strategy (Coarse t = 8.06 23 < .001 Information condrtron) Level of aggregation of information: Buyers know marginal production costs and investment cost separately (Fine Information), or know only the sum of marginal production costs and investment cost (Coarse Information). Buyer strategy: Buyer’s firm strategy is to cover only supplier’s marginal production cost (Self- interested Strategy), or Buyer’s firm strategy is to share the net surplus equally with the supplier and cover supplier’s marginal production costs and investment cost (Fair Strategy). 47 I used planned contrasts to test Hlb through Hld by examining pairwise mean comparisons. As detailed in Table 2-6, Panel B, when buyers are assigned a self- interested strategy, the mean offer in the coarse condition is statistically significantly greater than the mean offer in the fine condition (t = 3.18, p = .002 one-tailed). Thus, Hlb is supported. Additionally, the mean offer difference between fine and coarse information is not significant for buyers in the fair strategy condition (t = .32, p = .75); thus, results are consistent with ch. Finally, when buyers possess coarse cost information, the mean offer of buyers in the fair strategy condition is statistically significantly greater than the mean offer of buyers in the self-interested strategy condition (t = 8.06, p < .001). Thus, Hld is supported. In sum, providing buyers with coarse cost information increases the offers made by buyers with a self-interested strategy, but does not change the offers of buyers with a fair strategy. 2. 5.3.2 Investment Decisions — H2 and H3d-d H2 predicts a main effect of level of aggregation of cost information on suppliers’ investment decision. As predicted and detailed in Table 2-7, Panel A, results from a logistic regression provide evidence that suppliers who knew they were paired with buyers who possessed coarse cost information were more likely to invest than suppliers who knew they were paired with buyers with fine cost information (Wald statistic: 7.68, p = .006). 48 Table 2-7 Tests of H2 and H3a-d Dependent variables: Supplier investment decision and investment amount Panel A — Investment Decision. - Test of H2 Hypothesis Wald Statistic [3 Significance Main effect of aggregation level on likelihood of investing 7'68 2'0 '006 Panel B - Investment Decision. - AN OVA Test of H3a" Sum of Mean F- Significance Eta Source Squares df Square statistic of F squared gfigegamn 2751322751 1 2751322751 16.30 .000 .246 Buyer 3860968660 1 3860968660 22.87 .000 .314 strategy Aggregation level x buyer 2751322751 1 2751322751 16.30 '00" '246 Test of H3a strategy Explained 9575254375 3 3191751458 18.91 .000 .53 Error 8439560439 35 168791208 Total 18014814814 53 Adjusted R2 = .503 Panel C - Investment Decision“ — Planned Contrast Tests of H3b-d Hypothesis Comparison Statistic df p-value H3b Coarse versus Fine Information (Self- < .001 interested Strategy condition) t = 5.20 26 (one- tailed) H3c Coarse versus Fine Information (Fair t = 1 24 1 0 Strategy condition) ' H3d Farr versus Self-interested Strategy t = .53 25 .603 (Coarse Informatlon condrtlon) LVariables are defined in Table 2-6. Results from logistic regression are qualitatively equivalent (Wald statistic: 3.85 significant at .05 level). 49 H3a predicts that the effect the level of aggregation of cost information of the buyer has on the supplier idiosyncratic investment depends on buyer strategy. To examine the interaction AN OVAs are run on investment level with level of aggregation of cost information and buyer strategy as between-subj ect factors. Results provide evidence of an interaction (F = 16.30, p < .001). Thus, H3a is supported as detailed in Table 2-7, Panel B and Figure 2-5. Figure 2-5" Suppliers’ Mean Investment 40.000 - Buyer a , strategy: a I 9 —— Fair 2 ’I’ .E , ------ Self- ? interested 2 , ' 10.000 - ' l T Fine Coarse Level of Aggregation of Information . Variables are defined in Table 2-4. 50 I used planned contrasts to test H3b through H3d by examining pairwise mean comparisons. Results are presented in Table 2-7, Panel C. When suppliers are likely to be matched with buyers with a self-interested strategy, the mean investment for the coarse information condition is statistically significantly greater than the mean investment in the fine information condition (t = 5.20, p < .001). Thus, H3b is supported. Additionally, the mean investment difference between fine and coarse information is zero when buyers have a higher probability of being assigned a fair strategy and all participants invest. Thus, H3c cannot be rejected. Finally, when buyers possess coarse information, the mean investment when buyers have a higher probability of being assigned a fair strategy is not statistically significantly difl‘erent than the mean investment when buyers are more likely be assigned a self-interested strategy (t = .53, p = .60). Thus, H3d cannot be rejected. 2.5.4 Supplementary Analysis Post experiment questions were used to assess the soundness of the investment strategy of the suppliers. As detailed in Table 2-11 and 2-12 Panels B, participants appropriately took into account the potential effect of their actions on their firm net cash and their own payoffs and did not ignore the 25% chance that the buyer’s firm might have a different strategy. Trade efi‘iciency (i.e., agreement fiom trade) cannot be directly measured since no participant takes on the role of suppliers in the trade task and suppliers do not actually reject or accept offers. However, trade efficiency is of concern because it has been proposed (Gul 2001) and observed (Sloof, Oosterbeek, and Sonnemans 2007) that partial sharing of information impairs trade efficiency. In this dissertation and experimental instructions, suppliers are assumed to be self-interested. Accordingly, given that an 51 idiosyncratic investment has already been made, any offer that is greater than marginal production costs should be accepted by suppliers (i.e., efficient trade). Moreover, as detailed in the trade predictions above, it is unlikely that buyers will offer less than marginal production costs. Thus, in contrast with Gul’s predictions, I propose that trade will be efficient. The minimum offer acceptable by a self-interested supplier is $10,000, i.e., the marginal production costs, in this experiment. Therefore, any offer that is equal to or greater than $10,000 will be accepted and will, accordingly, result in efficient trade. With coarse cost information, the lowest offer observed was $13,500. In other words, all buyers offered more than suppliers’ marginal production costs suggesting that acceptance by self-interested suppliers would follow. Hence, the results of this experiment are consistent with efficient trade despite aggregation of information. Additionally, $30,000 is the modal offer observed. Interestingly, based on the figures used in this experiment, $30,000 is also the lowest offer that provides sufficient incentives for a self-interested supplier to invest when he is faced with a 75% chance of being paired with a buyer who is assigned a self-interested strategy. It was proposed earlier that, in the coarse information condition, self-interested buyers ’ offers would be equal to (F + M)c - 8 + a with 5 Z 0 and 6 significantly smaller than F. Thus, buyers’ trade offers should be significantly greater than M and close to the sum of marginal and investment costs (i.e., (F + M)c) for participants in the coarse and self-interested conditions. The mean offer of buyers assigned a self-interested strategy in the coarse information condition (i.e., $3 8,625) is not significantly different than the sum of marginal and investment costs (F + M) (i.e., $50,000) (t = 1.60, p = .14) (after eliminating one observation where the subject used his prior experience to estimate the 52 marginal production costs). Thus, results provide evidence that buyers who possess coarse cost information make offers close to the sum of marginal production costs and past investment cost, even when they are instructed to only cover marginal production costs and to maximize their firm surplus (i.e., self-interested strategy condition). Additionally, the mean investment of suppliers who had a higher probability of being paired with buyers with a self-interested strategy and coarse cost information ($34,286) is not significantly lower than socially optimal investment of $40,000 (t = 1.47, one-tailed p = .083). Hypotheses 2 and 3a-d above rely on the prediction that suppliers would anticipate receiving higher ofi'ers, from buyers assigned a self-interested strategy, in the coarse information condition than in the fine information condition. Consistent with this prediction, suppliers matched with buyers who are likely to have a self-interested strategy and who possess coarse cost information expect, on average, offers of $47,278. On the other hand, suppliers matched with probably self-interested buyers who possess fine cost information expect, on average, offers of $26,946. The mean difference is significant (t = 1.93, one-tailed p = .033). Additionally, the expectations of suppliers who had a higher probability of being matched with buyers with a fair strategy do not change with the level of aggregation of information (t = .22, p = .827). In sum, the prediction that the level of aggregation of the information possessed by buyers with a self-interested strategy affects the supplier’s trade expectations is supported. Post-experiment questions help eliminate the sunk cost bias as a potential explanation for the buyers’ offers being higher than predicted by economic theory. Participants were asked to indicate (on a scale of 1 to 9) their agreement with statements 53 that suppliers would reject offers that did not cover their expected production costs and machine cost, and that, once a seller has purchased a project-specific machine, this purchase should be reimbursed by the buyer no matter what the buyer’s firm purchasing strategy is (See Table 2-12). Participants selected their answer from a 1 to 9 Likert scale. Participants’ mean answers are not significantly greater than 5 in all conditions (mean = 4.66). This finding provides evidence that offers that are greater than marginal costs are not the result of the sunk cost bias of participants. 2.6 DISCUSSION AND CONCLUSIONS Successful alliances, networks and other types of close buyer-supplier relationships are often characterized by extensive sharing of information (Baiman and Rajan 2002a; Kulmala 2002). Management accounting information plays an important role in this information sharing as exchange of cost information has helped firms diagnose problems, reduce costs and improve performance (Carr and Ng 1995; Kumar 1996; Seal et al. 1999; Dekker 2003). Yet, contracts are incomplete and suppliers who disclose their cost information to more powerful buyers cannot use contract terms to protect themselves firlly against the potential self-interest of their counterpart. Anticipating ex post self-interested behavior fiom their counterpart, firms have, ex ante, avoided to make idiosyncratic investments that would be socially optimal (i.e., hold-up problem) (Williamson 1985). This research investigates the effects of the level of aggregation of supplier- provided cost information (i.e., fine or coarse) and of trading with buyers with certain firm strategies (i.e., self-interested or fair) on the hold-up problem. Results provide 54 empirical evidence that aggregating idiosyncratic investment and marginal production costs (i.e., coarse information) instead of disaggregating costs (i.e., fine information) helps suppliers overcome their reluctance to make idiosyncratic investments. First, because of information loss associated with aggregation, buyers with a self-interested strategy who possess coarse cost information are unable to determine the sunk past investment cost when making a trade offer. Consequently, their offers are significantly greater than marginal production costs and greater than the offers of buyers with fine cost information. Second, anticipating that buyers with coarse cost information will not be able to behave opportunistically ex post, the suppliers matched with such buyers are more likely ex ante to invest than suppliers matched with buyers who possess fine cost information. In essence, buyers’ information loss associated with coarse cost information helps curtail buyers’ self-interested behavior during trade and, consequently, provides investment incentives for suppliers; thus, mitigating the hold-up problem and improving overall efficiency of the transaction. Irnportantly, increasing the level of aggregation of cost information does not change the trade offers of buyers in firms with a fair strategy or the investment decision of suppliers who had a higher probability of being matched with them. Laboratory testing allows simplification of the context, identification of optimal behavior and manipulation and controlling of the buyer strategy (i.e., self-interested or fair), but it also suffers from the limitations associated with any experimental work. Additionally, coarsening cost information might reduce the potential benefits of providing cost information. Thus, it is important to examine the cost-benefit tradeoffs of coarsening cost information prior to implementing this mitigating mechanism. 55 This research contributes to the literature on hold-ups by intentionally abstracting from communication or any reputation concerns associated with close buyer-supplier relationships. As a result, this dissertation contributes to the literature on incomplete contracting and opportunism in buyer-supplier relationships by providing evidence that aggregated supplier-provided cost information provides suppliers with incentives to make a socially optimal idiosyncratic investment in the context of small nrnnbers bargaining. Finally, my research contributes by isolating and manipulating buyer strategy. By assigning buyers a firm strategy to guide their actions, this dissertation successfully induces trade offers consistent with economic predictions when there is no private information and buyers are assigned a self-interested strategy. Additionally I document an interaction between level of aggregation of supplier-provided cost information and buyer strategy. I assume that the level of aggregation of supplier-provided cost information is exogenous. Letting suppliers decide whether they would like to disclose fine or coarse information, and the effect of their decision on the relational contracting aspect of their relationship with buyers warrants further investigation. Additionally, to complement this research, the effect of other types of partial disclosure of information (e.g., through mechanisms such as reduced precision and reduced timeliness) on inter-firm relationships could be investigated. 56 APPENDIX A Detailed Calculations of Trade and Investment Predictions In this study, following a fair strategy is defined as reimbursing marginal production costs and investment costs and sharing the net surplus. Although Fehr and Schmidt’s (1999) theory of inequity aversion examines fairness at the individual level instead of the firm level, their results can be adapted to analyze expected individual outcomes of employees that follow a fair firm strategy. Fehr and Schmidt (1999) suggest that, for some individuals, their utility function includes disutility when his own payoff is not equal to their counterpart’s. Irnportantly, the Fehr and Schmidt’s model of inequity aversion proposes that disutility is greater when the payoff difference is disadvantageous than when it is advantageous to oneself. Hence, Fehr and Schmidt suggest that the utility function of player i is of the following form: Ui = Xi - (1i mflx(xj - Xiio) - 131 maX(Xi - X130) (1) i 9E j and j is the other player. 0., 2 [3i and 1 > Bi 2 0. xi is the net monetary payoff of player i. 0.; 2 Bi suggests that the buyer’s trade offer will not give the supplier an amount greater than half the net joint surplus. Accordingly, I expect the buyer’s payoffs to be larger than (or equal to) the supplier’s payoffs. Consequently, xs _<_ X}, and ab max(xS - xb,0) = 0. Thus, from equation (1), it follows that the utility function of buyer b, Ub, is of the following form: Ub = Xb - Bb max(xb - X530) (2) Let G represent the gross surplus generated by the supplier’s idiosyncratic 57 investment (i.e., the surplus fiom trade), M the predicted marginal production costs, and F the supplier’s investment cost. G > M + F. In other words, investment is socially efficient. Let I represent the supplier’s investment decision: I = 1 if investment takes place, I = 0 otherwise. X; represents the net monetary payoffs of player i. The surplus from trade is as follows: G if I = 1 0 if I = 0 Let T equal the trade offer of the buyer to the supplier. I define a self-interested buyer as a buyer who intends to make an offer that just covers production costs M plus 8.8 For a self-interested buyer, equation (2) applies with [31, = 0.b Consequently, the utility function of the buyer who is self-interested is of the following form: Ub = xb Ub = xb = G - T if trade takes place. Alternatively, Ub = xb = 0 if trade breaks down. Trade Predictions First, I examine the effect of the level of aggregation of supplier-provided cost information. Williamson’s (1985) predictions suggest that the self-interested buyer who knows the supplier’s marginal production costs, M, and past idiosyncratic investment a a is an immaterial amount offered so that the supplier is not indifferent between accepting and rejecting the offer. b Alternatively, if the utility function of a buyer takes the form of U}, = xb - Bl, max(xb - xs,0) with [3b <.5, the buyer will maximize his utility by making offers of M + 8. Accordingly the buyer will behave in a manner similar to buyers whose utility frmction includes 13b: 0. Thus, buyers with 3b < .5 are included in the above definition of self-interested buyers. 58 cost, F, (i.e., fine information) will offer just enough to cover marginal production costs plus a (or M + e). Anticipating this, the supplier will not make the socially optimal idiosyncratic investment. I propose that a self-interested buyer who possesses coarse cost information c in the form of the sum of the supplier’s marginal production costs and past idiosyncratic investment cost (i.e., (F + M)c) will offer more than a self-interested buyer who possesses fine cost information. Knowing that a self-interested supplier has an acceptance threshold of M, a self-interested buyer would like to offer M + e (where a is immaterial), but he has no prior information about the supplier’s true marginal production costs, M, or the investment cost, F. The buyer wants to make a low offer so as to maximize his payoffs, but, being risk averse, does not want to risk losing his share of the surplus from trade (i.e., G -T) by making an offer that does not cover marginal production costs. Hence, his tradeoffer, T, isasfollows: T=(F+M)c-8+ewithS=E(F)-0.6is suchthatfiZO. 6 represents the buyer’s expected value of F (E(F)) reduced by the adjustment the buyer makes for his uncertainty, 0. The higher the buyer’s uncertainty about M and F’s true values, and the higher the surplus fi'om trade, the higher 0, and the lower 6. Additionally (F + M)c - 8> M for values of F sufficiently large.0 In essence, uncertainty about the true marginal production costs and fear of losing the surplus fi'om trade will cause a self- interested buyer to include most of the cost of the supplier’s past investment in his offer and offer close to (F + M)c. In sum, the information loss associated with coarse information limits the buyer’s potential self-interested behavior; thus, resulting in higher c The hold-up problem would be a non issue if idiosyncratic investments were small. 59 offers than fine information. Second, I examine the effect of buyer strategy. Extant research (Carmichael and MacLeod 2003; Ellingsen and Johannesson 2004; Ewerhart 2006) suggests that fair buyers will reimburse past investment costs.d Consequently, I define a fair buyer as a buyer who intends to share the net joint surplus (i.e., potential surplus fiom trade less marginal production costs and past investment cost) and to reimburse the supplier for marginal production costs and investment cost. Thus, for a fair buyer, equation (2) applies with Bi > 5” Hence, the utility frmction of a fair buyer is of the form: Ub=xb-Bbmax(xb-xs,0)=G-T-Bb(G-T-(T-(F+M)),witthZ.5 When the buyer shares the surplus equally (i.e., offers G/2 + F0 + M/2), G - T - (T - (F + M)) = 0. As a result, the buyer’s utility is maximized and Ub = G - T when the buyer offers a price that covers up to half of the net surplus and reimburses the supplier for his marginal production costs and investment cost. Thus, the buyer does not need to identify F and M separately in order to make his offer. Consequently, the utility of a fair buyer is maximized when he shares the net surplus equally whether he possesses fine (i.e., F and M) or coarse (i.e., the sum (F + M)c) cost information. Thus, offers of fair buyers should not be affected by whether investment and marginal costs are disclosed disaggregated (in the form of F and M) or d Alternatively, Troeger (2002) and Ellingsen and Robles (2002) have presented evolutionary models where, as a result of repeated interactions, sunk costs are included in the buyers’ offers and investment is efiicient. Still, not only do those papers assume much repetition of transactions whereas the hold-up problem is typically defined as a one-shot problem, but these papers also fail to explain why the srn'plus would be shared between buyer and supplier (Ellingsen and Johannesson 2004). c The case of 5b =.5 leads buyers to be indifferent between any value of T provided that XS S xb 60 aggregated (in the form (F + M)c). As a result, the offer made by fair buyers who have coarse cost information is not expected to be significantly different from the offer made by fair buyers who possess fine cost information. Investment predictions Suppliers will not know buyers’ type (i.e., self-interested or fair) with certainty. Recall that, consistent with econorrrics models, suppliers are assumed to be self- interested. Accordingly, equation (1) applies with as = BS = 0. Consequently, suppliers’ utility fimction is of the form: Us = xS = T- F - M Let q be the probability that the buyer is self-interested. (1 - q) is the probability that the buyer is fair. (i) When the buyer possesses fine cost information, f, the supplier, s, has q probability of receiving T = M + a and (l-q) probability of receiving T = [(G- F - M)/2 + F + M] during trade. Thus, the supplier’s expected utility from trade (E(Usf» is the following: E(Usf) = q(M + e)+(1- q)[(G - F - M)/2 + F + M] - F - M. The supplier will invest if his expected utility is positive. E(U3f) 2 0 is equivalent to: (G-F-M)/2FZq/(1-q) (3) Hence, the higher the probability of the buyer being self-interested (i.e., q) or the higher the investment cost (i.e., F), the less likely the supplier will invest. (ii) When the buyer possesses coarse cost information, c, in the form (F + M) c, the supplier, s, has q probability of receiving T = (F + M)c - 8 + 8 and (1 - q) 61 probability of receiving T = (G - (F + M) c)/2 + (F + M)c during trade. Thus, the supplier’s expected utility from trade (E(Usc» is greater than or equal to q((F+M)c—8+a)+(l -q) [(G- (F+M)c)/2+(F+M)c] -F-M. Hewill invest ifhis expected utility is positive. E(Usc) 2 0 is equivalent to: (G - F - M)/28 Z q/(l - q) (4) The value of 8 is an empirical question. It remains however that 8 < F as explained above. Consequently, E(Usc) > ECUsf) and the supplier is more likely to invest in the coarse than in the fine condition. 62 APPENDIX B Documentation of the Process Followed The final experiment was conducted subsequent to several tests that resulted in changes to the form of the game, the instructions and the way the experiment was administered. This evolution is detailed below in Figure 2-6. Figure 2-6 Evolution of Testing Phase 1: pre- Phase 2: pilot Phase 3: final pilot test test collection Trade task: Trade task: Trade task: One-shot One-shot ultimatum Game form not changed. ultimatum game game (adapted) Explanation of incentive Game repeated Game not repeated compensation changed. Instructions Measure of social modified motives added. Investment task; Investment task; One-shot game Game form and Game not repeated instructions not changed. Trade offers from pilot test used to inform suppliers of possible outcomes. The effects of the level of aggregation of supplier-provided cost information and buyer strategy on the hold-up problem were tested by conducting a trade experiment and an investment experiment separately. While this departs from prior papers who combined both tasks in one experiment (e.g., Berg, Dickhaut, and McCabe 1995; Ellingsen and Johannesson 2004; Sloof, Oosterbeek, and Sonnemans 2007), this makes for a more 63 efficient use of subjects as prior experiments were unable to make use of one set of participants (namely, buyers) when suppliers had not invested. The evolution of Phases 1, 2 and 3 presented in Figure 2-6 is detailed below. Phase 1: Pre-pilot Tests I chose to examine trade first because suppliers base their investment decision on expectations from trade and I needed to be able to give suppliers information about the outcome of the trade negotiations. Once the Human Subject Review Board approval was obtained and protocol studies conducted with PhD and MBA students, undergraduate students enrolled in a business class and MBAs participated in the pre-pilot test conducted during summer of 2006. The Game Participants preferences for fairness were elicited by using a an instrument from social psychology developed by Messick and McClintock (1968) and adapted by Liebrand (1984). Trade was examined at a later date by means of an ultimatum game. Participants were randomly assigned to the role of buyer or seller. A buyer paired with a supplier made a one-time offer which could be rejected by the supplier. Subjects were re- paired and buyers negotiated with another supplier. The game was repeated nine times. Administration of the Experiment A computerized web-based experiment was conducted with several groups. This ensured portability of the experiment which could be run fi'om students’ individual laptops or any computer lab. Programming was written by a computer science student using Flash. Frequent testing by the administrator was performed. Computerization of the experiment enabled the researcher to record all actions taken by the participants, to test their understanding of the instructions and not allow them to proceed until they were able to answer questions correctly, and to assess whether participants cared more about their outcome than the their counterpart’s (as they had to click on cells to reveal payoffs of each party). Results, Problems Encountered and Remedies Although results were generally consistent with predictions that offers of participants with coarse cost information were higher than those of buyers with fine cost information. Several significant problems were encountered: - A high rate of no—shows and late arrivals made pairing of subjects difficult each time the experiment was run. - Participants had difficulty answering certain pre-experiment questions as shown by the frequency of wrong answers. Instructions were clarified and/or questions rephrased. - Participation of foreign students who took longer to comprehend instruction delayed the repairing and not all sessions could be run (in one case delay was such that only one session was run). - Although the sessions were supposed to be independent as subjects were re-paired, buyers seemed to make higher offers in response to prior rejections; thus, data could only be used from the first session. - Consistent with prior research (e.g., Berg, Dickhaut, and McCabe 1995), buyers who did not know the type of the supplier they were trading with (i.e., self-interested or fair) made offers higher than predicted by economic models even in the complete information condition. 65 - Participants seemed to focus on net payoffs and ignored the breakdown of marginal and sunk costs. - Cells were not balanced as fewer than 30% of the participants could be classified as fair. The problems encountered were addressed with Phase 2. Phase 2: Pilot Tests The Human Subject Review Board approval for the trade and investment task was received as a new project. Undergraduate students from managerial accounting classes, MBAs and PhD students participated in the pilot test conducted during fall of 2006. Changes Implemented Instructions for the trade task were revised and payoff tables removed to avoid participants focusing on payoff tables and to ensure that the aggregation manipulation was effective. Participants’ preferences for fairness were manipulated through firm strategy instead of i being measured. The Game The format of the ultimatum game related to the trade task was altered so that, consistent with economic models, buyers knew with certainty that suppliers would be self-interested. The administrator took the role of the supplier and any offer greater than marginal production cost was accepted. Since results from experiments were several sessions were conducted had not previously been independent, only one session was conducted 66 Investment decision was also a one-period decision. Backward induction was facilitated by giving participants example of possible trade outcomes. Administration of the Experiments Experiments were administered with paper and pencil so as to avoid delays due to programming and need for computer laboratory space to conduct experiments. Results and Problems Encountered Manipulations of aggregation of cost information and firm strategy were effective as documented with results fiom manipulation checks questions. Results were consistent with predictions. However, two observations could not be used because one subject in the coarse information condition used the incentive compensation formula to calculate exact marginal production costs and another indicated he refused to follow the firm strategy because of strongly held preferences. The problems encountered were remedied when Phase 3 was administered. Phase 3: Final Data Collection MBAs fi'om a managerial accounting class participated in the final data collection conducted during spring of 2007. Few changes to the instructions were needed and no additional Human Subject Review Board approval was necessary. Changes Implemented First, participants preferences for fairness were elicited subsequent to the experiment by using an instrument fi'orn social psychology developed by Messick and McClintock (1968) and adapted by Liebrand (1984). This measure was to be used as a covariate in the analyses. Second, the incentive compensation formula was altered so that 67 participants in the coarse condition could no longer infer the exact marginal production cost during the trade task. Third, actual trade offers obtained during the pilot test were used to give examples of possible outcomes to suppliers who made an idiosyncratic investment decision. The Game and Administration of the Experimem The format of the games and the way the experiments were administered were the same as for the pilot test. Results and Problems Encountered Manipulations of aggregation of cost information and firm strategy were effective as documented with results from manipulation checks questions. Results were consistent with predictions. 68 APPENDIX C Additional Data Collection and Analysis Some data were collected to be used as covariates in the analyses, to assess participants’ understanding of the instructions and to provide insight about the participants’ decision-making process should results not conform to expectations. The analyses are detailed below in tables 2-6 to 2-9. Table 2-8 Sample Demographics Panel A: Frequency Frequency - Trade T ask“ Frequency — Investment Task Male 38 39 Female 1 3 1 5 Undergraduate 16 0 Graduate 35 54 North America 35 34 Europe 1 2 Asia 14 12 Other 1 6 *One observation missing 69 no. we. 8:20.me agomuamm 02 ”we... m: a. me. ~ me. ~ :38 momSoo meuesooom Hummus-WE mo 69:: Z 2 m2 9% mom e3 e2 unease used 255 95 mo. 3 2.3 ween.“ 8:385“. 2:35.35 oz Ecuauuuaxm gush «5.582: mm mm; mm. 5 ~ A m: no; 5&8 womczoo weunzoooa Rummage mo BDEDZ *2 a? med mom 3m Nam swans “see 685 #5 me. as 2.8 ween—a 3:05am“. eaaocmewmm e Z Eeflfiuoaxm ink 25:. 3..an 2F.— oEan 0:5 :32 Nwouahm 3.6.3-? =aao>O 38.25 ham 33285.2; 325.50 59.52. womanaamefion he anemia—“EU :32 "m 3:5— 83:80 u.” uses 70 925-88 Hco. v AH . . . . . 80.50 E 88808.8 mvo. n a KH 0. .mod n. Om N. mo 0 we w nm N. .2. h be .3 eoHHSano .3. .05 0:09.28 8 H 8:45-88 H8. H8. v a . . . 80:8 3 :80 .0: FEE be Ho 8:03:88 0 a .NoH u. .wdH 0. wk. H. em 0 em w w w the. 65 “898.8 8 H .0188 N e0b.mH 05 5 So. So. v a . . . . . 82.8 e :80 .0: PEG be we :oHefiquoo v a dad n. .36 00 am N. mm m mm w on \1 no m be .2: 809.28 8 H .0385 X .0b.mH 05 5 05h v 0.....an m A :82 ":035035— ”:0H0300mnm 0.0.... 08:80.»:— So. v Q . . . . . 80.80 0H 88808.8 0: 61w u 0 mm 5 mm o 5H 5 on n no m be me 8:85:88 me. .05 E0388 8 H mac. n a .N n c So. v m . . . . 80.80 e :80 .0: PEG 0eHnH v 0880 .2; H. mm 5 mm w w mm c we \1 the. .8 88.388 be .05 0:02.38 8 H Hoo. v .H . . . . . .ow 0n :3. 8880988 m: .320 .1. e we n 2 H. Ne w mm m me \1 030805 be ..0.H.Ho be $03.0. .0HH0m 05 HH Amev 00:08.5... 02800:: 235%... 2 m A :82 0.2; 2.29 ":0350093— ”:0H0500nem 09.80 05h 0.....an 0::— :8—>H .flwueeum 07.00;. =20>O 308.5 ..Ham e00u0e0ueH1=0w 28m ..BHHH 9H 8:23.050 8330—70...— .A00ewm bweoem o H08. 088mm. bmeoem m50e0m080e H 53¢ 20% :33 Erie :o 8.800.: 0.03 2580.5 :onHo0H00enH .805 o. 820:: :03 0.6: 538 .05 £88..me 0:. .630. 8 80:. 08.588 8 H08 82.05me 89820:: $830308 8.885 8008 8 8a.. 0.03 8080.8 :onH00e-0..nH 0:05—0:00 80.50.. 8083500 ...... 0:32.25 8500:.er he uHabae< eta 033—. 71 82:80 mm 000888.300 :H $5.0m 08.3 omm.mw 3N5: 93.8 08m 95. v a .34 n0 8&0 :8 m 88m 08:88.0 8% 00 :02: 30: «8&0 . . . . . . , 0. 8b: 30000 8% H 000 8: 0.8.: 80% 8 :5 v a M: w n: N; e- 8 a: mm- :m m E a- my» 0%; .38 nouwsww”: 0288 Hows 3 88:0 003. 8&0 9 0H 30080 8m 9m 88m .HH 08h 8&0 :8 m 88m 3:2: 8 :5. v a .8.” u: 8.3 «8.0. :8: 33m 80.: so: 8 E? .080 80262: 0288 02,8 9 88:0 8% 8&0 0: 0H $00.80 8m Mm 88m .3 0::— A 00800 =0 "8:300anm 0.03 08:808— 80H8-08H . a . . . 030000 03 3 mg. n : .2: u 0 30 am VmH mo com ow 30 cm mom 0H N 8HH0m 88. 8 a 8&0 8032 0:0 0%th 05.: A 00.800 :0 "8:800anm 300,—. 08:. 00800 05..— 0280 05m 802 H0085 $000.9 H885 30.005 :80— 880805.28 08:00:: @008 8:6 c—é 03:8 3:08.800H 8050: 000885: 8.. 80:8:0 88—0....8883— .8 0.3223. 72 680on up 8: 28% 530% 80.582: 2: inch .338.“ Pa «33% =<... .. .voaoota 3 8: ufionm 530% Segmgfi 05 £ng .ooo.om 55 .632» 2a $235 =< ... a 3.3. 23m 83: 25.3 3.6m $38“ 233 soiououoag as a :23 macho Mm 595 E38 z. s." a 52 no . . . . . a; a B as “8 Mg :5» as :33 $226 838 A can So a a2 Q So 8 5 Q So 8 sum 3% 2a 33m 8053 as 3: :56 05 25% 8 mm 388% SE Pm ism ..— use» .. :5. v a .2: no . . . . . Home :5 m 25 Ea 8» 8 3:3 .526 3.800 A 8E a: So ooo No ooo o: ooo mo ooo o: 8.6 “Boom 93 Exam 5233 :30 «on =Eo>o 05 08% 9 mm $08.90. SE Pm Exam u— ?5 no. as 2.3 98:3 magnum? 2.35:»: oz ":ouauooonm :23 2.2582: 99.25 2F.— oanU «Eh :32 38.25 memory .1295 .3835 bur,— couauhfififiom 2.33.25 63:3 non—O 62.5230 oTN 035—. 73 o 28 80880988 03838 3 Emma. . on 5: m3. ta 8.2 3.2 5 2a :98 Mae 3 3858 2 me 8323 no I on _ I“ 85:08 98985 Bow 8 mo Hobo as $882 . . . . . . . . 5885888 ES ow _o vnH R mu“ ow mm $5 82 mm mm on ov >8 88 58 $8.6 >8 088838 8 o3§3~ do on @395 83980883 03838 on a. v a .25 u. 8.8 8.2 3.2 2.8 3.8 E Ea 5.8 Mann 8. Ba 38.8 B was» a 3:83 Home 5 32 08 8:8 8 883 8: Bo H bah A 339.88 .... 5% 28333an 923.083 . . . . . .oommaousqonmbn 0:388 ocmoomfoo FE 9. So. u a :Nd no 3 2 R 2 me em av 3 on o_ 05.8 “moo 05. 8o N B=om 88582 8 3853“ ._38m on 250% 8.6 MN 8 :5. v a 63 u. 3.2 8.2 8.8 8.: 9mm gum 3 BE“... 88 as as. SE S. E 8&8 :30 05 8853 882mm? 05 35:3 _ 3.8.5855 .... A ham Eoflfioonnm 3580 08..— oflaoo 98..— .:32 382% ”3588:: as; £55 3325 in 38.35.28 ”55:... 2: “.33.. 33...... 352. 2: deuce?“ .35: .805 8 $828.58 88 8083 83826 @838 mo oo§a8m 05 $88 Bo: 9 82¢ 8 3808083 838 88253 $88 oo_ 880:8 8 3me 0.83 358083 m TN 035—. gawk. 012,—. 3. .08:— 859985 :38 825mg:— uua 38...“. no 838323.»:— 8:53—38.— 74 m: a; mm; mud $4 $2 .850 .m.N Ho=om :85 8032 on m: 33H 36 amen mad Wm 250% £35 55 805 :30 3.8.5 >8 .532 m m< .8: 55 H 53 08:38 n 335989 N Bzom A35 3. «a anew—38¢ 8:35.55 «:35qu a Z Ecfifiooanm 3.890 2.3 3.890 05...— :32 38.25 "8588.3 338-9 .835 38.25 ha..— vouaohoufluzum $55.55 05 9888 53.83:. 5:39 2: .5235 < .25— .:-~ 2%... 75 8.5 Pm m: 003 Hm.mH ~56 H 3.2 $0.2 00>:m 5H3 0850 8 5000 :0: HH000>0 032 0 0800:0w 3 b88888 05 500053800 H . . . . . .va 805 00808 8050808800 038008 m: 3 NH mo 3 mm HH K H: as w :0 :80 3 5585250 >8 500053800 H 85 ma. 8 0500 M880 00:08.55 2805:»? :2 ”85500.80 5015-083 300 082008 0... . I 0. . I 0.2 8.2 00.3 8.2 00.8 005828.300 20 80 20 805002 No I 3 m I 0 00: 5303 m 00>:mH :05 50800800 003 H 3.800 a 5 A 05..— 805000085— 50300008 - H3 5:0 0H5 . . . . . H080 0n 58050 85 Mm 00>:mH >5 50:80 mm 800%“05 008000.55 3 H: mm H: mm H: mm NH 2 H 5000 :0: 05 580 85 >8 >5 508.80 5000 8058me 0 Z :0: 05 8002505 008000.55 05 50553 H 5039.85.25 u 5 A .55 8055028....— 3800 05."— 3.800 05% .802 @805 3808083 3307—. H8005 R0805 .55 5030000555 ”530:5 05 80:8 50802—0 350m 2: .8505 >55: .505 8 300085.33 5:0 308.55 803008 55008 .50 00805038 05 3030 5H0: 8 805.5 8 0808083 380 8002505 $80: 2: 088:0 5 5007.0 0003 08080505 300,—. 808302.— 5 .08.— 5055280 =uN 030B 76 m8 HH.H 0 0m; EN 2.. .8050 . I 8 . . I .m.m 80>8m 8054m$0H mmo I N m I 5 506 oo.m 09H HNH. 8.5 05 5805 £000 «08 0.8.5 >8 48008 0 m< 50500805855 A 8H0...H 808 55 m 80>8m 580 8085002: 80 0508 H . 8030 80800808800 058008 . . . . . .8000 508 05808 0 8800 0.8 on mm mo mm 2. mm 80 mm Hm Hm 0. 85 >8 80.5 b88830 05 508056800 H 95 ma. 8 0:00 N8080 008080.55 80058me 02 ”80530093— 08000 085 08000 08E 8002 E8085 ”38088000 300.? H8805 30.085 .55 50.00.838.25 ”5.30:5 05 w8080 50.0050 350: Hz: 508.5800 m H085 .HHIN 030,—. 77 .05000 80508508 on 50. v 8 .Hv.m "5 05.0 wmd $0 mad 3.0 50500880 0.N 80HH0m .8260 508 55 8.5 N 80HH0m >nH 5050008 808 80.50 >8 .50 x05 05 508020800 H . . . . . . . . 800 080 0 880800 000 508 00 8. 80.18 2 01 00... 000 00 m 000 8. 0 800% 8Fww§w0wo ofioafiewwgwmofi 50.00.658.55 v u 5 .508 "8050000880 3. 00080.88 0.505 a 0:8 0080 880800 00.0 03 02 N: a: 80 088082 B 80 808 2300 8:83 2888 508 0H 008080.55 055008080088 0 80050888 8 x05 0 8005 N 80:0m .05000 80508508 50500900 50.8030 00> 00.8 cod mmd N00 ems 508.305 80.50 80 N 80:0m 0x08 05 08000 80508508 50500800 0.N 80HH0m .50 0m50H305H >8 5008 H 08E 8 500008008555 5 502—le “8050500880 00.800 08:— 00.800 08% 8002 30.085 80.55 380.80 .80...— 50.00..38_.800 2000 8888 0; ”08050080 808008-308 0.00.! 050.5. "< .080.— .0080 >Hw8050 m 580 008005 >Hw8050 858000800 H 555 2000 5005.5 5808-0 80 50.80008 0803 08050088 8206058008 08055800 800.500 080580080.— w85H0880H0H00: 50 805080008— NHIN 030,—. 78 .5080>00 08 28080 A0000 80508508 50500800 580 8000 08.8008 003 00000 Hm 3088-088 8. v 8 .mmN "8 05.5 00m mad 50:0 2; 505808 =0 0080008 08808000088 08 5303 5000 088008 008088 580 00000 80508508 505008000 80>00 508.. m 80>8mH 80.5 80.50 >80 50>0508 H 80.50 00 838-003 :0. ... 0 .008 .1; 00.0 00.0 00.0 :0 00.0 32 0 ea 208 2 008 0008080 028000 08 .8 85055088 088 008 5803 m 80>8m 50>0HH08 H 5000080088 .20m v .800— "805000083— 000. n 0 .0 _ .0 l . . . . . 002000 8802200 820.88 0 028 0.88 mm 00.8000 v 088 .m v 8002 05 m 50 8 N0 N m0 8 3 N 885 0.mH 80>8mH 5085 008080 00mm 05 508088 H ~05 me. 00 30088080 0080.00.55 08005880 08 .m v 8002 ”8050000888 08 80.50 088 50808.08 H .8 80.5 80.50 088 0 A .002 00.0 08.0 00.0 3.0 00.0 09088 20 08808 20 802030 8.8 82808 8088 08 5803 8000 008 0.mH 80>8mH 50>0508 H no. u 8 88% H5 . . . . . .>H80>0 8000 508 =080>0 05 08080 5803 m 59m 00 200 v 3.000 .0 A :82 00 0 00 0 N0 0 00 0 00 0 as 000:8 as, 305. 8 0.88 283 A05 3. 00 0:00a080 .H5 008080.55 08005890 08 .m A 8002 "80500008088 008000 08:— 008000 083 8002 >w000hm H8005 300085 .808 50000802885 20% 80088 0-H ”08050080 80580050008 800E 0808.00.05 “m .0808 50885800 NnuN 030,—. 79 APPENDIX D Experimental Materials Part 1. Trade instructions (Fine Information/Self-intemted Strategy and Fair Strategy) You are assigned the role of Buyer. The Buyer buys parts fi'om sellers and assembles them into a final product sold to the outside market. You have an opportunity to undertake a project with a Seller. This project required the Seller to purchase a machine to improve the quality of the parts the Seller manufactures. The machine the Seller purchased is project-specific and can only be used to make the parts you need. It has no value outside this project with you. You cannot purchase these parts from another seller and the Seller cannot sell these parts to another buyer. Your task is to decide how much to ofl‘er the Seller for the parts. Your ofi'er should be guided by your firm’s purchasing strategy. Your firm’s cash is calculated by deducting the amount you offer the Seller from the revenue you can earn for selling the final product. You will work through an example before deciding how much to offer the Seller. 80 .05000 80508508 50500880 805088 80300 008 0005 5080 8050 80 80000 05 >H08HH88 08 80=0m 08 .H. 808508058800 0 0808 508800 88 80.50 05 800.008 80 80000 :3 80=0m 08.H. .>w08050 88080.88 0.8.85 80> >8 5058w 08 58080 80.50 80> .08. H m .80 08080808 8 0808 088 80.8 80.50 80 80=0m 080 0808 853 80 > 8000 088008 580 3000 80508508 50000880 0.80:0m 05 3088 80 > 808508 585 25 88:00 805 080.008 50500880 80> 3088 80 > .885 08 805 8000 0>50w08 8 88008 5803 580 00000 80508508 50000880 8260 508 0005 5080 80.50 80 80000 05 808588 08 80=0m 05 .0080H.H .0508 805 8000 0.885 80808 00585008 05 08 H00w 0.80HH0m 08H 08080808 05 50500880 0H 508508 H085 80> 08800308 05 580 5000 088008 80888 .08000 80508508 50500880 80838 03088 80:0m 08.H. 088008 058008050088 08.5 50008088 >50080 008 80:0m 08 .H. >2 >3 >9 8N0 A: om “050.8 055 80.5 8000 0.885 0.80:0w 80.50 050000.. cm "050.: 055 80.: 00=0m 8000 0.885 .80 > 008 8088.0 000 80 5 A S 8 50500800 I 80.50 3000.80 8000 “050.8 0880 80.5 880080880H 0.885 0 008 80 00 088038 088008 880.080 0.8.80 0.00 800 00 88.0 02%020WQ0M0000 .8020 028.80 80:83:00 080000 80:0m 80.50 0808 80> 0.80:0m 3088 80> 3088 80> 00.50 - 0880>08 . 50020 02808 0050..» 055 80.... 8000 0.885 80> .3208 5050005 080 800.008 08 .50 0080080 08880008 580 085085 08.H. "108.40% 81 Your compensation is composed of two parts. 1) A fee of $5 that you collect for participating in this research project. 2) In addition, you can earn some incentive compensation. Your incentive compensation is calculated as follows: your firm’s cash fiom this trade/30,000. 82 INSTRUCTION SHEET Practice example: Your offer should be guided by the purchasing strategy of your firm. Self-interested Strategy condition Your firm’s purchasing strategy is to offer sellers just enough to cover expected production costs. Your firm usually ggg' ores past machine cost. Your goal is to maximize your firm’s cash. Fair Strategy condition Your firm’s purchasing strategy is to share the net cash evenly between seller and buyer firm. Your firm usually reimburses the Seller for past machine cost and expected production costs. Your firm calculates overall net cash as follows: Expected product revenue — expected production costs —— machine cost. Below is a practice example that highlights the decision you will be making and its possible outcomes. Please propose an offer consistent with your firm strategy (4) and calculate the outcome of the trade (5) in the timeline below. You will then look at the effect of this offer on your incentive compensation. Seller X has already purchased a project-specific machine to manufacture parts for you. Seller’s goal is to maximize his firm’s cash from this trade. Hence, Seller X is unlikely to accept an offer that does not cover expected production costs and would result in negative cash for his firm. Your expected revenue fi'om selling the final product is $140,000. You know the Seller’s expected production costs are $10,000 and machine cost was $30,000. 83 cm "058.: 0E. 80.50 0800?.— 808 8000 0.8:: x ._0=0m 80=0m 80 "050.5 055 8.0.: 8000 0.8.5 80> I I 6 E I I a a E - n 0000 008 80508508 50800880 I 00.50 80.50 8on n 8000 088008. 08.9.; n 088 "050.: 80> .3880 0.8.85 580 808.2 m M 08000 080.300 808508 .8808 X 80=0m 055 8.0.... 8080 0.8:: x 80:0m 80> >8 50558 80.50 80 80508508 50800880 50800880 II "III - 80.5 0 80:0m 0808 80> .3000 0.00:0m 3088 80 > 3088 80> 88.9.5 H 80.50 I 88.9.; 080000 80:0m 80508508 50800880 580 "050.5 8000 088008 00088882 055 8.0.5 8000 0.8:: 80> 885 80> .885 80:00 580 088 8003808 >80>0 8000 808 0.8080 08 08 3080.80 08.85 80 > 8088.8 .888 9% 358.5 $5 8.0.: 8000 0.8:: x ._0=0m 8080 0800.000 8% "050.5 80:0w 055 .80.... 8000 0.8:: 80> - n 3000 80.50 0 8 aoflozfinfin 630253 I hobo . 909.0me H 500 van—ONE ooodv~m H 05: a ”050.: 055 80> 3080.80 0.8.8.8 580 80.0— m M 08000 088038 808508 Em808 x 80=0m 80.: 8000 0.8.8. X ._0=0m 80> >8 5058» 8080 80508508 50800880 50800880 I III - 000 a 5:00 0808 8> .08 £2.00 >68 2; >65. 2% 000655 N 80.50 I coed: m "050.: 055 8.0.: 8000 0.8....— 80> 080000 80:0m 058008 00.8083 885 80 > .08000 80508508 508008080 880 05 8880 “08.80.50 05 08 38080 0.885 80> 808.8% 880.080.85.805 84 Recall that your incentive compensation is calculated as follows: Your firm’s cash from this trade/30,000. If following your offer, the Seller’s firm’s cash is greater than $0, Seller X accepts your offer. You receive: ($140,000 — offer)/30,000 = ($140,000- /30,000 = S If following your offer, the Seller’s firm’s cash is less than or equal to $0, Seller X rejects your offer. You receive: $0. 85 QUESTIONS Before you decide how much to offer the Seller, please complete the section that follows. The questions are intended to capture how well you understand the important dimensions of this task. page number: If you are uncertain how to answer the question please consult the appropriate 1. Please check one: I am a Seller. I am a Buyer. Please circle the extent to which you agree or disagree with each statement. 2. My firm’s purchasing strategy affects the cash I will derive from trade. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 3. The seller can use the project-specific machine to make parts for another buyer. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 4. If following my offer, the Seller’s cash from trade is $0 or negative, he will reject my offer. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 5. If the Seller rejects my offer, my incentive compensation will be $0. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 6. I am confident that my computation of my firm’s net cash is correct. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 7. I am confident that my computation of my compensation is correct. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 86 p.1 p.1 an: N— p.2 p.2 p.2 OFFER DECISION Your task: Making an offer to Seller Z Here are details about this prospective project with Seller Z. Seller Z has already purchased a project-specific machine. This machine has no value outside this project with you. Seller Z will incur expected production costs if he/she accepts your offer. Seller’s goal is to maximize his firm’s cash from this trade. Hence, Seller Z is unlikely to accept an offer that does not cover expected production costs and would result in negative cash for his firm. Recall that your offer should be guided by the purchasing strategy of your firm. Self-interested Strategy condition Your firm’s strategy is to offer sellers just enough to cover expected production costs. Your firm usually ignores past machine cost. Your goal is to maximize your firm’s cash. Fair Strategy condition Your firm’s purchasing strategy is to share the net cash evenly between seller and buyer firm. Your firm usually reimburses the Seller for past machine cost and expected production costs. You know you can sell the final product for $170,000. You know the Seller’s expected production costs are $10,000 and the machine cost was $40,000. Please make an offer to Seller Z (your trade decision) and answer the following questions. How much do you offer Seller Z? What will your firm’s cash from this trade be if Seller Z accepts your offer? What will your firm’s cash from this trade be if Seller Z rejects your offer? What is the lowest offer you think Seller Z would accept? 87 WRAP UP QUESTIONS During the task you just performed, how much did you offer Seller Z? Please explain how you came to this decision. Please answer the following demographic questions. All information will be kept anonymous and confidential. 1. Sex (circle one): M F 2. What degree are you currently pursuing (circle one)? Undergraduate Masters MBA Ph.D. 3. Grade point average: 4. National/cultural background: United States or Canada _ Europe _ Asia __ Latin America _ Other (please specify) __ 4. How many 3-credit managerial accounting classes have you taken? 88 How important where each of the following in your decision? Please allocate points to the following factors so that the points total 100. Base your allocation on how important these factors were when you made your trade decision. Point allocated 1. I believed the difference between the cash earned by my firm and the cash earned by Seller Z’s firm should be small. 2. I wanted to reimburse Seller Z for the cost of the project-specific machine he/ she purchased. 3. I did not want to make too low an offer because it would be rejected and my firm’s cash and my incentive compensation would be $0. 4. I wanted to maximize my firm’s cash and my own compensation. 5. Seller Z purchased a machine and I did not. As a result, my firm’s cash fi'om this trade should be lower than Seller Z’s. 6. Making an offer of a few thousand dollars allowed me to maximize my firm’s cash and my incentive compensation. 7. Other. Please explain: Total points allocated (Total must equal 100) Answer the following questions related to the decision you just made regarding Seller Z. Please circle the extent to which you agree or disagree with each statement. (l= strongly disagree and 9= strongly agree): 1. My firm’s purchasing strategy is to share the net cash from trade evenly between Seller and Buyer firms. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 2. I knew exactly what Seller Z’s expected production costs were. Strongly . Strongly disagree agree 1 2 3 4 5 6 7 8 9 3. I used my knowledge of Seller Z’s expected production costs to make Seller Z an ofi‘er that just covered expected production costs. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 89 4. Once a seller has purchased a project-specific machine, this purchase should be reimbursed by the buyer no matter what the buyer’s firm purchasing strategy is. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 5. My firm’s purchasing strategy is to offer sellers just enough to cover expected production costs. Strongly ' Strongly disagree agree 1 2 3 4 5 6 7 8 9 6. Seller Z took a risk in purchasing a project-specific machine because he/she might not be reimbursed for his/her purchase. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 7. I knew with certainty what was the lowest offer Seller Z would accept. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 8. I knew the total of Seller Z’s expected production costs and machine costs but did not know Seller Z’s exact expected production costs. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 9. I believed that Seller Z would reject offers that did not cover his/her expected production costs AND machine cost. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 10. I believed that the machine cost was a sunk cost and should not affect the amount I offered Seller Z. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 11. I considered the risk of my offer being rejected by Seller Z if it did not cover Seller Z’s expected production costs. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 9O Part 2: Trade instructions (Coarse lnformation/Self-interested Strategy and Fair Strategy) You are assigned the role of Buyer. The Buyer buys parts from sellers and assembles them into a final product sold to the outside market. You have an opportunity to undertake a project with a Seller. This project required the Seller to purchase a machine to improve the quality of the parts the Seller manufactures. The machine the Seller purchased is project-specific and can only be used to make the parts you need. It has no value outside this project with you. You cannot purchase these parts from another seller and the Seller cannot sell these parts to another buyer. Your task is to decide how much to offer the Seller for the parts. Your offer should be guided by your firm’s purchasing strategy. Your firm’s cash is calculated by deducting the amount you offer the Seller from the revenue you can earn for selling the fmal product. You will work through an example before deciding how much to offer the Seller. 91 .8000 80000008 0000090 0058: 00>00 000 0000 00.0 0&0 :0 80000 00 300500 mm 00=0m 2:. 0800008000 0 00008 00880 80 0&0 05 000.00 .8 80000 :05 00=0m 2:. .3085 900000.80 0.800 80» .3 0008m 00 08000 0&0 .80 > .25. 5 ,8 0.8080008 8 8.80 08 00.0 0&0 :0 00=0m 08 00008 :05 00 > .8000 .805 308. 00% .0030: .0000 088008 00 8000 00000008 008090 0.008% 05 00:00 305 0.00 00 00 > 0.00008 88m 05 was—0m 80.0 000000 0000098 .80» 305— 00 > 80¢ 02 .8.“ 0000 0.603: 8 0.0000 0803 000 $0.00 0030000.:— 0000098 00.60 008 0000 005 0&0 :0 80000 3 E0008: 8 00=0m 05 .0803 .0008 80.0 0000 0.8.0.0 00.000.— 088088 00 8 03m 0.00:0m 00,—. 000.08% 8 000.0090 8 00000.8 88.0 .80» 0002,00 05 0:0 000 080008 00502 .8000 00000008 0000090 00.008 0305‘ 00=0m 05. 080008 0500900088 08 000000.80 000.080 0.0: 00=0m 0E. AC $0 AC A3 A: cm "000.: 0.5 80.... 0&0 80008 08.80 {0:00 0:00 8 00.5 0.0 EB. :08 0.8.5 80> 8000 00000008 03 I Gav A Amv A Amy CV 3880 .000 0005 80:05 85. "000.: 3.: 80...— w802800 0 80¢ 00» 00mm .0000 05:08: 00:00.. 003008 089800.200 0% so» a 80% s 0.8 80:02.. 0:02.. 8085 00:30:00 080000 00=0m 00.00 00.08 00 > 0080080 0.00:0m 308— 00> 00.00 - 00:05. 308. .0: 00 00 > 000008 0808xm 02:. 0.0 :50 :03 0.8.;— 80> .3200 028000 0.8 000M009 05 m0 00800000 0300000 080 082080 05. 3:800 000.0...— 92 Your compensation is composed of two parts. 1) A fee of $5 that you collect for participating in this research project. 2) In addition, you can cam some incentive compensation. Your incentive compensation is calculated as follows: your firm’s cash from this trade/30,000. 93 INSTRUCTION SHEET Practice example: Your offer should be guided by the purchasing strategy of your firm. Self-interested Strategy condition Your firm’s purchasing strategy is to offer sellers just enough to cover expected production costs. Your firm usually ignores past machine cost. Your goal is to maximize your firm’s cash. Fair Strategy condition Your firm’s purchasing strategy is to share the net cash evenly between seller and buyer firm. Your firm usually reimburses the Seller for past machine cost and expected production costs. Your firm calculates overall net cash as follows: Expected product revenue — expected production costs - machine cost. Below is a practice example that highlights the decision you will be making and its possible outcomes. Please propose an offer consistent with your firm strategy (4) and calculate the outcome of the trade (5) in the timeline below. You will then look at the effect of this offer on your incentive compensation. Seller X has already purchased a project-specific machine to manufacture parts for you. Seller’s goal is to maximize his firm’s cash from this trade. Hence, Seller X is unlikely to accept an offer that does not cover expected production costs and would result in negative cash for his firm. Your expected revenue from selling the final product is $140,000. You do not know either the Seller’s expected production costs or machine cost. However, you know their total is $40,000. 94 ca 800.: 0.00 80... 080 “.80..— N 02.0w 8&0 0808.. cm 800... 8=0m 0.00 80... 080 0.8....— 80> / 6 $0 5 5 I A: u - n 3000 8&0 00> 08.3% 80.03 m 080008 00.80008 008098 I 8&0 .8088 0.8.0.. .80.. >0 u .08. 0.000 308. I 00880 08000 X 800m "000.: 000.8 8&0 8 800m 00.08 00 > .800 080008 80008 008098 0.0. 80.... 0000 0.8.... x 8.8m 8&0 00> .0800 00.80008 .8 0800 0080008 308. 00> u - 080000 800m 008098 000 .000 080008 008098 0.8..0m 08.8; I .80 I 80.9.; 32582 80 05> 80 322 so: 8 8> 800... 8:00 000 8.80 000380 0.0. 80.... 080 0.8.... 80> .0008 080 80 080m 8 0. 308.00. 0.800 80> 80.0.0.0 .008. cm "000... a... as. .03 8:: x 3:00 30° flame a "06:6 —— m 0.0. 80.... 080 0.8.0.. 80> 6 I E 5 S 5 n - u 880 8&0 00> .8088 000.03 I 0%“me 080008 00.80008 008098 I. 8&0 0.8.0.. .80 8 >0 000.0» 8&0 n 80.0. .000... 308. M00008 008098 .0w000 x 800m .000... 00 800m 808 00 > .800 00> 800 080008 0 00 0.0. 80.... 080 0.8.0.. x 8.8m 8&0 00.0008 880». 80 80> .8 0800 00.80008 3 8. > I - 080000 800m .0800 00.80008 008098 008098 m.800m 08.9.. a u 8&0 I 08de 800... 0.0. 80... 080 0.8.... 80> 8>00 0. 000000 .00“ 8&0 308. 80 00 00> 0. m. .3088 0.800 80> 80.0.0.0 00808.0..830 5 9 Recall that your incentive compensation is calculated as follows: Your firm’s cash from this trade/30,000. If following your offer, the Seller’s firm’s cash is greater than $0, Seller X accepts your offer. You receive: ($140,000 — ofi‘er)/30,000 = ($140,000- /30,000_= S If following your offer, the Seller’s firm’s cash is less than or equal to $0, Seller X rejects your offer. You receive: $ 96 QUESTIONS Before you decide how much to offer the Seller, please complete the section that follows. The questions are intended to capture how well you understand the important dimensions of this task. If you are uncertain how to answer the question please consult the appropriate page number: 1. Please check one: I am a Seller. I am a Buyer. p. 1 Please circle the extent to which you agree or disagree with each statement. 2. My firm’s purchasing strategy affects the cash I will derive from trade. Strongly Strongly disagree agree 1 1 2 3 4 5 6 7 8 9 p' 3. The seller can use the project-specific machine to make parts for another buye Strongly Strongly disagree agree 1 1 2 3 4 5 6 7 8 9 p' 4. If following my offer, the Seller’s cash from trade is $0 or negative, he will reject my offer. Strongly Strongly disagree agree 1 1 2 3 4 5 6 7 8 9 2' 2 5. If the Seller rejects my offer, my incentive compensation will be $0. Strongly Strongly disagree agree 2 1 2 3 4 5 6 7 8 9 p' 6. I am confident that my computation of my firm’s net cash is correct. Strongly Strongly disagree agree 2 1 2 3 4 5 6 7 8 9 p‘ 7. I am confident that my computation of my compensation is correct. Strongly Strongly disagree agree p. 2 l 2 3 4 5 6 7 8 9 97 OFFER DECISION Your task: Making an offer to Seller Z Here are details about this prospective project with Seller Z. Seller Z has already purchased a project-specific machine. This machine has no value outside this project with you. Seller Z will incur expected production costs if he/she accepts your offer. Seller’s goal is to maximize his firm’s cash from this trade. Hence, Seller Z is unlikely to accept an offer that does not cover expected production costs and would result in negative cash for his firm. Recall that your offer should be guided by the purchasing strategy of your firm. Self-interested Strategy condition Your firm’s strategy is to offer sellers just enough to cover expected production costs. Your firm usually ignores past machine cost. Your goal is to maximize your firm’s cash. Fair Strategy condition Your firm’s purchasing strategy is to share the net cash evenly between seller and buyer firm. Your firm usually reimburses the Seller for past machine cost and expected production costs. You know you can sell the final product for $170,000. You do not know either the Seller’s expected production costs or machine cost. However, you know their total is $50,000. Please make an offer to Seller Z (your trade decision) and answer the following questions. How much do you offer Seller Z? What will your firm’s cash from this trade be if Seller Z accepts your offer? What will your firm’s cash from this trade be if Seller Z rejects your offer? What is the lowest offer you think Seller Z would accept? 98 WRAP UP QUESTIONS During the task you just performed, how much did you offer Seller Z? Please explain how you came to this decision. Please answer the following demographic questions. All information will be kept anonymous and confidential. 1. Sex (circle one): M F 2. What degree are you currently pursuing (circle one)? Undergraduate Masters MBA Ph.D. 3. Grade point average: 4. National/cultural background: United States or Canada _ Europe __ Asia _ Latin America _ Other (please specify) _ 4. How many 3—credit managerial accounting classes have you taken? 99 How important where each of the following in your decision? Please allocate points to the following factors so that the points total 100. Base your allocation on how important these factors were when you made your trade decision. Point allocated l. I believed the difference between the cash earned by my firm and the cash earned by Seller Z’s firm should be small. 2. I wanted to reimburse Seller Z for the cost of the project-specific machine he/shgurchased. 3. I did not want to make too low an offer because it would be rejected and my firm’s cash and my incentive compensation would be $0. 4. I wanted to maximize my firm’s cash and my own compensation. 5. Seller Z purchased a machine and I did not. As a result, my firm’s cash fi'om this trade should be lower than Seller Z’s. 6. Making an offer of a few thousand dollars allowed me to maximize my firm’s cash and my incentive compensation. 7. Other. Please explain: Total points allocated (Total must equal 100) Answer the following questions related to the decision you just made regarding Seller Z. Please circle the extent to which you agree or disagree with each statement. (l= strongly disagree and 9= strongly agree): 1. My firm’s purchasing strategy is to share the net cash fi'om trade evenly between Seller and Buyer firms. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 2. I knew exactly what Seller Z’s expected production costs were. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 3. I used my knowledge of Seller Z’s expected production costs to make Seller Z an offer that just covered expected production costs. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 100 4. Once a seller has purchased a project-specific machine, this purchase should be reimbursed by the buyer no matter what the buyer’s firm purchasing strategy is. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 5. My firm’s purchasing strategy is to offer sellers just enough to cover expected production costs. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 6. Seller Z took a risk in purchasing a project-specific machine because he/ she might not be reimbursed for his/her purchase. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 7. I knew with certainty what was the lowest offer Seller Z would accept. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 8. I knew the total of Seller Z’s expected production costs and machine costs but did not know Seller Z’s exact expected production costs. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 9. I believed that Seller Z would reject offers that did not cover his/her expected production costs AND machine cost. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 10. I believed that the machine cost was a sunk cost and should not affect the amount I offered Seller Z. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 II. I considered the risk of my offer being rejected by Seller Z if it did not cover Seller Z’s expected production costs. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 101 Part 3: Investment instructions (Fine lnformation/Self-interested Strategy and Fair Strategy) You are assigned the role of Seller. The Seller manufactures parts and sells parts to buyers. You have an opportunity to undertake a project with a Buyer. This project could be mutually beneficial. This project requires you to purchase a machine to improve the quality of the parts you manufacture. If you purchase the machine, the Buyer will purchase the parts from you and assemble them into a final product sold to the outside market. You cannot sell these parts to another buyer and the Buyer cannot purchase these parts from another seller. Your task is to decide whether to purchase the machine that improves the quality of the parts you manufacture for the Buyer. This machine is project-specific and can only be used to make the parts the Buyer needs. It has no value outside this project with the Buyer. Your goal is to maximize your firm’s net cash, and thus your own compensation, by capturing as much of the net cash as possible. Your firm’s net cash is calculated by deducting your costs from the amount the Buyer offers you for the parts. You will work through some examples before making your investment decision. 102 03000.00 9000008.. 0...... 0.80.5 00 00000 0. 8&0 .80. .m ..0 0.008080. 0. 00.08 00. 00.. 8&0 00 00.. 000.08 8.00m .3 8&0 00 00... 00.08 .00 ...3 8.30 I 00.0008 0... 00000.00 .00 00 00.0 .. c 0 8&088000 0 00.08 000000 .00 8&0 00. 80.00.. 00 80000 ...3 00 > .0. .0000 00.0008 80.. 000 0.000 00.80008 008098 80.. 0305. 8.00.. .0 .80008 .00.. 00. 8.0.00 80.... 00008.. 008098 0308. 8.00m E .8000 00. 00 00.0808 00 0. 00008.. 80008 00.. 000 .000 000008 000 .0800 00.80008 008098 .00.. 3000. 00 .r .00.. 80.... 0.08 800000.08 0. 00000800. 00 ...3 8.00m 00... 00.80.8008 00.. 0.8.. 0... ..0 3.80.0 00. 0000.00. 3.0.0.8003 00.0008 00... 00.0.8000 0.00.08 00 ...3 80.008 00. 0.8000 00 > I 00.0008 0... 0000000.. 00.. .— om 2.0.8 80 0.8.5 8.0...— ce .0000 80 0.8.... 8.0.... 8&0 80.00.. 0% .0000 80 0.8.5 80> .000 00.0002 I 00 > .0000 80 0.8.... 80> 6 .3 E I 8&0 .. 30.000 .000 00.0008 osco>o.. «0300.5 wands—9:5 2.0% 9.0 mumoo 05:95.. «0.60.5 .0000 8.. 0.8.... 8.80 . 0 u 000 00.80008 008098 008098 ..ObO MD OFF—G. 8.5%: ...:Oh mace—u— h0~A=m macs—v— kph—am .000 00.0008 I 80000 00> & . m. 0.000 00.80098 008098 I 8&0 .0000 .00 0.8.... 80> .0 ... 00.0008 .000 .0: 00 00> ‘ 5 0800.00 _0.0...0000 0.00.08 0 0.008 0. 00.0008 000 00> 30.00 00.880 0.0 80.08 00. ..0 00800.00 0.0.0000 000 08.08.. 00... “0:080 80.0.... 103 Your compensation is composed of two parts. 1) A fee of $5 that you collect for participating in this research project. 2) In addition, you can earn up to $10 more of incentive compensation based on the net cash generated for your firm. You are endowed with 40,000 experimental dollars to start. The calculation of your incentive compensation is as follows: (40,000 + net cash)/ 10,000 104 INSTRUCTION SHEETS Examples of prior experience with similar projects: Bach Buyer’s offer depends on the purchasing strategy of his/her firm. Below are details about previous experiences you’ve had with two other buyers when you made a project-specific purchase. Example 1: You purchased a proj ect-specific machine for $3 0,000 to manufacture parts for Buyer X because you believed Buyer X would trade in a manner that was mutually beneficial. Buyer X’s final product had expected revenues of $140,000. Yom' expected production costs were $10,000. Buyer X knew that your expected production costs were $10,000 and your machine cost was $30,000. Buyer X’s firm purchasing strategy is to offer sellers just enough to cover expected production costs. Buyer X usually ignores machine costs. Buyer X assumes you normally will not accept an offer that does not cover production costs. 105 8.: "83:8 + 83$ 60282 2,3 283 so» .223... 2: £25.. .2. 2.. a; = 8.5 “83:88.3- 25.3 60282 so» 5:: 9x .35. 3.8.3.. .5» = 2.; "83:88.3- 83$ 62602 a; as... ....x 5.5 82.88 a; = .Eofiaowqo BEE .30» mm ooodv 6863398 B: + 2563 5.323.. 2. 182523 a stanza—=3 95:02: .59» .2: =33— cw 2.28 «2. 95.5 N hob—m Ste 60.? 30> 8:63- 2.23 «2. 95.: 28> 33.2w 5 cs T 5 4 Q r A 3 u 25.. 5 I 25de . n 80 . u . 2.23 .254 E auto x owmhnmmfi ease» ocoowwwwe coo one u 3: x 95.: cob—m Exam .35. .880 EB. 2565 n 8355 382.5... oat—owe Emacs hobo 98:268.“ .550 $80 552695 965— K cob—m so» 32.35; 25.3%. E008 30> 8 353 X Exam 8398 So.» 333:8 on 2:8 u 83. mm who a? “8.85 85:3 I 255 a u 83 a 5_ x m .5» 838m 2.28 «2. ”...—.5 ...—.5 106 a £0.00 .0: 0.0:: X 00.95 a £000 00: 0.5.5 08> 0&0 000.000 6 r :0 6 :IIII 5 . € 0 u 08.03 n . ... . a ..0 0 08000.0 .80 00200:. .50» oooowwwme ooo one H 2.000 00: 0.5.5 00.25 0 a 0 000 0% X 099mm .02; .380 000 ooo.o_m u 380 «0:008 00000 ax 0 0:22.20 29.00 a n > 0050.605 .580 00200005 00.00008 0305. x 00.95 :0» _0_0c0:0o m 0. 3003 X 00.9m— 50» 0305. x .055 N£052: 0: 2:00 . .0088 000:2. £000 .0: 0.5.;— 0=0> 00> 00:000m .0080 030020000 00>00 00: 800 005 0&0 :0 E0000 00: EB £0.80: :0» 008080 x 00.3mm .380 0:300:— 0000:w_ b.0505 x 00.3: .0080 830—50.:— 00000000 0080 8 £9.05 «03. 80:0» 8&0 3 0m 3000.50 at: 0.x 00.3m— 005 05:03 .ooo.omm 003 080 002008 .50» 0:0 ooo.o5 000 380 00302008 0000098 .50» 005 0305— X 00.55 .ooo.o_m 203 380 00303008 0000096 30> .ooo.o3m .00 80:26.. 0000098 00: 00300.5 Bum 0.x 00.9mm .3305: £38.08 003 005 00808 0 E 000.: 2003 X 008m 02,0200 :0.» 080000 X 003mm 00m 00.00 003003008 00 ooo.oma 00.“ 002008 2.3000300 .85 0 00005050 00> .000—0.. 0:202: 05 5 Amy 0.00.: 05 we 0800050 05 0.0.3.00 00 2.0 A3 00.08 5.3 X 005.5 00:0 05 0050.... 3 8000089 X 00.9m— Efiafinefi 0.: 00: .x 00.95 5?: 0050.098 08.:— 050» 00: 107 0:00.30 0000.... 00:0m .000 00.00.. 00003.00. 0.000 .00 ..000>0 00.0 00000.0 00... 00.00 00 00.000 00 0. ~30000.0 90.02.0000. 000.0 9N 09.0.. .0863 003 0000 000.008 0000. .000 000.05 0003 00000 00.80.0000 00000900 000% 005 305. N 0000.. .0006; 0003 00000 00000000. 000000000 000 > .8060; .00 00000>00 000000000 .000. 000.0000 .00.. 0N 00>0m 0.00.0002. 3.0308 003 005 00000.: 0 0. 0.0000 0.003 N 000.3. 005:2. 00.. 000000.. N 0000.. 00.0 0.00.. 000000000000 0. 086mm 00.. 00.0.0000 050000-000 .0000. 0 000000000. 00 V "N 0.0.0000H "00.0.00:— 00.. ...w000. 000 .00.. 00.. ... 000.00 0.." 00.00.. 000.00 00.. .. 000....0 0.x 00.00.. 00.0000 00.. ... 000300000800 0.50000. 000.» 80.00.00 0000.0. 000.030.0000 .030. 0000. 0. ooodv 68.35.0000 .00 + 8060. “030:0. 00 0000.00.00 0. 00.000.000.500 0300000. 000» .0... ...—00¢ 108 000.000 u 000.0... + 000.0. + 0800.00 I ocod. I 0860.8 n .000 000.000. + 0.000 00000000.. 00.00000 + QC000 00.0.0000 - 0.000 000000000 00.00000 I 000000.... 00.0.0 000.008 + 000.0... ”00382 2.00 0.003 000 .2050... 20 0.000.. 00.. 00.. 000 0. 00.00 "000.00.000.00- 000.000 "00282 000 5000 ..N .0000 038.02 =00 0. 00.00 "000.00.000.00. 000.000 .0208. 80 0.0000 ..N .00.... 08082. =00 0 0000030000 .00.... 000.. 0. 000.00 0006:2000 .00 + 806$ ”030:0. 00 00.0.0200 0. 003000000000 02.000... 000.. .0... ..0005 cm 2.000 .00 0.0.0.. N 00.0.5 80.3.0 2.000 .0: 0.0.00 000> 00.0.0 80.00.. .00 $0 .00 .111 .00 A .0 86.60% H MOD 0.... OOOGOEOMMOM COO.OV — a N OOOAVMw H 000.000 .. 000.010 000.000 0.0.00 N 0 .._ 2.025.. 2.280. 0.0000 00.005 .00.... ....000 0:0 000.... a .I. 0.000 .00000.. 00.00000 00.. ._0_0...0:0.. 0.000 .00 0... 00.0.0 02.000000 00.00000 0305. N 00.00.. ...—00.0.: 0.. 2000 0. 0.003 N 00005 000.. 03000. N 000.05 .020... 005:2. 00.. 0000005 2.000 .0: 0.0..... N 00.00.. 00.0.0 .00000 00 > 30.03 H 000.30 I ooo... _ m I 000.000 2.000 .0: 0.0.0.. 000 > 109 m 2.000 .00 0.0.0.. N 09.05 a 2.000 .00 0.0.0.. 000 > 000.0 000.0. oocdm H 256: w H . H mu la 000....0 N .000 000.00.: 000.. 0000.0. 0:.:0%.mm%mm.m00 a 000.5 .00.... ...:0>0 0:0 08...... H 0.000 .000000 00.00000 0.0.. ._0_0...0:0.. "Swan «0: PE...— N huh—am— ...mwo .0: 0... 035m :O_.o=UO.Q EEOUQXD 950:.— N hohzm 37.2.2... On. 2:00 Ila H .0000 0. 0.003 N 00 ..05 .00.. 030:. N 00...... 000.000 00>0:0n m .0080 00> 00.. 0000005 2.000 .00 0.0.0... 000 > ....00>0 0.0.... 00:0m 000 09.05 0003.2. ..000 .00 ..00030 0... 0000.... .0... 00....0 :0 00.0.0 0. 0. 30.00... 0...... EN 09.05 .0... 0E000< .8on 003 .000 000.000. 000.. 000 80.0.0 000 0.000 00.000000 00.0005 000.. .0... 0305. N 09.05 .80... . m 0003 0.000 000000000 00.00000 000 .r .0860; ..0 00000>00 00.00000 00.. .000000 .00... 0.N 09.05 0.0.0000... ....03000 003 .0... 000000. 0 0. 0000. 0.003 N 000.05 00>0..0.. 00.. 000000.. N 09.05 00.. 0.000 000.00.000.00 0. Soda 00.. 00.0000. 0....0000-.00_.000 0 000000000 00 > 30.0.. 0....00... 0... 0. Am. 0000. 0... ..0 00.00.00 0... 30.00.00 0. 000 A... 00.00. ...3 N 09.05 00....0 0... 8.0000 0. 000000000 N 09.05 08.00.0000. 0... 003 .N 00005 ....3 0000.00000 00.00 000.. 00.. 110 “00.0000. 0... 2.9.0.. .00 00.. 00.. .— "00h... 0.N 00...... 30...... 00.. .— 058385 00.... so. 0 80.0.. 08.00208 .2. + 08.8. n "00...... 0.N 09.0.. .00000 00.. .. "00300000080 03.0000. 000.. 80.00.00 0000... 030:0. 00 00.0.0300 0. 00300000500 03.0000. 000.. .0... :33. 1]] QUESTIONS Before you make an investment decision, please complete the section that follows. The questions are intended to capture how well you understand the important dimensions of this investment task. If you are uncertain how to answer the question please consult the appropriate 1. Please check one: Page number: I am a Seller. I am a Buyer. Page 1 Please circle the extent to which you agree or disagree with each statement 2. The Buyer’s firm purchasing strategy affects the net cash I will derive fi‘omt Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 Page 1 3. I can use the project-specific machine to make parts for another Buyer. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 Page 1 4. If I do not purchase the project-specific machine, my firm’s net cash is $0. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 Page 1 5. In the Buyer X example, I am confident that my computation of my firm’s n1 correct Strongly Strongly disagree agree 1 2 3 4 5 6 7 g 9 Pages 2 & 3 6. In the Buyer Z example, I am confident that my computation of my firm’s nc correct. Strongly Strongly disagree agree Pages 4 1 2 3 4 5 6 7 8 9 & 5 7. I am confident that my computation of my compensation is correct. Strongly Strongly disagree agree Pages 2 l 2 3 4 5 6 7 8 9 t05 112 INVESTMENT DECISION Your task Buyer B asked you to make a project-specific purchase decision. This project could be mutually beneficial. Here are further details about your prospective project with Buyer B. You have not worked with Buyer B before. Self-interested Strategy You believe that there is a 75% chance that Buyer B’s firm purchasing strategy is to offer just enough to cover expected production costs. You believe there is a 25% chance that Buyer B’s purchasing strategy is to make an offer that shares the overall net cash between Buyer and Seller firms evenly. Fair Strategy You believe that there is a 75% chance that Buyer B’s firm purchasing strategy is to make an offer that shares the overall net cash between Buyer and Seller firms evenly. You believe there is a 25% chance that Buyer B’s firm purchasing strategy is to offer sellers just enough to cover expected production costs. The amount of sales revenue that Buyer B can expect for the final product is contingent upon whether you purchase the proj ect-specific machine: Your machine cost Buyer B sales revenue $0 (no purchase made) $0 $40,000 $1 70,000 If you purchase the machine — This machine has no value outside this project with Buyer B. In addition to the machine cost of $40,000, you expect to incur production costs of $10,000 if you accept Buyer B’s offer. If you do not purchase the machine — Your firm’s net cash will be $0. Please use the information Buyer B knows to answer the following questions. Buyer B knows he/she can sell the final product for $170,000. 113 Buyer B knows that your expected production costs are $10,000 and your machine cost is $40,000. If Buyer B’s firm strategy is to offer just enough to cover expected production costs, what do you think Buyer B will offer you? What will your firm’s net cash be if you accept Buyer B’s offer? If Buyer B’s firm strategy is to share the overall net cash between Buyer and Seller firm evenly, what do you think Buyer B will offer you? What will your firm’s net cash be if you accept Buyer B’s offer? Please make your investment decision. Select the cost of the machine you would like to purchase: _$0 (equivalent to no purchase made) _$4o,ooo How much do you anticipate Buyer B will offer you? What is the lowest offer you would accept? 114 WRAP UP QUESTIONS During the task you just performed, did you decide to purchase the project-specific machine? _Yes __ No Please explain how you came to this decision. Please answer the following demographic questions. All information will be kept anonymous and confidential. 1. Sex (circle one): M F 2. What degree are you currently pursuing (circle one)? Undergraduate Masters MBA Ph.D. 3. Grade point average: 4. National/cultural background: United States or Canada _ Europe _ Asia _ Latin America _ Other (please specify) __ 4. How many 3-credit managerial accounting classes have you taken? 115 How important where each of the following in your decision of whether to purchase the machine? , Please allocate points to the following factors so that the points total 100. Base your allocation on how important these factors were when you made your purchase decision. Point allocated 1. I believed the difference between the net cash earned by my firm and the net cash earned by Buyer B’s firm should be small. 2. I was concerned that Buyer B would not reimburse me for the project-specific machine cost. 3. I considered my opportunity to earn an incentive compensation greater than $4. 4. I considered the opportunity to generate a large overall net cash to share with Buyer B’s firm. 5. I considered the opportunity for my firm to earn a positive net cash. 6. I took into account my risk of earning an incentive compensation lower than $4. 7. I made an investment and Buyer B did not. As a result, my firm’s net cash should be larger than Buyer B’s. 8. Other. Please explain: Total points allocated (Total must equal 100) Answer the following questions related to the investment decision you just made regarding Buyer B. Please circle the extent to which you agree or disagree with each statement. (l= strongly disagree and 9= strongly agree): 1. I believed purchasing the machine and subsequently trading with Buyer B was likely to be mutually beneficial. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 2. Buyer B’s firm purchasing strategy is to share the net cash evenly between both fmns. Strongly ' Strongly disagree agree 1 2 3 4 5 6 7 8 9 116 3. I would invest if I knew with certainty that Buyer B would share the overall net cash evenly. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 4. Buyer B’s firm purchasing strategy is to cover expected production costs. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 5. I believed Buyer B knew exactly what my expected production costs were. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 6. Even if Buyer B’s firm policy was to offer just enough to cover production costs, there was a chance I might get reimbursed for the cost of the machine I purchased. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 7. I believed Buyer B would use his knowledge of my expected production costs to make me a low offer. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 8. I believed any offer from Buyer B that covers expected production costs and ignores machine cost would be unreasonable because all related costs (past machine cost and expected production costs) should be covered. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 9. I believed Buyer B’s net cash would be much higher if I purchased the machine and accepted his offer than if I rejected his offer Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 10. I believed Buyer B knew the total of my expected production costs and machine costs but did not know my exact expected production costs. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 117 11. I ignored the 25% chance that Buyer B’s firm might have a different purchasing strategy. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 118 Part 4: Investment instructions (Coarse Information/Self-interested Strategy and Fair Strategy) You are assigned the role of Seller. The Seller manufactures parts and sells parts to buyers. You have an opportunity to undertake a project with a Buyer. This project could be mutually beneficial. This project requires you to purchase a machine to improve the quality of the parts you manufacture. If you purchase the machine, the Buyer will purchase the parts from you and assemble them into a final product sold to the outside market. You cannot sell these parts to another buyer and the Buyer cannot purchase these parts from another seller. Your task is to decide whether to purchase the machine that improves the quality of the parts you manufacture for the Buyer. This machine is project-specific and can only be used to make the parts the Buyer needs. It has no value outside this project with the Buyer. Your goal is to maximize your firm’s net cash, and thus your own compensation, by capturing as much of the net cash as possible. Your firm’s net cash is calculated by deducting your costs from the amount the Buyer offers you for the parts. You will work through some examples before making your investment decision. 119 3&0 :0 :0» 00.08 :0: :05 Exam I 0:2an 05 008—0...:— «0: 00 :0.» .— .:020008000 0 0028 8:50 :5 0&0 05 000 _.0.: :0 E0000 EB 20> 0900800 m800208: 8:0 0.00.?m :0 000.3 mm 00000 .80.; .00 3:080:08 8 0:0: 05 .8.“ 0&0 :0 :0» 000008 Exam .0000 .205 0.30:0. 0m 0000 0:288 .80» :0 8000 8005008 0000090 80% 0052 305— 00: 0000 Exam 00:00:: 08m 2.0 328 80.0 080.00.: 0000098 030:0— 000:5 .0095 02: .3 00:80:0w 0n 0: 00:05.. 00:00:: 05 0:: 0000 0:288 05 .3000 8000000: 0000098 80.» 305 00> .00» 800 3:0: m800208: 8 000000008 0a :05 Basm 02,—. 0800.858 :0.» 3.8: 05 m0 5:050 05 00000008 3208333 0:20.08 05. 2208.252 30:88 02 :05 003.00: 05 05:02 3% I 0:238 05 03:00:: :0.» «— cm 2.08 .0: 0.8.6 00.9m— ._0.¢0 :00? 0000 0:282 I 00> 2.000 00: 0.80: 80> 6 1 _ 0&0 - .8880 00:00: 00:00.5 920.0208: " 000.0: 0.8.. 00.? , 8580003 __ a .0000 0&0 00x08 08m $00 urge—Ow:— 1 g 30> 3000 8000005 080098 I 0&0 £000 00: 0.8.6 80> cm 2000 .0: 0.8.5 00.95 cm 2.000 .0: ...—Eu 80> A: A3 c; CU Am. A: 0:208: .92 .0: 00 00> $0 6 1 5 E ...:2 :05 6.305. 0: .0000 0:_ :09” 00:30.. .0008: 0:.meehMWMWM .0 5000 500000.. 0000096 0000098 .50» 030:0. 08m 0. 0:205: >3 00> 30:0. :0: 0000 Erasm— .30_0n 00:800 000 000.88 05 .«0 0080080 02600: 0:0 08—080. 0E. 3:300 000.0...— 120 Your compensation is composed of two parts. 1) A fee of $5 that you collect for participating in this research project. 2) In addition, you can earn up to $10 more of incentive compensation based on the net cash generated for your firm. You are endowed with 40,000 experimental dollars to start. The calculation of your incentive compensation is as follows: (40,000 + net cash)/ 10,000 12] INSTRUCTION SHEETS Examples of prior experience with similar projects: Each Buyer’s ofi‘er depends on the purchasing strategy of his/her firm. Below are details about previous experiences you’ve had with two other buyers when you made a project-specific purchase. Example 1: You purchased a proj ect-specific machine for $30,000 to manufacture parts for Buyer X because you believed Buyer X would trade in a manner that was mutually beneficial. Buyer X’s final product had expected revenues of $140,000. Your expected production costs were $10,000. Buyer X did not know either your expected production costs or your machine cost. However, he knew their total was $40,000. Buyer X’s firm strategy is to offer sellers just enough to cover expected production costs. Buyer X’s firm usually ignores machine costs. Buyer X assumes you normally will not accept an offer that does not cover production costs. 122 00.3. "80.008 + 08.9.. 03.82 2,2 283 a; .228... 2: £0.50 .2. 02. 0% = 0.20 "08.00.800.00- 08.9.. 00:82 a; 50.. 0.x 30.0 038.9. 0% : 03.0 n 08.0088: + 80.9.. 05 3:00 "08.2.0003- 80.3.. 08303 02082 8» ......00 0.x .00.: 02:83 0% 8 8083005 .002. 80% 0. 08.9. 6006:2000 00: + 086$ ”0.30:0. 00 0000.00.00 0. 8500:0083 02800:. ...—0.» 0020 :30: cm 2.000 00: 0.8.... x 00.9.: 00.00 000.0.— §.enm.. 00> 2.000 00: 0.8....— 80> 08.30" 6 :1 0.. 6 a. E 08.30-08.900 oh 08.5: n .833 80.80 n 08.9.; I 08.8» n 80.3% I 80.00; 80:... 0:0 80.?“ :003002 .58 0.020 0305. 0.. 00:05.. 082008 29002 "00000.. 0:00:00 :0 000.200 .0000 0:2008 000% 00000.8 0000098 00» 20.0.0052 2000 .0: 0.8.... x 00.95 0&0 02 .3000 5.000000: :0 00000 00000008. 0305. x 00%0m 2.0808 02 0.000 3...; u .5on 000000 00 > 00030 20:00:: 0. X 000009”0 .50» 305. «00.0.... 00>0=02 - 08.20 - 80. _ a 8. 30.0 838: .080 .2. 080 x 30.0 .50 3.380 80.5- :0000008 00>00 . u 08.000 3 053 x 00.0 I 000....» I 002mg 80...”. 80000.. 2000 00: 0.8.5 80> 123 w ...000 .0: 0.0:... X 00...... 0 "Swan «0: FE: L=¢> n IE. Ill 0 n m 80.... .0093. ..000 .0: 0.0:... X .0»...— an 0 0. mu m 60.... .0095. ..000 .0: 0.0:... ..:0> .0....0 .000. Ste ...0000 :0 > 25.00% N 25505 H . _0.0. ..0... 0305. 0::0>0. 000 0mm M 000.50.. 8:080 20...... 0.. ..000 0:300... 8:00.... 00.0093 05...”... ...w:0.. 0.. 0.000 :0..0:00... .0 0.000 02.0000... 030:. X .0z:m .. :0 .0.0...0:0.. 50.. ..0 0.8.000: 00.0093 .00.. 305. __0:w:.m_ 0.. 0.:00 0. X .003. 00:000m. .0: 0000 X .003. .000. 00>0._0.. . :0» 00:000.. 0.08 02.0000... .0>00 0. 0.:03 X .003. 0.000 500000... .0>00 .0: 00 .0... 0.0...0 ...0000 .0: ...? 3.08.0: :9. 005:000 X .0000. 0.000 0:...00... 00.000. ....000: ...... 0.X .003. 0.000 :0..0:00.... 00.0093 00.000 0. 00:05 .00.. 0.2.00 00...: 0. 0. 30.0.0 E0... 0.X 00...;— .0... 0.00000. .ooodcm 0. .80. ..0... 0.50.... 0... 0.0532. ..000 0:...000. 50.. .0 0.000 02.0000... 00.0098 .00.. .05.... .505. .0: 0000 X .003. 08.0.0 203 0.08 :0..0:00... 00.0096 50> 68.9.5 ..0 30:05.. 00.0096 00.. .0000... .0:... 0.X 0.3m. ..0.0...0:0.. 0:033. 003 .05 5:00.: 0 :. 000.. 0.003 X .055 00>0..0.. :9. 000000.. X .003. .o.. 9.0.. 08.00.0000. 0. 086mm .0. 0:...008 050000.003... 0 0000005.. :0 ..r 330—0.. 00:00... 0... 0. Am. 000... 0... ..0 0500.00 0... 30.00.00 0. 0:0 A... 3.0:. ......» X 000.5 .0:: 0... 3.00.... 0. 00000000.. X .99.. 0050.58... 0... 002 .X .025 ......» 00:0...095 00.... .50.. 003 124 00.50.00 05.... ..0..0m 0:0 .055 500.50.. ..000 .0: ..0..0>0 0... 00.0..0 .0... 00....0 :0 00.05 0. 0. 30.2.0 5...... 0.N .055 .80de 00.5 .0.0. ..0... 30:0. 0.. 50.00302 .00.. 055.005 50.. .0 0.000 5050500... 00.0098 50.. .0:..0 30:0. .0: 0.0 N .055 .0865 0.03 0.000 5050500.... 00.0098 50> .ooodv . m ..0 35:05. 080098 00.. .0500... .0:... MN .055 ..0.0...0:0.. 3.0555 003 .0... .0505 0 :. 000.. 0.50.... N .955 00>0..0.. 50.. 005000.. N .055 .0. 0...... 05.00.5505 0. 000.03 5.. 05.00:. 05.02.0000 .8... 0 0000505.. 50> "N 0.5500.— .0:...005 0... 5.050.. .0: 00.. 50.. .— 20....0 0.X 00.55 .00_.0.. 50.. .. m 0:0 w 5003.05 v.0...0 0.X 00.0.5 ...0000 50.. .— .:0300505500 05500:. 50.. 80.50.00 0000... 55053005 .055. .50.. 0. 2.0.0.. 6006.508 .0: + 80.00. ”0.50:0. 00 00.0—50.00 0. 50300500500 0.0.5005 .50.. .0... ..0005 125 08.80 H 08.0.. + 088.9. .. 000.005. H.000 000.00.: 000 0.000 0000500... 00.00.08 ..0 .0.0. + m\..000 000.00.: 000 0.000 0000500... 00.0093 ..0 .80. I 05005.... 80.0 $8.008 + 80.0.0 "03.82 2,0. 2.53 so. .2000... 20 0.0.5.. .2. 02. .5. 0 8.: "000.00.08.00 80.0... "02082 8. .0.... ....N .00... 038...: .5. 0 0000 "000.00.08.09. 08.0... "02082 .5. .3... ..N .0...... 03038 .5. .. 000830000 .0...:. .50.. 0. 000.0... .80.... 5.000 .0: + 80.00. ”030:0. 00 00.0.5300 0. 00..0000..0.00 0.0.0000. .50.. .0... ..000~. cm 2.000 .0: 0.005 N 3.5.. .0b0 .00_.0. 000.3% 2.000 .0: 0.0.... .50 > E I 00 5 , .0 TI: .0 . 000.9% . I . g: cmw fl . H .50. .3... 0305. 8o 90; l coo omm 000.00% I 00060; ....ooo com 0.0....0 N 0... .000 002.000. 05:0>0. u 000.00.: ...w50.. 2.000 .00 0.0:... N 3.5.. .0..5m. .05.... ....00>0 .0 0.000 0 0.. 0.500. .. .0500... 00.00.“...0 50.. ..0.0...000.. . .0....0 ...00000 .0H30... 0....050 00.00....0 .50.. 3000. 0305. N .0 5m ....05.50. 0.. 0.500 ...... 0mm n ...0000 50> . .00 N .0 5m .00 0000 N .0 ..5m. 60.00... 00>0..0.. 000.03 I 000.05 I 000.0% 50.. 005000m. "Ana “0: 95¢ .56» 126 a 2.000 .0: 0.0.... N .0..5m. a 2.000 .0: 0.0.... .50 > .0.....0 .00_.0._ 000.com .I. .06. ..0... 030.0. 0.. 80.0.... n 000.00. M .. . m ..000 0.....00... .50.. 0500.5. 00.02.00 ...w50.. mu 0.0..0 N .0 5m 05.... .0 0.000 0000500... .0500... 00.00000 50.. .0.0...0:0.. m .0..0 ....00>0 ..000 .00 0... 00.00....0 .50.. 30.0. 030.5. N .0..5m. ....05.5.: 2.000 .00 0 0.... N .0..5m. .0080 50> 0.0.0 0. 0.003 N 0.5m. .0: 0000 N .0..5m. 0.. 0.500 .Illw 80.8... 00>0..0.. H 50.. 005000m. m 2.000 .0: 0.0.... .50> ....:0>0 0.0.... .0..0m 0:0 .0..5m. 0003.2. ..000 .0: ..0.0>0 0... 00.0.0 .0... .0....0 :0 3.0.: 0. 0. 30.0.... 0.... 0.N .0..5m. .0... 0.0500< 600.com 0. .0.0. ..0... 0300.. 0.. ..0>030.. ..000 000.00.: .50.. .0 0.000 00.8500... 00.0098 .50.. .0....0 300.. .0: 0000 N .0..5m. 686.. 0.03 0.000 00.8500... 0800...... .50 > .0860... .0 00500>0. 00.00....0 00.. .0500... .0:... 0.N .0..5m ..0.0..000.. ....05.5.: 003.0... .00000. 0 0. 000.. 0.503 N .0..5m. 00>0..0.. 50.. 005000.. N .0..5m. .0. 0...... 05.00.5000. 0. 80.03 .0. 0:...00.: 006000.00 ..0... 0 0000005.. 50> 30.0.. 0:..00... 0... 0. Am. 000.. 0... ..0 00.00.50 0... 80.50.00 0. 000 A... 3.00. ...3 N 3.5.. .0....0 0... .0.00... 0. 00000000.. N .0..5m. :0..0.:.0..:. 0... 00.. .N .003. ....3 00:0..0000 .0.... .50.. 00.. 127 "00.000... 0... 00050.. .00 0.... 00.. .— ".0...... 0.N .90.... .000... 500 .. 0583020 .00... so. 0. 08.8 08.00208 .0: + 08.90 H ".0...—0 0.N .00.... .0000: 00% .. "00.300.00.000 03.0000. .500 0.0.00.3 0000... 030:0. 0.. 0000.00.00 0. 00300000080 02.0000. ...—0.. ...... :33— 128 QUESTIONS Before you make an investment decision, please complete the section that follows. The questions are intended to capture how well you understand the important dimensions of this investment task. If you are uncertain how to answer the question please consult the appropriate page number: 1. Please check one: I am a Seller. Page 1 I am a Buyer. Please circle the extent to which you agree or disagree with each statement 2. The Buyer’s firm purchasing strategy affects the net cash I will derive from 1 Strongly Strongly disagree agree Page 1 1 2 3 4 5 6 7 8 9 3. I can use the proj ect—specific machine to make parts for another Buyer. Strongly Strongly disagree agree Page 1 l 2 3 4 5 6 7 8 9 4. If I do not purchase the project-specific machine, my firm’s net cash is $0. Strongly Strongly disagree agree Page 1 l 2 3 4 5 6 7 8 9 5. In the Buyer X example, I am confident that my computation of my firm’s nc correct Strongly Strongly disagree agree Pages 2 1 2 3 4 5 6 7 8 9 & 3 6. In the Buyer Z example, I am confident that my computation of my firm’s nc correct. Strongly Strongly disagree agree Pages 4 l 2 3 4 5 6 7 8 9 & 5 7. I am confident that my computation of my compensation is correct. Strongly Strongly Pages 2 disagree agree to 5 l 2 3 4 5 6 7 8 9 129 INVESTMENT DECISION Your task Buyer B asked you to make a project-specific purchase decision. This project could be mutually beneficial. Here are further details about your prospective project with Buyer B. You have not worked with Buyer B before. Self-interested Strategy You believe that there is a 75% chance that Buyer B’s firm purchasing strategy is to offer just enough to cover expected production costs. You believe there is a 25% chance that Buyer B’s purchasing strategy is to make an offer that shares the overall net cash between Buyer and Seller firms evenly. Fair Strategy You believe that there is a 75% chance that Buyer B’s firm purchasing strategy is to make an offer that shares the overall net cash between Buyer and Seller firms evenly. You believe there is a 25% chance that Buyer B’s firm purchasing strategy is to offer sellers just enough to cover expected production costs. The amount of sales revenue that Buyer B can expect for the final product is contingent upon whether you purchase the proj ect-specific machine: Your machine cost Buyer B sales revenue $0 (no purchase made) $0 $40,000 $170,000 If you purchase the machine — This machine has no value outside this project with Buyer B. In addition to the machine cost of $40,000, you expect to incur production costs of $10,000 if you accept Buyer B’s offer. If you do not purchase the machine — Your firm’s net cash will be $0. Please use the information Buyer B knows to answer the following questions. Buyer B knows he can sell the final product for $170,000. 130 Like Buyer X and Buyer Z, Buyer B does not know either your expected production costs or your machine cost. However, Buyer B knows their total is $50,000. If Buyer B’s firm strategy is to offer just enough to cover expected production costs, what do you think Buyer B will offer you? What will your firm’s net cash be if you accept Buyer B’s offer? If Buyer B’s firm strategy is to share the overall net cash between Buyer and Seller firm evenly, what do you think Buyer B will offer you? What will your firm’s net cash be if you accept Buyer B’s offer? Please make your investment decision. Select the cost of the machine you would like to purchase: _$0 (equivalent to no purchase made) _$40,000 How much do you anticipate Buyer B will offer you? What is the lowest offer you would accept? 131 WRAP UP QUESTIONS During the task you just performed, did you decide to purchase the project-specific machine? _Yes _ No Please explain how you came to this decision. Please answer the following demographic questions. All information will be kept anonymous and confidential. 1. Sex (circle one): M F 2. What degree are you currently pursuing (circle one)? Undergraduate Masters MBA Ph.D. 3. Grade point average: 4. National/cultural background: United States or Canada __ Europe __ Asia _ Latin America _ Other (please specify) _ 4. How many 3-credit managerial accounting classes have you taken? 132 How important where each of the following in your decision of whether to purchase the machine? Please allocate points to the following factors so that the points total 100. Base your allocation on how important these factors were when you made your purchase decision. Point allocated 1. I believed the difference between the net cash earned by my firm and the net cash earned by Buyer B’s firm should be small. 2. I was concerned that Buyer B would not reimburse me for the project- specific machine cost 3. I considered my opportunity to earn an incentive compensation greater than $4 4. I considered the opportunity to generate a large overall net cash to share with Buyer B’s firm. 5. I considered the opportunity for my firm to earn a positive net cash 6. I took into account my risk of earning incentive compensation lower than $4 7. I made an investment and Buyer B did not. As a result, my firm’s net cash should be larger than Buyer B’s. 8. Other. Please explain: Total points allocated (Total must equal 100) Answer the following questions related to the investment decision you just made regarding Buyer B. Please circle the extent to which you agree or disagree with each statement. (l= strongly disagree and 9= strongly agree): 1. I believed purchasing the machine and subsequently trading with Buyer B was likely to be mutually beneficial. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 2. Buyer B’s firm purchasing strategy is to share the net cash evenly between both firms. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 133 3. I would invest if I knew with certainty that Buyer B would share the overall net cash evenly. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 4. Buyer B’s firm purchasing strategy is to cover expected production costs. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 5. I believed Buyer B knew exactly what my expected production costs were. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 6. Even if Buyer B’s firm policy was to offer just enough to cover production costs, there was a chance I might get reimbursed for the cost of the machine I purchased. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8. 9 7. I believed Buyer B would use his knowledge of my expected production costs to make me a low offer. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 8. I believed any offer from Buyer B that covers expected production costs and ignores machine cost would be unreasonable because all related costs (past machine cost and expected production costs) should be covered. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 9. I believed Buyer B’s net cash would be much higher if I purchased the machine and accepted his offer than if I rejected his offer. Strongly . Strongly disagree agree 1 2 3 4 5 6 7 8 9 10. I believed Buyer B knew the total of my expected production costs and machine costs, but did not know my exact expected production costs. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 134 11. I ignored the 25% chance that Buyer B’s firm might have a different purchasing strategy. Strongly Strongly disagree agree 1 2 3 4 5 6 7 8 9 135 3 REVIEW OF THE HOLD-UP LITERATURE IN THE CONTEXT OF STRATEGIC DECISION-MAKING 3.1 INTRODUCTION "Suppose the production of a particular product requires a large capital equipment which is, however, specialized insofar that it can only be used for the particular product concerned or can only be readapted at great cost. Then, the firm producing such a product for one consumer finds itself faced with one great risk - that the consumer may transfer his demand elsewhere or that he may exercise his monopoly power to force down the price... " C oase correspondence to Ronald Fowler in I 932. In this essay, I review and classify analytical and empirical research aimed at finding and testing mitigating mechanisms to hold-ups. This research is organized around strategic decision-making such as organizational design, trade decisions, and resource allocations. Such a review is warranted despite recent reviews by Shelanski and Klein (1995) and Coeurderoy and Quélin (1997) for at least three reasons: (1) First, those reviews presented case studies and archival research, but did not incorporate findings from analytical or experimental studies. (2) Second, extant reviews did not extend outside the fields of law, economics, marketing, and management. (3) Third, research has evolved in two new directions recently: Research has focused increasingly on relational contracting and has just started to investigate how investor private information affects opportunism. I propose that integrating findings from various disciplines is necessary as each discipline has preferred areas of investigation, but only provides a partial picture of the status of research in other disciplines. For instance, findings from intra-firm and transfer pricing research conducted by accounting are not included in analyses from economics or 136 supply chain. Additionally, results from other disciplines can help point to weaknesses and suggest boundaries of prior research. To that effect, Che and Hausch (1999) show that establishing an initial contract (a solution to hold-ups proposed by accounting research (cf. Edlin and Reichelstein 1995)) encourages idiosyncratic selfish investment but is ineffective with cooperative investments. Finally, since research in accounting is organized by types of strategic decision, scholars risk missing insights fiom research conducted to address another decision even though both address hold-up problems. For instance, aggregation of information helps mitigate opportunism in a resource allocation setting (cf. Arya et al. 2000), but it is not until the first essay presented herein that the effectiveness of this solution was examined with a trade decision. Accordingly, this essay has three main objectives. First, although transaction cost economics has provided the dominating framework for analyzing hold-ups, this essay reviews the various theoretical perspectives that have been used to support analyses of hold-ups highlighting their common features and differences. Second, I integrate research from the fields of economics, experimental economics, accounting, finance and supply chain and organize it around decisions such as organizational design, trade decisions, and resource allocations. Third, I suggest avenues for future investigation. Firms make strategic decisions such as organizational design, trade decisions, or resource allocations on a regular basis. These strategic decisions are often associated with investments in idiosyncratic assets (i.e., relationship-specific investments) by one or both of the parties involved. Relationship-specific investments can generate a surplus when the investor subsequently deals with a specific economic partner. However, these investments lose most, if not all, their value in alternative uses. Economic theories (i.e., 137 transaction cost economics, property rights, and agency theories) predict that the investor anticipates that the specific economic partner will behave in an opportunistic manner by appropriating most of the surplus generated by the investment. As a result, the investor will bear the cost of the relationship-specific investment and only receive a small portion of the surplus created by his investment. Accordingly, he under-invests (i.e., hold-up problem).1 The hold-up problem studied herein is composed of three elements: (a) a socially optimal relationship-specific investment by one party which creates quasi-rents, (b) lack of credible commitment by the non-investor to refrain from appropriating those quasi-rents, and (c) opportunism leading to appropriation of a portion of the quasi-rents by the non-investor. Scholars propose that opportunistic behavior by the non-investing party can be curtailed if interdependence between the parties is increased or there is protection of the investor. Thus, various theories prescribe integration and centralization, detailed and long-term formal contracts, allocation of property rights, incentives, and social norms as ways to reduce the likelihood that the surplus generated by the investment will be appropriated. Scholars also developed analytical models and tested the predictions of those theories with empirical studies. While their findings are generally consistent with the economic and relational contracting theories, they also show that information characteristics can play a role in mitigating hold-ups. Section 3.2 presents the hold-up problem in greater detail and provides theoretical predictions from the field of economics and law with presentation of transaction cost 1 Although the term of hold-up was first used by Goldberg (1976), the concept originated with Coarse (1937) and evolved from Goldberg’s original meaning of appropriation of surplus to underinvestrnent in idiosyncratic assets because of anticipation of subsequent appropriation. 138 economics, property rights, agency and relational contracting theories. Section 3.3 details investigations and findings from research aimed at curtailing the hold-up problem. It is organized around strategic decisions of organizational design, trade decisions, and resource allocations. Section 3.4 concludes. 3.2 HOLD-UP AND OVERVIEW OF THEORIES Early economic investigations initiated by Coase (1937) examined relationships between firms and focused on explaining antecedents of organizational boundaries and contracting arrangements by investigating governance structures ranging fiom arm’s length relationships (price based governance and outcome-based contracts) to integration (governance through a unified authority structure). The premise of this research stream is that organizational boundaries are the result of cost minimizing decisions. Theories were refined much later with transaction cost economics (Williamson 1975), property rights (Hart and Moore 1990), agency (Jensen and Meckling 1976), and relational contracting (Macneil 1980) theories. Theories that analyze the issues associated with relationship- specific investments are presented below following a detailed description of relationship- specific investments and hold-ups. 3.2.1 Description of Hold-ups Per Williamson (1983; 1991), relationship-specific investments take the form of site specificity (e.g., collocation of electricity generating facility next to a coal mine (Joskow 1987)), capital asset specificity (e. g., special tooling (Monteverde and Teece 1982)), a dedicated asset for one customer, human capital (e. g., training of personnel to distribute or service products (Anderson 1985; Jensen and Rothwell 1998)), brand name 139 capital (e.g., goodwill (Anderson 1994)), and temporal specificity (e.g., threats of delays as a way to extract price concessions (Masten, Meehan, and Snyder 1991; Pirrong 1993)). When a firm makes a relationship-specific investment, the more specific the cost is, the lower the value of the investment in alternative uses.2 Expectations of self-interested behavior once a relationship-specific investment has occurred lead to under-investment in idiosyncratic assets and the hold-up problem as detailed in Figure 3-1 below. Figure 3-1 Simplified Timeline of the Hold-up Problem Stage 1 Stage 2 0 1 2 T l l Incomplete Relationship-specific Surplus division contract signed investment choice Adapted fi'om Hart and Moore (1988) A contract was negotiated at time 0 (for a predetermined quantity of a product, or price, or provision of certain effort). The price or quantity cannot be specified with certainty ex ante because they depend on the resolution of certain parameters that cannot be predicted (e.g., success of R&D efforts). At time 1, one party must decide whether to make a non-contractible idiosyncratic one-period investment that will create a surplus 2 Irnportantly, what determines whether an investment is relationship-specific is not whether the costs incurred are fixed or variable, but whether the asset can be redeployed and what its value in alternative uses is. 140 should a trade or compensation agreement follow. Time 1 represents the investment stage. The investor’s goal is to maximize his payoff fiom negotiation. Accordingly, his investment decision is based on expectations from the negotiation to be conducted at time 2. In other words, the investor uses backward induction to make his investment decision. The parties divide the surplus at time 2. Time 2 represents the negotiation stage. Once a relationship-specific investment has been made, it is less attractive for the parties to negotiate outside the relationship than within. Additionally, negotiating parties cannot be fully protected by contracts because contracts are inherently incomplete. Contract incompleteness arises, first, because of uncertainty. Uncertainty means that it might not be possible to include all possible contingencies in contracts. Second, performance might be difficult and costly to monitor as outcomes are unobservable or unverifiable (Klein 1980). Because of the incomplete nature of contracts, court enforcement of contracts is difficult (Klein 1996) and contracts cannot fully protect the investor fiom opportunism by his counterpart. Thus, a bilateral monopoly is created ex post whereby the most powerful party can appropriate the surplus generated by the investment if cooperative behavior is not consistent with self-interested behavior (Tirole 1988). An example of the hold-up problem suggested by the automobile industry follows. Suppose that an automobile supplier must decide whether to invest in R&D that can lead to valuable improvements in a component of a specific model of automobile produced by an original equipment manufacturers (OEM). R&D is costly and does not have any value outside the relationship with this OEM.3 The supplier’s R&D investment decision is based on expectations of payoffs from eventual purchase of the component by 3 In this setting, R&D constitutes an idiosyncratic asset, i.e., relationship-specific asset. 141 this OEM. Additionally, contracts are incomplete and the OEM cannot credibly commit to reimbursing the supplier for the R&D investment. The supplier realizes that once the R&D investment is made, it becomes a sunk cost that may be ignored in trade negotiations with the OEM. In other words, the supplier anticipates that the OEM will only pay for the component’s marginal production costs and not the sunk investment in R&D. Foreseeing the potential for the OEM’s self-interested behavior during trade negotiations, the supplier under-invests in R&D in relation to the socially optimal investment.4 Henke finds evidence to that effect when conducting an annual survey of American automobile suppliers and American OEMs relationships (cf. Henke 2004, 2006). Analyses of hold-ups are grounded in four main theories. Transaction cost economics, property rights, agency and relational contracting theories view hold-ups are an agency problem whereby self-interested behavior dominates to the detriment of cooperative behavior. All propose to explain antecedents to hold-ups and offer possible solutions to the hold-up problem. 3.2.2 Transaction Cost Economics Transaction cost economics (thereafter TCE) focuses on efficiency and minimizing the transaction costs that are incurred ex ante when contracts are written and negotiated, and ex post as a result of monitoring and enforcing contracts or of any contract breach (cf. Klein, Crawford, and Alchian 1978; Williamson 1975, 1979, 1985). The unit of analysis is the transaction. TCE is based on assumptions of bounded rationality of the negotiating parties (i.e., contracts are inherently incomplete) and 4 Making a socially optimal investment (i.e., the investment that maximizes the potential surplus from trade net of investment costs) defines investment efficiency. 142 uncertainty (environmental, technological and related to information asymmetry). Transaction cost economics also assumes that, as a result of incomplete contracting, parties cannot credibly commit to not renegotiate although it would be optimal to do so. Finally, an important premise of transaction cost economics is that parties are motivated by self-interest. Thus, given the opportunity, the non-investor will behave in an opportunistic manner: Since the idiosyncratic investment generates quasi-rents (i.e., value above and beyond value in next best use (Klein, Crawford, and Alchian 1978)), the non- investor will appropriate the quasi-rents generated by the idiosyncratic investment. Thus, the idiosyncratic nature of the asset creates a safeguarding problem of the quasi-rents and mechanisms must be designed to minimize the risk of opportunism by the non-investing party. TCE predicts that when uncertainty is high, frequency of transactions is high and significant relationship-specific investment are necessary, firms (who intend to minimize the sum of transaction costs and production costs) are more likely to be organized in hierarchies (i.e., vertical integration), to write detailed and long- terrn contracts, or use breach remedies in the form of penalties to help protect the investor (i.e., governance structure) (Williamson 1975, 1985). In essence, TCE focuses on mitigating ex post opportunistic behavior through governance structure. TCE provides theoretical support for the majority of the empirical research concerned with hold-ups (cf. Shelanski and Klein 1995). In their review of TCE’s empirical research, Shelanski and Klein (1995) and Coeurderoy and Quélin (1997) note that, while studies experience some measurement problems related to uncertainty and the degree of specificity of the investment, the evidence overwhelmingly lends support to TCE's theoretical predictions. 143 3.2.3 Property Rights Theory The preperty rights literature (cf. Grossman and Hart 1986; Hart and Moore 1990) also assumes incomplete contracting, non-contractible investment, and self-interest (thus, potential opportunistic behavior). However, property rights theory deviates from TCE in so far as property rights theory emphasizes that ownership of an asset confers its owner residual rights of control over the asset and, thus, is associated with the rights to appropriate returns from the assets. The unit of analysis is the asset. Additionally, property right theory assumes that, since knowledge of payoffs is common and since bargaining is costless, the ownership of the property rights and the contract will dictate the distribution of the surplus. Consequently, allocation of the surplus will be efficient. In other words, property rights theory focuses on avoiding the distortion of the ex ante investment instead of attempting to limit the ex post opportunism that occurs when the smplus is distributed. This leads Whinston (2001, 184) to argue that property rights theory is better suited to the analysis of intra-firm hold-ups than TCE. 3.2.4 Agency Theory The agency literature similarly relies on assumptions of self-interested behavior. Yet, an important distinction fiom TCE is that agency theory focuses on ex ante mechanisms and contracting to help anticipate the hazards due to separation of ownership and control and to align interests of principal and agent (cf. Jensen and Meckling 1976; Lambert 2001). The unit of analysis is the contract. Contract terms and incentives are the focus of agency analyses. Thus, contrary to transaction cost economics and property rights theory which assume that contracts are incomplete, agency theory assumes that contracts are complete, costless and, accordingly, enforceable by courts (Baiman 1990). 144 Risk preference of the principal and agent, and private information of the agent play an important role in agency theory predictions. Agency literature suggests that incentives and extensive exchange of information help curb opportunism by the agent. In summary, agency theory points to contracting as a simple solution to hold-ups (e. g., Edlin and Reichelstein 1995). Overall, various economic theories do not systematically lead to different predictions, but some of the variables of interest that differ from one theory to the other can help explain certain findings. For instance, Eisenhardt (1989) observes that the risk neutrality of the principal, a variable of interest of agency theory, provides a good explanation for Walker and Weber’s (1987) finding that uncertainty does not affect make or buy decisions for components when assets were highly specialized (i.e., dies). However, those findings could not be explained by transaction cost economics. Whereas economic theories such as TCE, property rights and agency theories study the limits to opportunistic behavior through formal mechanisms (such as governance structure, assignment of ownership rights, or incentives), relational contracting theory relies on informal mechanisms (e.g., social norms, context of the relationship) to explain limited opportunistic behavior. 3.2.5 Relational Contracting Theory Relational contracting theory takes into account the historical and social context in which the relationship takes place (Macneil 1980). The relationship between two firms or divisions is the unit of analysis. Relational contracting does not View each transaction as independent, but as embedded in a relationship where there is mutual interest. Relational contracting assmnes that preservation of the relation matters and cooperation 145 is instrumental. Additionally, it assumes that breach remedies cannot be specified and that investment is not contractible ex ante. While informal contracts such as a verbal agreement and a handshake are extreme examples of relational contracts (Palay 1985), Macneil (1986) proposes that relationships between firms include a relational dimension that complements the formal aspect of contracts. MacNeil fiirther specifies that “relational exchange creates circumstances where the long-run individual economic (material) interests of each party conflict with any short run desires to maximize individual utility respecting the goods in any particular exchange; the more relational the exchange, the more artificial becomes the idea of maximization” (1986, 578). This can be associated with the idea of reputation, trust, or social norms. A number of firms (e.g., Xerox, General Motors, Black & Decker, and Nieman Marcus) have demonstrated their focus on relational contracting by concentrating their transactions with a few carefully chosen suppliers (Sheth and Sharma 1997). Sheth and Sharma argue that drastically reducing the size of the firm’s supply base (by as much as 90% for Xerox) is associated with increased trust, cooperation and social bonds which are sine qua non for competitive positioning. Although Williamson (1979; 1985) touches on the social component of exchanges that are neither strictly arm’s length nor hierarchies, relational contracting theory develops much further the process by which the social dimension of the relationship mitigates the likelihood of opportunistic behavior (Bradach and Eccles 1989). Literature in the fields of economics, accounting, finance, and supply chain has used the theories presented above to create models and gather empirical evidence about alleviating the hold-up problem encountered when firms make decisions ranging from 146 organizational design and trade decisions to resource allocations. Their findings are detailed in the section below. 3.3 STRATEGIC DECISION-MAKING ASSOCIATED WITH IDIOSYNCRATIC INVESTMENTS The analytical models and empirical evidence presented below highlight developments to the hold-up literature subsequent to Shelanski and Klein’s (1995), and Coeurderoy and Que'lin’s (1997) reviews. Findings of analytical research are detailed in Table 3-1. Findings of empirical studies are presented in Table 3-2. In tables 3-1 and 3-2, the hold-up literature is organized around the strategic decisions of organizational design, trade, and resource allocations and by the types of mitigating mechanisms proposed by each study. This organization is motivated by two factors. First, each strategic decision chosen is representative of a stream of accounting research. Second, mitigating mechanisms can be grouped in two categories. Formal mechanisms of vertical integration, joint ownership, contracts, pricing mechanisms, incentives and interdependence are mostly representative of economics theories (i.e., transaction costs economics, property rights, and agency theory). On the other hand, investigations of informal mitigating mechanisms are supported by private information and relational contracting theory. 147 Quote e883“ was 8330M»: $268 .5 “5:532: 03560383 X 283 :5 .835 beam was @33330 “58. Ammoomv sausage ace... .5 68:8 325 x x x 5am so. 38am Aooamgoficm 050on 98 mmwmg Goa S coufioaxov 3:582 583 23 “85:8 325 N fiofionomcm a. 58m Gown—hue seafloomono .8ch Being “coma: a. “och—:8 325 N N 63 3 36m much Shrug: 2833—. 0.3:. 33 9 526033 282 1&3 3533530 3 a m w. “J mm MJWM. WJQWM mm mm mmmm wwmmm ...... W. m. .m. n W S m... u s a d «Emu—3:82 meets—em guano...— EESHE ”Eva—.282 38.5..— mEmiaAoo—z unflawE—Z mere—emu can wax—am Earn—a: Tm «Emu. 148 889nm :osmchomfi 838 was 8.3 x 883 .E 5 50¢. $38me E05582: Goa C no woman mfiwt 866% we .8982? X 5.8m 6% 588m cocasuomfi mo 093 no woman Saw: :oEooe .«o aouuoo=< N Q88 5on 28:32.“ 3.533— mfioeon 0:95 Ea 83:52: no noumEn—Eou x @003 micogmm 5% 8%: vagina use 38636 :0 e83 8:359:00 x 8:5 8 E23. 53% message no Essa? x 883 msaoeam coumESmE 38 mo bzmnmmtg :o wEufifiou Ammo: mica 5.35.: comma—-38 no 83383 Z x .3 8 mzfioEwm Ammoav Howbaoo 3E5 X E22238m 6% 56m much Shrubs 2332:. our; 3 3 m. W n. u. H I m. I m. I mummmwmwmummm n. m n m. m A. d m w. m. r M m m. u m. m m w m w. m. m. m m m... .... w s m m. a w. m m a a u o m. ma u w s m u a .0 253.382 mecca—em 38.3.5 1:585 mania—€02 38.8.”— uoauunou Tm 033—. 149 .3 2“ fluogmgfi cup—ogmomfl can? mflflogo “flow can 333332 320 mo 3:315 N N 2.33 him mach—85o HEB vofimmgmmu mm 3:25ng guacamofi $sz 93:83 3E x x 88$ .3 s 83022 $5938 bang: ~23 33683 pa 35.53%: 38 C oumHoamBE :3: and 533.525 awe N >188: Q 830 $50.58 33885 53, 3368mm 0.5 358635 93 C oueofimofl fig Ea acacia: gauge x :25 a. 833 in»... 35.3.3.5me 3 3 m. m1 u. n. H I J . 3%. m; w mmd 0mm nunmwmmmmwumm a m a _u n W m m. u. m. w mm mm. m. a w m m2 m m m . . .m. u u S u. m ” MW um. mania-.82 mucuanBO 38.5.:— nEnmnaaoo—z 38.5,.— mfianaauo—z mauawE—z 2.5-2:: can p.2—am Evian—m— N-m 033m. 150 puma Reggae 02 3338:: 8383— 388% ea 3 Ba 8:25? 28 833%? x x :89 5:5 .coumqbomfi 2.9% 8 86:0 93 Eng? mascdooom 8:80 X X $88 81.3 a. BEG $055 was Haggai: oBmEumnocb X X 988 ..m 3 ~85 A33 acmwoqfifi c2326 5335 was noumoidfifioo X X av nomwfizm b60582 was as: 58on62 mo souSoomxm X $3: .3 Ho whom magma—coo 33833—8 98 mmoquoEEOo 88800 X X Gloomy .E 8 58¢th “25 nwofifihmcoo _mow .Eogmoém 12835 X X Amoomv c8593 a. 9:. 55083888 98 ~355an mo sou—2898 .coufiongoo X Good swim a. 5.3 “5:533: #8325 was 23823 Hozmmsm X X 63: quE a. 955w 253er 022,—. 3 3 m. no. u. u. “a I fl 1. I mummmdwmmdmmm ”mummm.mmmmnmmg m. u a. m e m u m. m. m a u m s W. m _u. m. a P a s an n, m o m. m. m. u m S m S u. u s s m m. man—£2382 8333930 13:85 mfifiuaaooz 38.5,.— Bgzso 3. 2...; 151 Analytical models and empirical evidence identify and investigate mitigating mechanisms to hold-ups. These studies appear in economics, accounting, finance, supply chain and marketing fields and are organized in the tables around strategic decisions of organizational design, trade and resource allocations. Each category of decision provides a setting in which hold-ups are likely to occur albeit with some variations. 3.3.1 Organizational Design Empirical economics research has traditionally aimed to test the validity of TCE predictions by examining the organizational design choices firms make. In this setting, hold-up can occur when one firm must decide whether to make a relationship-specific investment followed by trade with a specific firm. The investment is not contractible, contracts are incomplete and thus, it is not possible for the trade partner to commit to not behave opportunistically. As a result, the firm decides whether to use integration or detailed contracts to avoid potential opportunism by its trade partner. Economics research has provided empirical evidence supporting TCE's claim that governance structures that take the form of vertical integration and detailed, long-term contracting encourage investment in idiosyncratic assets (when uncertainty is high) (cf. Shelanski and Klein 1995). Results of empirical research subsequent to the reviews by Shelanski and Klein (1995) and Coeurderoy and Quélin (1997) are summarized in table 3-2. Jensen and Rothwell (1998) have added to this research by conducting detailed task- related analyses of the operation of nuclear power plants and examining what factors lead firms to use their own employees instead of subcontractors. They find that, consistent with TCE predictions, plants are more likely to internally source production of critical tasks (i.e., vertical integration) when monitoring is difficult and firm-specific investments 152 are high. Accounting and finance research has recently contributed to this stream of research by investigating how relationship-specific investments affect firms’ outsourcing decisions. Coles & Hesterly (1998) is one of the early studies that examines the interaction between two important attributes of transactions, namely uncertainty and relation-specific investment. From a mail survey sent to hospitals, Coles and Hesterly measure the degree of routinization of performance characteristics and rate of technological change (i.e., uncertainty), and physical and human asset specificity. Their results are consistent with TCE’s predictions in so far as increasing uncertainty in the presence of asset specificity significantly increases the probability that the transaction will be internally sourced (for private hospitals). In sum, Coles and Hesterly propose that vertical integration serves as a formal safeguard against hold-ups as shown in Table 3-2. Anderson, Glenn and Sedatole (2000) also study the relationship between uncertainty and relationship-specific investment (i.e., dies) and outsourcing decisions, albeit in the automobile industry. The premise of their study is that all in-process dies are transaction-specific; consequently, the sourcing decision should be driven by uncertainty. They improve on prior measures of uncertainty by measuring part and subassembly complexity, and the difficulty of measuring contract performance. In conflict with theoretical predictions and prior empirical evidence, they find that the greater the uncertainty, the greater the likelihood of outsourcing. To explain this counter intuitive result, they suggest that their study provides insights about the sourcing decision for parts given the existence of capacity, whereas prior empirical evidence examined investments in long-term capacity. In other words, the sourcing decision they examine is not based on 153 minimizing the sum of production and transaction costs because their setting violates the separability of production and transaction costs, an implicit assumption of TCE. Their finding is classified under integration in Table 3-2. Additionally, an important contribution of their study is their attempt to measure opportunistic behavior by the non- investor (i.e., fi'equency of rework by supplier required and delay between firm request for delivery and actual delivery). They do not find evidence of opporttmistic behavior by OEM following idiosyncratic investment in dies made by suppliers and attribute this finding to relational contracting (earlier involvement of suppliers in relationship). Dyer’s study (1997) of the relationship between Japanese automobile OEMs and their suppliers has also documented the importance of relational contracting as a safeguard to hold-ups. Dyer (1997) conducts an exploratory study of fifty Japanese and American supplier-GEM relationships. The results from interviews and surveys provide evidence that, in Japan, relationships deepen over time and are characterized by information sharing and trust (i.e., dimensions of relationship characteristics) between Japanese OEMs and small groups of suppliers. Since Japanese OEM can use promises of high volume and repeated business to encourage idiosyncratic investment, the need for ex ante contracting and ex post negotiating over the surplus is greatly reduced. Additionally, Dyer finds evidence of joint ownership in idiosyncratic assets by OEM and supplier (a formal mechanism). In sum, Dyer finds preliminary evidence in support of relational contracting theory as well as property rights theory. Hence, as shown in Table 3-2, consistent with transaction cost theory, research provides evidence that organizational design decisions attempt to mitigate hold-ups mostly through formal mechanisms such as vertical integration (i.e., internally sourcing 154 products). Additionally, this stream of research demonstrates the importance of informal safeguards to hold-ups such as close relationships and provides support for relational contracting. While archival empirical papers provide evidence consistent or reconcilable with the theories presented above, Whyte (1994) proposes that cognitive biases of the investors, and not fear of hold—ups, are at the source of integration decisions. Results from an experiment in which subjects were asked to make a vertical integration decision demonstrate the influence of the sunk cost bias in the choice of governance structure, thus providing an alternative explanation of integration choices. Roodhoofi and Walop (1999) provide additional evidence about how the sunk cost bias affects outsourcing decisions. In the experiment conducted by Roodhooft and Walop, manager-subj ects incorporate past sunk costs and asset specificity to make their sourcing decision. Thus, both asset specificity and existence of sunk cost hinder outsourcing. Still, Troeger (2002) shows, with an evolutionary model, that the sunk cost bias can actually encourage investment in idiosyncratic assets over the long-run as investors benefit from the non-investors propensity to include sunk costs in their offer decision. 3.3.2 Trade Decisions The trade decisions (i.e., price and quantity negotiations) that follow a relationship-specific investment have been the subject of extensive scrutiny in the context of inter-firm as well as intra-firm transactions. Tables 3-1 and 3-2 group research on trade decisions by trade between firms (i.e., inter-firm research) and trade between divisions (intra-firm and transfer pricing research) as these two streams have evolved in different directions as suggested below. 155 3.3.2.1 Interflrm Trade Decisions Researchers have studied hold-ups between firms using analytical and empirical methods. Analytical papers are presented first in Table 3-1. They are followed by empirical papers in Table 3-2). In this setting, hold-up can occur when one firm must decide whether to make a relationship-specific investment followed by trade with a specific firm. Inter-firm hold-up papers rely on assumptions of noncontractible investment, incomplete contracting and resulting inability of the non-investor to commit to not behave opportunistically. 3. 3. 2. I . 1 Analytical models Scholars model hold-ups between two parties in a single period. Analytical models have initially focused on using contractual mechanisms to alleviate the hold-up problem. Edlin (1996) and Edlin and Reichelstein (1996) borrow from the literature on remedies to breach of contracts and propose to counterbalance the underinvestment associated with relationship-specific investment with the overinvestrnent created by breach remedies (this stream of literature argues that breach remedies encourage investment even when it is not efficient for the parties to trade). More specifically, Edlin (1996) demonstrates that, under expectation damages (i.e., damages that provide full compensation of victim of breach), relationship-specific investment will be encouraged when parties can sign an initial contract that specifies very high quality and quantity and the non-investor makes an up-fi'ont payment. This up-fiont payment creates interdependence between the parties and makes it more attractive for the non-investor to finish performance of the contract than to breach it. Thus, the investor is the only one likely to breach the contract. As such, the investor has incentives to maximize his share 156 of the surplus that will remain once the damages have been paid. As a result, he invests efficiently. Irnportantly, the formal mitigating mechanisms proposed by Edlin (i.e., contracts and interdependence as shown on Table 3-1) assume that courts have sufficient information to estimate expectation damages. Edlin and Reichelstein (1996) build on Edlin’s finding to introduce specific performance as an additional breach remedy with one-sided and bilateral investment. They conclude that, with one-sided investment, 3 non-contingent fixed price contract provides investment incentives under specific performance or expectation damages breach remedies. On the other hand, bilateral investment can only be encouraged with application of specific performance. In sum, choosing an initial contract (a formal mechanism) and the appropriate breach remedies prior to investing should alleviate the hold-up problem provided damages can be estimated ex ante. Although the nature of the idiosyncratic investment is not expanded upon in most papers (with the notable exception of Baiman and Rajan (2002a)), noteworthy insights from Che and Hausch (1999) point to limitations of the value of initial contracting as a mitigating mechanism to hold-ups. Che and Hausch highlight how the effectiveness of contracting arrangements, as a safeng to opportunistic behavior, varies depending on whether the investment made generates a benefit for the investor (i.e., selfish investment such as investment resulting in cost reduction for the investor) or whether it generates a benefit for the non-investor (i.e., cooperative investment such as R&D investment). Che and Hausch point out that although several papers argue that incomplete contracting may solve the hold-up problem, such a solution is only effective with selfish investments. 157 In their 2002 paper, Baiman and Rajan survey the literature that addresses how to mitigate inter-firm hold-up problems and illustrate those problems in term of bargaining power (i.e., making a selfish relationship-specific investment weakens the investor’s bargaining power). Baiman and Rajan propose that a simple non-contingent contract that specifies the quantity to be exchanged and the price at which the exchange will take place eliminates the hold-up problem. As in the intra-firm setting studied by Edlin and Reichelstein (1995), Baiman and Rajan argue that this initial contract changes the relative bargaining position of the investor in so far as it improves the investor’s status quo outcome; thus, leading the investor to receive what was guaranteed in the initial contract plus a share of the surplus from renegotiations. Such an initial contract (a formal mechanism) would however have no value with cooperative investments. As shown in Table 3-1, Baiman and Rajan also point to additional formal mechanisms such as joint ownership of the asset and interdependence with the buyer taking an equity stake in the supplier. In sum, as presented in Table 3-1, the aforementioned papers present some form of formal contracting as a solution to the hold-up problem in the tradition of agency theory. Some economists (e.g., Che and Hausch 1999), however, argue that although complete contracting would eliminate hold—ups, many contracts are incomplete and associated with cooperative investments. Accordingly, mitigating mechanisms other than contracting must be considered. Scholars have begun investigating information asymmetry as an informal mechanism to induce socially optimal investment. Gul (2001) examines how information asymmetry (i.e., investment is unobservable to the non-investor) can provide an 158 alternative to increasing the investor’s bargaining power and, accordingly, helps solve the hold-up problem. The intuition is that private information of the investor allows him to obtain information rents which provide incentives to invest. The private information of the investor, however, hinders trade negotiation and the hold-up problem is not mitigated. Still, Gul demonstrates that when negotiations are costless and the non-investor makes repeated take-it-or-leave-it offers, the hold-up problem is eliminated when the idiosyncratic investment is unobservable to the non-investor. The empirical investigation of the mechanisms proposed above has been rather sparse: While analytical models have focused on alleviating trade opportunism and encouraging relationship—specific investment through contractual mechanisms and private information of the investor, empirical research has introduced evidence supporting social norms and trust as mitigating mechanisms to hold-ups. The empirical literature can be organized in two streams. First, archival research focuses on relational contracting and informal mechanisms such as the importance of the quality of the relationships between the trading partners. Second, experimental research borrows fiom negotiation and economics literature and examines how characteristics of the negotiators (e.g., various forms of fairness and propensity to trust) and private information of the investor might help curtail hold-ups. 3. 3. 2. 1.2 Archival empirical evidence Archival empirical literature related to inter-firm trade is presented in Table 3-2. A survey of chemical manufacturers by Stump and Heide (1996) investigating trade behavior highlights that ex post opportunistic behavior can be averted by supplier selection (thus, qualification programs are an important element of buyer-supplier 159 relationships) and formal interdependence encouraged through bilateral investments. A112 and Brush (2000) administered surveys to machinery and electronic equipment OEMs. They find evidence that the positive relation between asset specificity and uncertainty and negotiation costs is mediated by informal relational elements such as collaboration, expectations of continuity, and communication. Still following this stream of research, Jap and Anderson (2003) surveyed buyer- supplier relationships of four Fortune 50 manufacturing companies. Buyers and suppliers were asked to report on opportunistic behavior they have encountered during trade. J ap and Anderson find that goal congruence, interpersonal trust and bilateral idiosyncratic investment act as safeguards against opportunism. In essence informal relationship characteristics (i.e., goal congruence and interpersonal trust) and formal interdependence (i.e., bilateral investment) serve as a commitment device and are associated with relationships with longer time horizon. Krishnan, Miller, and Sedatole’s (2007) examination of formal contracts a Fortune 500 firm has with its customers leads them to conclude that the supplier is more likely to own idiosyncratic asset when contracts are collaborative or more complete (thus, lending simultaneous support to informal methods of relational contracting and to more formal methods of transaction cost economics). Additionally, Krishnan et al. show that, as suggested by TCE and agency theory, collaborative contracting is more likely to be observed when monitoring and performance measurement is difficult. In sum, as with decisions related to organizational design, archival empirical research related to inter-firm trade decisions finds evidence consistent with predictions of economic theories. It also provides overwhelming evidence demonstrating the importance 160 of relationship characteristics, and supporting relational contracting theory, as shown in Table 3-2. 3.3.2.1.3 Experimental empirical evidence Empirical evidence related to trade decisions appears in Table 3-2. Experimental empirical research has examined how personal characteristics of the negotiators and private information can help mitigate hold-ups. The seminal work of Berg, Dickhaut, and McCabe (1995) provides an early investigation of hold-ups in an experimental setting. In the tradition of experimental economics, they conduct a two-stage version of a dictator game. The first set of subjects decides how much of their show up fee to send to an anonymous counterpart (i.e., relationship-specific investment). Any money they forward to their counterpart is tripled. Their counterpart then decides how much (if anything) s/he would like to return to the sender. The game is not repeated and communication is not allowed so as to avoid any reputation concerns. Whereas economic theory predicts that subjects should not send money to their counterpart as s/he will appropriate the whole amount, Berg et al. observe that 30 out of 32 subjects forward money. Additionally, Berg et al. observe that 16 out of 28 returned at least as much as they received (the remainder returned trivial amounts or nothing). In a second experiment, knowing the previous behavior of the group that received the money changed the dispersion of the offers and increased the correlation between the amount sent and the amount returned. Berg et al. conclude that informal mechanisms related to personal characteristics affected hold-ups. In particular, the subjects’ expectations of reciprocity led them to trust their counterparts and send a portion of their show up fees to other subjects (i.e., invest). Additionally, some subjects reciprocated (i.e., did not behave opportunistically). Variations of this 161 experiment have since been conducted (e. g., Malhotra 2004; Cox 2004) and obtained similar results. Ellingsen and Johannesson (2004) enrich Berg et al.’s analysis by examining the informal effect of communication (a dimension of relationship characteristics) on hold- up. Their observations and conclusions are consistent with Berg et al.’s. More specifically, they find some investment and some evidence of inequity aversion where non-investors reimburse the investment cost and share the surplus with the investor when communication is not possible. Moreover, in Ellingsen and Johannesson’s experiment, the investor is not made worse off by investing than by choosing not to invest. With communication, offers are even higher. In sum, as shown in Table 3-2, Ellingsen and J ohannesson find that relationship characteristics (in the form of communication) and individual characteristics (in the form of inequity aversion) help mitigate hold-ups. Sloof, Oosterbeek, and Sonnemans (2007) test Gul’s model and the prediction that private information in the form of unobservable investment reduces investment inefficiency. They construct a two-stage experiment and, keeping the surplus generated by the investment constant, they vary the informal information characteristic of investment observability (private infomation) in combination with the cost of the investment. As predicted by Gul, Sloof et al. observe that, when investment cost is high, subjects do not invest when the investment is observable, but do invest when it is unobservable. However, when investment cost is low, subjects invest. Those results lead Sloof et al. to conclude that economic predictions hold when fairness and reciprocity consideration are low (i.e., high investment costs for a fixed surplus), but not when there is room for fairness (i.e., low investment costs for a fixed surplus). They also observe 162 that, during trade, some subjects reimburse the investor for his investment and share the surplus. Sloof et al. further suggest that their study documents an interaction between fairness (i.e., a personal characteristic) and observability of the investment (i.e., private information) that affects the likelihood of investment. Drake and Haka (2007) conduct an experiment in a setting that includes product architectural interdependence and asymmetric information between two trading partners. First, Drake and Haka study the effects the type of accounting system used (i.e., coarse as Volume Based Costing or finer as Activity Based Costing) and the economic setting of the firm have on the propensity of negotiating individuals to share information. Second, they examine how those two factors affect the individuals’ ability to capture available trade efficiencies. They provide evidence that, because of individual concerns about inequitable outcomes (personal characteristics in Table 3-2), information asymmetry (i.e., private information) that takes the form of coarse accounting systems mitigates trade opportunism. Thus, they lend support to the role of coarse information in hold-ups but do not measure the participants’ individual preferences for fairness. In sum, experimental papers provide evidence that personal characteristics of the individual such as social preferences and private information of the investor help mitigate hold-ups. They show that, under certain circumstances, individuals include sunk cost in their calculations during trade and that investors, anticipating that they will not be worse off if they invest, invest in relationship-specific assets to a greater extent than predicted by economics. Researchers should, however, be cautious before concluding that, because of fairness preference of individuals, hold-ups are not as severe as predicted by economics. As suggested by Liedtka (1989), managers are not left to follow their own 163 preferences but their actions within a firm are guided by corporate culture and strategy. Thus, the role of personal preferences for fairness as safeguard against hold-ups might have been somewhat overstated by extant experimental research. Miller (2007) attempts to address this limitation by investigating the effects of firm induced strategy (an informal relationship characteristic that takes the form of fair or self-interested strategy) on hold-ups. Additionally, she builds on Sloof et al.’s study by examining whether private information in the form of aggregated supplier-provided cost information, another informal mechanism, alleviates hold-ups. She documents that aggregation of cost information has a positive effect on the investment decision of the supplier. Additionally, she finds evidence that the level of aggregation of supplier- provided cost information and firm strategy interact to effect trade and investment decisions. Overall, findings of archival and empirical research find some evidence consistent with economic predictions. They also provide evidence that characteristics of the relationship, of the individual and private information offer additional safeguards against hold-ups as shown in Table 3-2. They further suggest (as proposed by relationship contracting theory) that social norms and social preferences of the negotiators mediate the relation between characteristics of transactions, such as asset Specificity, and opportunistic behavior. They have, however, to a great extent ignored how firm strategy might be used to induce individual to behave in a manner that might not be consistent with their personal preference. Interestingly, research on inter-firm trade does not propose that incentives can be used to mitigate hold-ups. Yet, Kumar (1996) presents 164 anecdotal evidence that firms such as Procter & Gamble use incentives to motivate managers to refiain from holding-up their trading partners. 3.3.2.2 Intro-firm Trade Decisions and Transfer Pricing Accounting research has overwhelmingly focused on investigating safeguards to hold-ups in intra-firm relationships. In models of intra-firm trade, i.e., transfer pricing, divisional managers choose whether to make an investment that will benefit the entire firm, but has little value with external customers (i.e., a relationship-specific investment). At the time the investment must be made, the managers do not possess sufficient information to determine the size of the transfer needed (i.e., investment is not contractible), contracts between the divisions are incomplete and the non-investing division cannot commit to not behave opportunistically. Analysis of transfer pricing mechanisms has suggested that negotiated transfer pricing lead to underinvestrnent as the divisions split the surplus generated by the investment made by one division. Initial transfer pricing models proposed commitment (i.e., absence of renegotiation) between the divisions (Rogerson 1992) or a centralized mechanism of profit allocation with intervention of headquarters as remedies to the hold-up problem. More recently, scholars have investigated how contracting, transfer price mechanisms, and incentives can help alleviate hold-ups. Their findings are summarized in Table 3-1. 3. 3. 2. 2. I Contracting and Transfer Price Mechanisms The first set of analytical papers discussed below assumes information symmetry between investor and non-investor divisions. As shown in Table 3-1, they propose that formal mechanisms of initial contracts and cost-based transfer pricing alleviate hold-ups, 165 but must be supplemented by other mechanisms in order to address problems of trade distortion and moral hazard. Edlin and Reichelstein (1995) assume that two divisions of the same firm have equal bargaining power. They find that investment will be efficient if the divisions sign a fixed-price contract prior to investing (provided the investor can insist the contract will be fulfilled) and negotiate on quantity once all uncertainty has been resolved. In sum, as shown in Table 3-1, formal contracts can help mitigate hold-ups in an intra-firm setting. Furthermore, when divisional managers are subject to moral hazard, Edlin and Reichelstein show that division negotiation of the transfer payment and compensation of the managers based on divisional income helps align the interests of the firm and the managers. Baldenius, Reichelstein and Sahay (1999) do not investigate contracting as a safeguard against trade opportunism. Instead, Baldenius et a1. turn to pricing mechanisms and compare the effectiveness of negotiated and cost-based transfer pricing as a safeguard against hold-ups. Under their model of cost-based transfer pricing, the selling division presents the cost report and the buying division decides how many units to purchase. Baldenius et al. point out that, although cost-based transfer pricing encourages investment in idiosyncratic assets by one division, it creates an additional problem. First, should the selling division invest and have monopolistic pricing power (i.e., trade distortion), the buying division will reduce the quantities purchased; thus, reducing the selling division investment incentives. Second, should the buying division invest, the selling division can appropriate part of the return generated by the investment by charging higher transfer prices (assuming that costs of the selling division are not 166 verifiable). However, when the selling division is constrained in the costs it can report, cost-based transfer pricing is preferred. In sum, whether a negotiated or cost-based transfer pricing mechanism is preferred depends in part on the verifiability of the cost information provided by the selling division. Baldenius (2000) extends the analysis to models with asymmetric information. He concludes that bargaining power should reside with the investing division to alleviate hold-up and that bargaining power should reside with the division with the most private information to minimize trade distortion. Formal contracts can provide the means to allocate the bargaining power to the supplier or the buyer. In sum, as presented in Table 3-1, this stream of transfer pricing literature relies on formal contracting and pricing mechanisms to mitigate hold-ups. Still, empirical research could provide insights on how contracts are actually used between divisions to encourage relationship-specific investments. 3. 3. 2. 2.2 Incentives Building on findings from agency theory, a few papers presented in Table 3-1 have analyzed how incentives can be used to formally align the interests of division managers and the firm. Anctil and Dutta (1999) investigate incentive compensation as safeguard against hold-up under information symmetry. They demonstrate that the optimal compensation contract is based on divisional profit as well as firm-wide profit with the former allowing for some risk-sharing between risk-averse divisional managers. Thus, they conclude that formal incentives can help divisions mitigate hold-ups. Baldenius (2006) further expands the analysis under asymmetric information to include the scenario where managers payoffs are comprised of incentives and empire 167 building (i.e., managers derive benefits of control fi'om the assets they manage provided the assets are productive). Baldenius suggests that the propensity of managers to take advantage of their private information to distort trade (i.e., sellers ask for more than their reservation price and buyers bid less than their reservation price) is reduced by their wish for trade to take place so they can derive empire building benefits. Baldenius demonstrates that, even in the absence of firm-wide profit sharing, low—powered incentives result in division managers valuing empire benefits more. As a result, cooperative bargaining is encouraged and, consequently, relationship-specific investment takes place (i.e., alleviating hold-up problems). In sum, counterbalancing formal incentives with empire building can help safeguard divisions against hold-ups. Analytical research has shown that formal mechanisms such as contracts, transfer price mechanisms and incentives can help mitigate hold-ups when firms make intra-firm trade decisions as shown in Table 3-1. However, our understanding of how divisions actually deal with those issues could benefit fi'orn empirical research. 3.3.3 Resource Allocations Resource allocation decisions are intra-firm decisions that can require an idiosyncratic investment by the manager who presents a capital project to the principal. In this setting, hold-up can occur when the manager must acquire firm-specific information about a capital project. The manager’s project search is followed by the firm choosing which capital project to allocate resources to. The manager acquisition of information is often noncontractible, and the principal cannot commit to reward the manager for his project search investment and not behave opportunistically. 168 Although much acquisition of information by the manager conducting a project search is specific to the firm the manager works for, most of the capital budgeting literature assumes away the idiosyncratic nature of the gent’s investment. In much of the resource allocations literature, the manager is endowed with private information about the productivity of the capital project (Lambert 2001, 79; Antle and Fellingham 1997, 905). In that stream of literature, the effort decision of the manager is based on the information he possesses and the investment decision is based on the information elicited by the principal, but the manager does not need to exert any effort to acquire information. A few capital budgeting papers assume that investment in information search is endogenous. They propose that the agent needs to exert effort to gain knowledge about the proposed project and that this knowledge only has value for this specific firm (i.e., relationship—specific investment). The first group of papers (Lambert 1986; Kim 2006) assumes that the principal can use contracts to commit to share the quasi-rents with the manager. Lambert (1986) proposes that allocation of risk to the agent can, in certain circumstances, motivate the agent to exert effort to acquire information and to select the best project. Kim (2006) shows that when information acquisition costs are large, auditing the agent’s report (independent of whether the agent reports high or low productivity) reduces the opportunity costs of becoming informed. Auditing and allocation of more capital to the project help motivate the agent to invest in information search. In sum, although Lambert and Kim assume that acquisition of information is firm-specific, they propose that contracts can serve as a commitment device. As a result, opportunistic behavior by the principal is not considered and hold-ups are not examined. 169 Still, managers who invest in fnm-specific acquisition of information risk not deriving any benefit from their investment as the principal or other managers might appropriate the quasi-rents generated by their investment. The second group of papers (Baiman and Rajan 1995; Arya et al. 2000; Stein 2002) assumes that the principal cannot commit to reward the manager for his firm-specific investment in information acquisition. Thus, they encompass the main characteristics of hold-up problems (i.e., relationship—specific investment that creates appropriable quasi-rents and lack of commitment by the non-investor not to appropriate the quasi-rents). They do, however, vary in the role they give to opportunistic behavior. Results from this stream of research are presented in Table 3-1 under the heading of resource allocations. Stein (2002) models how bank managers can be encouraged to acquire firm- specific information about the projects of their customers. He argues that, when information is soft (i.e., cannot be easily passed on to headquarters), the manager who exerts effort to acquire information risks having his research effort wasted if headquarters chooses not to allocate funds to the projects he proposes. Stein recommends that formal decentralization, classified in Table 3-1 under integration, is the preferred mechanism to encourage manager investment in information acquisition as it gives the manager authority to allocate fimds and to obtain the quasi-rents fi'om his investment. Baiman and Rajan (1995) model a capital investment decision where the manager is required to invest in firm-specific human capital, thus obtaining private information about a capital investment project for his firm. The manager’s firm-specific investment is followed by a capital investment in the project. The project outcome is a function of the level of investment, the manager’s effort and some state outcome. Baiman and Rajan 170 examine two possible inefficiencies. On the one hand, assigning capital investment decision rights to the manager can result in moral hazard. On the other, assigning capital investment decision rights to the owner can result in the owner acting opportunistically and appropriating the manager’s quasi-rents when the owner designs the manager’s compensation package. Baiman and Rajan conclude that the size of the manager’s firm specific human capital investment drives the assignment of capital investment decision rights (i.e., centralized if no specific investment, shared if moderate investment and delegated to the manager if large investment). Thus, as shown by Stein in the context of soft information, Baiman and Rajan propose that allocation of decision rights (a form of integration) can under circumstances alleviate hold-ups. Arya et al. (2000) address the same problem of potential appropriation of the manager’s quasi-rents, but turn to how the owner might be able to commit not to behave opportunistically. They propose that a manager can be motivated to increase his search for a profitable project (i.e., firm-specific human capital investment) if the owner’s information system provides coarse and late information. Such a system would create slack for the manager, and, thus, would provide incentives for the manager to increase his project search effort. Arya et al. conclude that the interaction between fineness and timing of information helps mitigate the hold-up problem because it creates private information that guarantees the manager that the principal will not be able to appropriate the surplus generated by his investment in project search. Thus, in addition to providing support for economic theories by showing the importance of allocating decision rights (a formal integration mechanism), analytical papers point to informal mechanisms of private information as a potential mitigating 171 mechanism to hold-ups as shown in Table 3-1. However, our understanding of how hold- ups might be mitigated when resource allocation decisions are made could benefit from more analytical as well as empirical research. 3.4 CONCLUSIONS Transaction cost economics, property rights, and agency theories have until recently provided the framework for analyzing hold-ups. They have suggested solutions to the hold-up problem in the form of formal mechanisms such as vertical integration, detailed and long-term contracts, allocation of property rights, and incentives. While analytical research and early empirical research (cf. Shelanski and Klein 1995; Coeurderoy and Quélin 1997) have, to a great extent, supported the predictions of economic theories, recent developments in empirical research have focused more on relational contracting theory and informal mechanisms. They have documented that characteristics of the relationship and of the individual limit firms’ and individuals’ tendency to behave opportrmistically and to appropriate the surplus generated by an idiosyncratic investment. Furthermore, observing that most individuals do not behave in a self-interested manner in laboratories, experimental research has gone as far as questioning if hold-up problems are as severe as economic theories lead us to believe. Still, hold-ups are an organizational problem and it would behoove us to understand better the relationship between fairness preferences of individuals and inter- firrn and inter—division investment decisions. Additionally, recent theoretical developments (cf. Arya et al. 2000; Gul 2001) suggest that further investigations of private information and of the effects information characteristics (e. g., fineness, precision, timeliness) have on hold-ups should be a fruitful avenue for future research. 172 The proliferation of analytical research that views contracts as a safeguard against hold- ' ups compared to the scarcity of the related empirical research (with the notable exceptions of Joskow (1985), Anderson et al. (2000) and Krishnan et al. (2007)) suggests that additional empirical investigation of contracts would help enrich our understanding of the dangers associated with incomplete contracting and of ways to curtail risk of ex post opportunism. Finally, although relationship-specific investments are incurred in the areas of organizational design, trade and resource allocations, some streams of research assume away the idiosyncratic nature of the investment (e. g., resource allocations) or provide little empirical investigation of hold-up mitigating mechanisms (e.g., intra-firm trade and resource allocations). Conducting additional research in these areas would give us a more in-depth understanding of how to address the hold-up problem. 173 REFERENCES Anctil, R., and S. Dutta. 1999. Negotiated transfer pricing and divisional vs. firm-wide performance evaluation. The Accounting Review 74 (l):87-104. Anderson, E. 1985. The sales person as outside agent or employee: a transaction cost analysis. Marketing Science 4 (3):234-254. . 1994. Transaction cost analysis and marketing. In Transaction Cost Economics and Beyond, edited by J. Groenewegen. Boston/Dordrecht/London: Kluwer Academic Publishers. Anderson, E., and B. Weitz. 1992. The use of pledges to build and sustain commitment in distribution channels. Journal of Marketing Research 29 (1):] 8. Anderson, S., and H. Dekker. 2005. Management control for market transactions: the relation between transaction characteristics, incomplete contract design, and subsequent performance. Management Science 51 (12): 1734-52. Anderson, S., D. Glenn, and K. Sedatole. 2000. Sourcing parts of complex products: evidence on transactions costs, high-powered incentives and ex-post opportunism. Accounting, Organizations and Society 25 (8):723-749. Antle, R., and J. Fellingham. 1997. Models of capital investments with private information and incentives: a selective review. Journal of Business Finance and Accounting 24:887-908. Artz, K., and T. Brush. 2000. Asset specificity, uncertainty and relational norms: an examination of coordination costs in collaborative strategic alliances. Journal of Economic Behavior & Organization 41 (2):337-362. Arya, A., J. Fellingham, J. Glover, and K. Sivaramakrishnan. 2000. Capital budgeting, the hold-up problem and information system design. Management Science 46 (2):205-216. Baiman, S. 1975. The Evaluation and Choice of Internal Information Systems Within a Multiperson World. Journal of Accounting Research 13 (l):1-15. . 1990. Agency research in managerial accounting: a second look. Accounting, Organizations and Society 15 (4):341-371 . Baiman, S., and M. Rajan. 1995. Centralization, delegation, and shared responsibility in the assignment of capital investment decision rights. Journal of A ccounting Research 33 (supplement):l35-164. 174 . 2002a. Incentive issues in inter-firm relationships. Accounting, Organizations and Society 27 (3):213-23 8. . 2002b. The role of information and opportunism in the choice of buyer-supplier relationships. Journal of Accounting Research 40 (2):247-278. Baldenius, T. 2000. Intrafirm trade, bargaining power, and specific investments. Review of Accounting Studies 5 :27-56. . 2006. Ownership, incentives and the hold-up problem. The Rand Journal of Economics forthcoming. Baldenius, T., S. Reichelstein, and S. Sahay. 1999. Negotiated vs. cost-based transfer pricing. Review of Accounting Studies 4 (2):67-91. Bensaou, M. 1999. Portfolios of buyer-supplier relationships. Sloan Management Review 40 (4)3544. Bensaou, M., and B. Anderson. 1999. Buyer-supplier relations in industrial markets: When do buyers risk making idiosyncratic investments? Organization Science 10 (4) :460-48 1 . Bensaou, M., and N. Venkatraman. 1995. Configurations of interorganizational relationships: A comparison between U.S, and Japanese automakers. Management Science 41 (9): 1471-1492. Berg, J ., J. Dickhaut, and K. McCabe. 1995. Trust, reciprocity, and social history. Games and Economic Behavior 10:122-142. Bourgeois, L. J ., and D. Brodwin. 1984. Strategic Implementation: Five Approaches to an Elusive Phenomenon. Strategic Management Journal (pm-1986) 5 (3):241-264. Bradach, J ., and R. Eccles. 1989. Price, authority, and trust: from ideal types to plural forms. Annual Review of Sociology 15:97-1 18. Brenner, T., and N. Vriend. 2006. On the behavior of proposers in ultimatum games. Journal of Economic Behavior & Organization 61 (4):617-631. Carnerer, C., and G. Loewenstein. 1993. Information, fairness, and efficiency in bargaining. In Psychological perspectives on justice: Theory and applications, edited by B. Mellers and J. Baron. Cambridge: Cambridge University Press. Carmichael, L., and W. B. MacLeod. 2003. Caring About Sunk Costs: A Behavioral Solution to Holdup Problems with Small Stakes. Journal of Law, Economics, and Organizations 19 (l):106-118. 175 Carr, C., and J. Ng. 1995. Total cost control: Nissan and its UK. supplier partnerships. Management Accounting Research 6:347-365. Chang, C., and G. Ive. 2007. The hold-up problem in the management of construction projects: A case study of the Channel Tunnel. International Journal of Project Management in press. Che, Y., and D. Hausch. 1999. Cooperative investments and the value of contracting. The American Economic Review 89 (1): 125-147. Coase, R. H. 1937. The nature of the firm. Economica 4 (16):386-405. Coeurderoy, R., and B. Quélin. 1997. L'économie des cofits de transactions: un bilan des études empiriques sur l'intégration verticale. Revue d'Economie Politique 107:145-181. Coles, J. W., and W. S. Hesterly. 1998. Transaction costs, quality, and economies of scale: examining contracting choices in the hospital industry. Journal of Corporate Finance 4 (4):321-345. Cooper, R., and R. Slagmulder. 2004. Interorganizational cost management and relational context. Accounting, Organizations and Society 29:1-26. Cox, J. 2004. How to identify trust and reciprocity. Games and Economic Behavior 46:260-281. Das, T. K. 2006. Strategic alliance temporalities and partner opportunism. British Journal ofManagement 17 (l):1-21. De Dreu, C., and P. van Lange. 1995. The impact of social value orientations on negotiator cognition and behavior. Personality and Social Psychology Bulletin 21 (11):]178-1188. Dekker, H. 2003. Value chain analysis in interfirm relationships: a field study. Management Accounting Research 14 (1):1-23. Demski, J. S., and H. F rimor. 1999. Performance measure garbling under renegotiation in multiperiod agencies. Journal of Accounting Research 37: 1 87-214. Diekmann, K., A. Tenbrunsel, P. Shah, H. Schroth, and M. Bazerman. 1996. The descriptive and prescriptive use of previous purchase price in negotiations. Organizational Behavior and Human Decision Processes 66 (2):] 79-191 . Doz, Y., and O. Baburoglu. 2000. From competition to collaboration. In Cooperative Strategy, edited by D. d. R. Faulkner, M. New York: Oxford University Press, Inc. 176 Drake, A., and S. Haka. 2007. Does ABC information exacerbate hold-up problems in buyer-supplier negotiations? The Accounting Review Forthcoming. Dyer, J. 1997. Effective interfirm collaboration: how firms minimize transaction costs and maximize transaction value. Strategic Management Journal 18 (7):535-556. Edlin, A. 1996. Cadillac contracts and up—front payments: Efficient investment under expectation damages. Journal of Law, Economics, & Organization 12 (1):98-118. Edlin, A., and S. Reichelstein. 1995. Specific investment under negotiated transfer pricing: an efficiency result. The Accounting Review 70 (2):275-291. . 1996. Holdups, standard breach remedies, and optimal investment. The American Economic Review 86 (3):478-501. Eisenhardt, K. 1989. Agency theory: An assessment and review. Academy of Management Review 14 (l):57-74. Ellingsen, T., and M. Johannesson. 2004a. Is there a hold-up problem? Scandinavian Journal of Economics 16 (3):475-494. . 2004b. Promises, threats and fairness. The Economic Journal 114:397-420. . 2005. Sunk costs and fairness in incomplete information bargaining. Games and Economic Behavior 50:155-177. Ellingsen, T., and J. Robles. 2002. Does evolution solve the hold-up problem? Games and Economic Behavior 39:28-53. Ewerhart, Christian. 2006. The effect of sunk costs on the outcome of altemating-offers bargaining between inequity-averse agents. Schmalenbach Business Review 58 (2): 184. Farrell, J ., and N. Gallini. 1988. Second-sourcing as commitment: Monopoly incentives to attract competition. The Quarterly Journal of Economics 103 (4):673-694. Farrell, J ., and C. Shapiro. 1989. Optimal contracts with lock-in. The American Economic Review 79 (l):51-68. Fehr, E., and K. Schmidt. 1999. A theory of fairness, competition and cooperation. Quarterly Journal of Economics 1 14:817-868. Floyd, S., and B. Wooldridge. 1992. Middle management involvement in strategy and its association with strategy type: a research note. Strategic Management Journal (1986-1998) 13 (Special issue):153-167. 177 Gantner, A., W. Gueth, and M. Koenigstein. 2001. Equitable choices in bargaining games with joint production. Journal of Economic Behavior & Organization 46:209-225. Goldberg, V. 1976. Regulation and administered contracts. The Bell Journal of Economics 7 (2):426-448. Granlund, M., and J. Taipaleenmaki. 2005. Management control and controllership in new economy firms - a life cycle perspective. Management Accounting Research 16 (1):21-57. Grossrnan, S., and 0. Hart. 1986. The costs and benefits of ownership: a theory of vertical and lateral integration. The Journal of Political Economy 94 (4):691-719. Gueth, W., R Schmittberger, and B. Schwarze. 1982. An experimental analysis of ultimatum bargaining. Journal of Economic Behavior & Organization 31367-388. Gul, F. 2001. Unobservable investment and the hold-up problem. Econometrica 69:343— 376. Gupta, A. K., and V. Govindarajan. 1984. Business unit strategy, managerial characteristics, and business unit effectiveness at strategy implementation. Academy of Management Journal 27:27—41 . Hackett, S. C. 1994. Is relational exchange possible in the absence of reputations and repeated contact? Journal of Law, Economics, & Organization 10 (2):360-3 89. Hart, 0., and J. Moore. 1988. Incomplete contracts and renegotiation. Econometrica 56 (4):755-785. . 1990. Property rights and the nature of the firm. The Journal of Political Economy 98 (6):]119-1158. Heide, J ., and G. John. 1990. Alliances in industrial purchasing: The determinants of joint action in buyer-supplier relationships. JMR, Journal of Marketing Research 27 (1)224—36. Helper, S. 1991. How much has really changed between US. automakers and their suppliers? Sloan Management Review 32 (4): 15-28. Henke, J. Jr. 2004. Poor Relationships Costing US Automakers. Planning Perspectives, Inc. 2004 [cited August 3 2004]. Available from ht_tp://www.ppil.com/static/2004-OEM-supplier study-finalpfd. . Study shows suppliers still prefer Toyota, Honda, but Ford and General Motors are improving. Planning Perspectives, Inc., September 2006 [cited 8/31/2006. Available from www.ppil .com/news/?im_2_rge=news. 178 Ijiri, Y. 1975. Theory of accounting measurement. Studies in Accounting Research 10. Ittner, C.., D.. Larcker, V. Nagar, and M. Rajan. 1999. Supplier selection, monitoring practices, and firm performance. Journal of Accounting and Public Policy 18 (3):253-281. J ap, S., and B. Anderson. 2003. Safeguarding interorganizational performance and continuity under ex post opportunism. Management Science 49 (12): 1 684-1701 . Jensen, J. B., and G. S. Rothwell. 1998. Transaction costs, regulation, and subcontracting at nuclear power plants. Journal of Economic Behavior & Organization 36 (3):369-381. Jensen, M. , and W. Meckling. 1976. Theory of the firm: Managerial behavior, agency costs, and ownership structure. Journal of Financial Economics 32305-360. J oskow, P. 1985. Vertical integration and long—term contracts: the case of coal-burning electric generating plants. Journal of Law and Economics 1 (1 ):33-80. . 1987. Contract duration and relationship-specific investments: empirical evidence from coal markets. The American Economic Review 77 (1):168-185. Kahneman, D., J. Knetsch, and R. Thaler. 1986. Fairness as a constraint on profit seeking: entitlements in the market. American Economic Review 76 (4):728-741. Kaplan, R. S., and D. Norton. 1996. Using the balanced scorecard as a strategic management system. Harvard Business Review January. Kim, D. 2006. Capital budgeting for new projects: On the role of auditing in information acquisition. Journal of Accounting and Economics 41 :25 7-270. Klein, B. 1980. Transaction cost determinants of "unfair" contractual relations. American Economic Review Proceedings 70:356-362. . 1988. Vertical integration as organizational ownership: The Fisher Body- General Motors relationship revisited. Journal of Law and Economics 4 (1):199- 21 3. . 1996. Why hold-ups occur: The self-enforcing range of contractual relationships. Economic Inquiry 34 (3):444-463. Klein, B., R. Crawford, and A. Alchian. 1978. Vertical integration, appropriable rents, and the competitive contracting process. Journal of Law and Economics 21 (2):297-326. 179 Krishnan, R., F. Miller, and K. Sedatole. 2007. Determinants of collaborative contracting in inter-firm relations. Kristof, A. 1996. Person-organization fit: an integrative review of its conceptualizations, measurement, and implications. Personnel Psychology 49 (1)21-49. Kulmala, H. 2002. Accounting in customer-supplier relationships - Developing cost management in network environment. Paper read at 3rd Conference on New Directions in Management Accounting: Innovations in practice and research, December 12-14, at Brussels. Belgium. Kumar, N. 1996. The power of trust in manufacturer-retailer relationships. Harvard Business Review 74 (6):92-106. Lambert, R. 2001 . Contracting theory and accounting. Journal of Accounting and Economics 32:3-87. Lambert, R. A. 1986. Executive effort and selection of risky projects. The RAND Journal of Economics 17 (1):77-88. Langfield-Smith, K. 1997. Management control systems and strategy: A critical review. Accounting, Organizations and Society 22:207-232. Langfield-Smith, K., and D. Smith. 2003. Management control systems and trust in outsourcing relationships. Management Accounting Research 14:281-307. Liebrand, W. 1984. The effect of social motives, communication and group size on behavior in an N-person multi-stage mixed-motive game. Eur0pean Journal of Social Psychology 14:239-264. Liedtka, J. 1989. Value congruence: The interplay of individual and organization. Journal of Business Ethics 8 (10):805-815. Macneil, I. 1980. The New Social Contract. New Haven, CT: Yale University Press. . 1986. Exchange revisited: Individual utility and social utility. Ethics 96:567-593. Malhotra, D. 2004. Trust and reciprocity decisions: The differing perspectives of tr'ustors and trusted parties. Organizational Behavior and Human Decision Processes 94:61 -73. March, J ., and J. Olsen. 1989. Rediscovering institutions: The organizational basis of politics. New York: Free Press. Masten, S., J. Jr. Meehan, and E. Snyder. 1991 . The costs of organization. Journal of Law, Economics, & Organization 7 (l):1-25. 180 Mayer, R., J. Davis, and F. Schoorrnan. 1995. An integrative model of organizational trust. The Academy of Management Review 20 (3):709-734. Messick, D., and C. McClintock. 1968. Motivational bases of choice in experimental games. Journal of Experimental Social Psychology 4 (1):1-25. Milgrom, P., and J. Roberts. 1992. Economics, Organizations, and Management. Englewood Cliffs, New Jersey: Prentice-Hall, Inc. Miller, F. 2007. Information aggregation, strategy and investment decisions in buyer- supplier relationships. Monteverde, K., and D. Teece. 1982. Appropriable rents and quasi-vertical integration. Journal of Law and Economics 25 (2):321-328. Ochs, J ., and A. Roth. 1989. An experimental study of sequential bargaining. The American Economic Review 79 (3):355-3 84. Palay, T. 1985. Avoiding regulatory constraints: Contracting safeguards and the role of informal agreement. Journal of Law, Economics, & Organization 1:155-185. Parayre, R. 1995. The strategic implications of sunk costs: A behavioral perspective. Journal of Economic Behavior & Organization 28:417-442. Parkhe, A. 1993. Strategic alliance structuring: a game theoretic and transaction cost examination of interfirm cooperation. Academy of Management Journal 36 (4):794-829. Pirrong, S. C. 1993. Contracting practices in bulk shipping markets: A transactions cost. Journal of Law and Economics 36 (October):93 7-976. Pittman, R. 1991. Specific investments, contracts, and opportunism: The evolution of railroad sidetrack agreements. Journal of Law and Economics 34 (2):565-589. Porter, M. E. 1985. Competitive advantage. New York: Free Press. Ring, P. S., and A. Van De Ven. 1992. Structuring c00perative relationships between organizations. Strategic Management Journal 13 (7):483. Roberts, J. 2004. The modern firm: Organization design for performance and growth. New York: Oxford University Press. Rogerson, W. 1992. Contractual solutions to the hold-up problem. Review of Economic Studies 59 (4):777-793. 181 Roodhooft, F ., and L. Warlop. 1999. On the role of sunk costs and asset specificity in outsourcing decisions: a research note. Accounting, Organizations and Society 24:363-369. Roth, A., and K. Murnighan. 1982. The role of information in bargaining: An experimental study. Econometrica 50 (5):1123-1142. Sako, M. 2004. Supplier development at Honda, Nissan and Toyota: Comparative case studies of organizational capability enhancement. Industrial and Corporate Change 13 (2):281-308. Seal, W., J. Cullen, A. Dunlop, T. Berry, and M. Ahmed. 1999. Enacting a European supply chain: A case study on the role of management accounting. Management Accounting Review 10:303-322. Shelanski, H., and P. Klein. 1995. Empirical research in transaction cost econonrics: a review and assessment. Journal of Law, Economics, & Organization 11 (2):335- 361. Shepard, A. 1987. Licensing to Enhance Demand for New Technologies. The Rand Journal of Economics 18 (3). Sheth, J ., and A. Sharma. 1997. Supplier relationships: Emerging issues and challenges. Industrial Marketing Management 26:91-100. Sloof, R., H. Oosterbeek, and J. Sonnemans. 2007. Does making specific investments unobservable boost investment incentives? Journal of Economics and Management Strategy (forthcoming). Stein, J. C. 2002. Information production and capital allocation: Decentralized versus hierarchical firms. The Journal of Finance 57 (5): 1891-1921 . Stump, R, and J. Heide. 1996. Controlling supplier opportunism in industrial relationships. Journal of Marketing Research 33 (4):431-441. Thomas, A., R. Litschert, and K. Ramaswamy. 1991. The performance impact of strategy-manager coalignment: an empirical examination. Strategic Management Journal 12 (7):509-522. Tirole, J. 1986. Procurement and renegotiation. The Journal of Political Economy 94 (2):235-259. . 1988. The Theory of Industrial Organization. Cambridge, Massachusetts: The MIT Press. 182 Troeger, T. 2002. Why sunk costs matter for bargaining outcomes: An evolutionary approach. Journal of Economic Theory 102:3 75-402. Walker, G., and D. Weber. 1987. Supplier competition, uncertainty, and make-or-buy decisions. The Academy of Management Review 30 (3):589—596. Whinston, M. 2001. Assessing the property rights and transaction-cost theories of frrm scope. The American Economic Review 91 (2): 184-188. Whyte, Glen. 1994. The role of asset specificity in the vertical integration decision. Journal of Economic Behavior & Organization 23 (3):287-302. Williamson, 0. 1975. Markets and hierarchies: Analysis and antitrust implications. New York, NY: The Free Press. . 1979. Transaction-cost economics: The governance of contractual relations. Journal of Law and Economics 22:233-261. . 1983. Credible commitments: Using hostages to support exchange. The American Economic Review 73 (4):519-540. . 1985. The economic institutions of capitalism: firms, markets, relational contracting. New York: Free Press. . 1991. Comparative economic organization: The analysis of discrete structural alternatives. Administrative Science Quarterly 36 (2):269-296. . 1993. Calculativeness, trust, and economic organization. Journal of Law and Economics 36:453-486. 183 lllllllllllllllll117111111111lflllllllll s 1293 02345 9539