 

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